System

Abstract

A system is provided.SOLUTION: The system includes means for collecting past materials, photographic materials, and handwritten memo data, means for cleaning and analyzing the collected data, means for learning a machine learning model based on the cleaned data, means for inputting information of a new business negotiation destination from a user, means for analyzing the input information and automatically generating a business negotiation material, means for comparing the new business negotiation destination with the past data and generating a proposal content, means for recognizing the handwritten memo data as a text and incorporating the text into the business negotiation material, and means for confirming and correcting the generated business negotiation material and finally outputting the material.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The technology of the present disclosure relates to a system. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] Creating business documents is a time-consuming and labor-intensive process. Furthermore, because the content and format of documents vary for each business meeting, it is difficult to quickly provide standardized, high-quality documents. Furthermore, it is difficult to utilize unstructured data such as handwritten notes, and there is a need for a way to effectively utilize this data. [Means for solving the problem]

[0005] The present invention provides a system that includes means for collecting past documents, photographs, and handwritten memo data, cleaning them, and analyzing them. It also includes means for training a machine learning model based on the cleaned data. It also includes means for inputting information about new business partners from a user, analyzing that information, and automatically generating business negotiation materials. It also includes means for comparing new business partners with past data to generate proposals, and means for recognizing handwritten memo data as text using OCR technology and incorporating it into the business negotiation materials. Finally, the system includes means for confirming and correcting the generated business negotiation materials and outputting them.

[0006] "Past documents" refer to documents and reports used in previous business negotiations or projects.

[0007] "Photo material" refers to image data used in business negotiations and presentations.

[0008] "Handwritten memo data" refers to data that has been digitized from handwritten content.

[0009] "Means of collection" refers to methods for obtaining past documents, photographs, and handwritten notes from databases, etc.

[0010] "Cleaning methods" refer to methods for removing unnecessary information from collected data and shaping it into a form suitable for analysis and learning.

[0011] "Means of analysis" refers to methods for understanding the content of the data and extracting the necessary information.

[0012] A "machine learning model" is an algorithm that automatically learns patterns and knowledge from data.

[0013] "Means of learning" refers to the method of training a machine learning model using past materials and data.

[0014] "New business partners" refer to new business partners or proposals with which we have not previously done business.

[0015] "Business negotiation materials" refer to materials such as presentations and reports used when conducting business negotiations.

[0016] "Means of automatic generation" refers to a method in which the system autonomously creates sales documents based on input information.

[0017] "Means of comparison" refers to the method of analyzing information about new business partners and comparing it with past data.

[0018] The "proposal content" refers to the solutions or services presented to the business partner.

[0019] "OCR technology" stands for optical character recognition technology, which converts characters in an image into text data.

[0020] "Means for checking and correcting" refers to a method for checking the generated sales negotiation materials and correcting them as necessary.

[0021] "Means of output" refers to the method of outputting the final completed document in a format that can be used by the user (such as a PDF or presentation file). [Brief explanation of the drawings]

[0022] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0023] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0024] First, the terms used in the following description will be explained.

[0025] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0026] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0027] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0028] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0029] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0030] [First embodiment]

[0031] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0032] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0033] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0034] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0035] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0036] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0037] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0038] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0039] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0040] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0041] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0042] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0043] The present invention relates to a system for automatically generating business negotiation materials, and is implemented as follows: The system is mainly composed of three entities: a server, a terminal, and a user.

[0044] Data collection and preprocessing

[0045] The server first collects past sales materials, photographs, and handwritten notes from the database. For example, the server uses an SQL query to execute a command such as "SELECT FROM past_materials" to retrieve past sales materials. The collected data is then cleaned and formatted to make it suitable for analysis. For example, unnecessary HTML tags and line breaks are removed, and the text data is normalized.

[0046] Data training

[0047] The server trains a machine learning model based on the cleaned historical data. Specifically, it performs TF-IDF vectorization to convert text data into numerical vectors, and then trains the model using algorithms such as logistic regression. This model is used to predict key elements of sales documents and proposal content.

[0048] Entering new business partner data

[0049] The user inputs information about the potential business partner (e.g., company name, industry, and needs) through the interface from a terminal. The input data is sent to the server and analyzed.

[0050] Generate sales documents

[0051] The server analyzes the data of new business partners and automatically generates sales documents based on that data. The server converts the input text data back into TF-IDF vectors and inputs them into a trained model to predict the proposal content. The prediction results are formatted as sales documents according to a template.

[0052] Proposal generation

[0053] The server compares the data of new business partners with past data and generates the most suitable proposals. These proposals are based on past success stories and learning data, allowing for an effective approach to new business partners.

[0054] Analysis of handwritten notes

[0055] The user scans their handwritten notes onto their device and sends them to the server, which then uses OCR technology to extract and clean the text from the handwritten notes, thereby incorporating the contents of the handwritten notes into the business meeting materials.

[0056] Check and print the final materials

[0057] The user can then review the generated sales documents on their device and make any necessary corrections. Finally, the server outputs the reviewed sales documents in PDF or presentation format and provides them to the user. This series of processes ensures that high-quality sales documents are generated efficiently.

[0058] Specific examples

[0059] For example, if a new business partner is a technology company seeking cloud solutions, the user inputs that information into the device. The server then analyzes the input signal and references relevant past business documents, photos, and handwritten notes to automatically generate the optimal proposal and materials. Finally, the user can review the generated business documents and make any necessary corrections, and the server outputs them in PDF format. This process allows for the rapid and efficient provision of standardized business documents.

[0060] The processing flow will be explained below.

[0061] Step 1:

[0062] The server collects past business materials, photographic materials, and handwritten memo data from the database. Specifically, it retrieves this data by executing a command such as "SELECT FROM past_materials" using an SQL query. The retrieved data is saved as business materials data, image data, and handwritten memo data, respectively.

[0063] Step 2:

[0064] The server cleans the collected data. For example, it removes unnecessary HTML tags and line breaks from past sales documents and normalizes the text data. For photo materials, it uses image recognition algorithms to trim unnecessary parts, and removes noise and emphasizes characters from handwritten memo data.

[0065] Step 3:

[0066] The server trains a machine learning model based on the cleaned data. Specifically, it performs TF-IDF vectorization to convert the text data into a numerical vector. It then uses classification algorithms such as logistic regression to train the model on past sales documents, generating a model that can extract the elements necessary for sales documents.

[0067] Step 4:

[0068] The user uses a terminal to input information about the new business partner (e.g., company name, industry, needs) through a dedicated interface. The input information is sent to the server as new business partner data.

[0069] Step 5:

[0070] The server receives and analyzes the new business negotiation data sent by the user. Specifically, it converts the new data into text format and performs TF-IDF vectorization. After that, it uses a machine learning model to automatically generate optimal business negotiation materials based on the analysis results.

[0071] Step 6:

[0072] The server compares the data of the new business partner with past data and generates the most suitable proposal. For example, based on past success stories and proposals, the server incorporates content to propose the optimal solutions and services to the new business partner into the business negotiation materials.

[0073] Step 7:

[0074] The user scans their handwritten notes into their device and sends the data to the server, which then uses OCR technology to extract text from the notes, cleans it, and formats it into a format suitable for analysis.

[0075] Step 8:

[0076] The user checks the generated sales negotiation materials on the terminal and makes corrections as necessary. Checks the contents of the sales negotiation materials and makes corrections if there is missing information or any parts that need to be corrected.

[0077] Step 9:

[0078] The server then outputs the final confirmed business documents in PDF or presentation format and provides them to the user, who can then download them and use them in actual business negotiations.

[0079] Example 1

[0080] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0081] The traditional process of creating sales documents is often done manually, consuming a great deal of time and effort. It is also difficult to effectively utilize past success stories and data, making it difficult to make effective proposals to new business partners. Furthermore, the manual process of incorporating information from handwritten notes into sales documents is cumbersome and inaccurate. There is a need for a system that can improve this current situation and automatically generate efficient, high-quality sales documents.

[0082] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0083] In this invention, the server includes: means for retrieving documents from a database using an SQL query; means for cleaning and analyzing the collected data and deleting unnecessary HTML tags and line breaks; means for performing TF-IDF vectorization to convert the cleaned text data into a numerical vector; means for training a machine learning model using an algorithm such as logistic regression using the TF-IDF vectorized data; means for a user to input information about a new business partner through a terminal; means for analyzing the input information and automatically generating business partner data as business partner documents; means for comparing the input information with past business partner data to generate optimal proposals; means for scanning handwritten memo data entered by the user and converting the converted handwritten memo data into text using OCR technology; means for importing the converted handwritten memo data into business partner documents; means for a user to review and correct the generated business partner documents; and means for outputting the reviewed documents in PDF format or presentation format. This enables efficient automatic generation of high-quality business partner documents.

[0084] An "SQL query" is a syntax for retrieving data from a database.

[0085] A "database" is a system for storing structured data.

[0086] "Cleaning" is the process of removing unnecessary information from data and preparing it in a form suitable for analysis.

[0087] An "HTML tag" is a part of the markup language used to describe the structure of a web page.

[0088] "TF-IDF vectorization" is a method for converting text data into a numerical vector.

[0089] A "numeric vector" is a format in which text data is expressed as numbers.

[0090] A "machine learning model" is an algorithm that learns patterns and relationships from data and makes predictions and classifications.

[0091] "Logistic Regression" is a statistical model for performing binary classification based on given data.

[0092] An "interface" is a screen or input form that allows a user to interact with a system.

[0093] "OCR technology" stands for optical character recognition technology, which is a technology that extracts character data from images.

[0094] "PDF format" is an abbreviation for Portable Document Format, a file format for displaying documents and images in a fixed layout.

[0095] This invention is a system that collects past documents, photographs, and handwritten memo data, and automatically generates business negotiation materials based on these. The system is mainly composed of three entities: a server, a terminal, and a user.

[0096] Data collection and preprocessing

[0097] First, the server uses an SQL query to retrieve past sales documents, photographs, and handwritten notes from the database. An example of a specific SQL query is "SELECT FROM past_materials." The collected data is then cleaned by removing unnecessary HTML tags and line breaks, and normalizing the text data. This prepares the data for analysis.

[0098] Data training

[0099] The cleaned data is transformed into a TF-IDF vector by the server and converted into a numerical vector. A machine learning model is then trained using algorithms such as logistic regression. This creates a model for predicting important elements of sales documents and proposal content.

[0100] Entering new business partner data

[0101] The user uses the terminal to input information about the potential business partner (company name, industry, needs, etc.) through the interface. The input data is sent from the terminal to the server and used for analysis.

[0102] Generate sales documents

[0103] The server analyzes the data of new business partners and automatically generates sales documents based on that data. During this process, TF-IDF vectorization is performed again, and the data is input into a trained model to predict the proposal content. The prediction results are formatted as sales documents based on a template. A specific output may be data embedded in a PowerPoint template.

[0104] Proposal generation

[0105] The server compares the data of the new business partner with past data and generates the most suitable proposal. This proposal is based on past success stories and learning data, making it possible to approach the new business partner effectively.

[0106] Analysis of handwritten notes

[0107] The user scans handwritten notes onto the device, which are then sent to a server where text data is extracted from the notes using OCR technology (e.g., Google® Cloud Vision API).Then, the text data is cleaned and incorporated into the sales meeting materials.

[0108] Check and print the final materials

[0109] The user checks the generated sales documents on their device and makes any necessary corrections. The server then outputs the final, checked sales documents in PDF or presentation format and provides them to the user. This series of processes ensures that high-quality sales documents are generated efficiently.

[0110] Specific examples

[0111] For example, if a new business partner is a technology company seeking a cloud solution, the user inputs that information via their device. The server then analyzes the input signal and references relevant past business documents, photos, and handwritten notes to automatically generate the optimal proposal and materials. Finally, the user can review the generated business documents and make any necessary corrections, after which the server outputs them in PDF format. This process ensures that business documents are provided quickly and efficiently.

[0112] Prompt Sentence Examples

[0113] "Generate the perfect proposal for new business opportunities with technology companies looking for cloud solutions."

[0114] In this invention, by using a generative AI model, negotiation materials are generated automatically and with high accuracy.

[0115] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0116] Step 1: Data collection and preprocessing

[0117] The server first uses an SQL query to retrieve past sales documents, photographs, and handwritten notes from the database. Specifically, it executes the query "SELECT FROM past_materials." The input is various materials in the database, and the output is the data collected by the server. The server then cleans the collected data, removing unnecessary HTML tags and line breaks, and normalizing the text data to make it suitable for analysis.

[0118] Step 2: Train the data

[0119] The server converts the cleaned data into a TF-IDF vector and then into a numerical vector. Next, it trains a machine learning model using algorithms such as logistic regression. The input is the cleaned text data, and the output is a machine learning model that predicts the important elements of sales documents and proposal content. For example, a model can be trained using data from past sales negotiations to build a model that predicts "sales negotiation elements with a high probability of success."

[0120] Step 3: Enter new customer data

[0121] The user inputs information about the new business partner (company name, industry, needs, etc.) through a terminal. This data is sent to the server in JSON format. The input is specific information about the new business partner, and the output is data sent to the server for analysis. For example, the user might input information such as "technology company" and "needs cloud solutions."

[0122] Step 4: Generate sales documents

[0123] The server analyzes data from new business partners, performs TF-IDF vectorization, and then inputs it into a trained model to predict proposal content. The input is information about new business partners, and the output is automatically generated business negotiation materials. As a specific example, a "presentation material proposing the scalability of cloud infrastructure" is generated based on a template.

[0124] Step 5: Generate proposals

[0125] The server compares the data of the new business partner with data from past negotiations, and generates the optimal proposal based on past success stories and related materials. The input is the information of the new business partner and past data, and the output is the proposal. For example, by referring to past successful cases in the same industry, the optimal proposal for the new business partner can be created.

[0126] Step 6: Analyze handwritten notes

[0127] The user scans their handwritten notes onto their device and sends the data to the server. The server then uses OCR technology to extract text data from the handwritten notes and cleans them. The input is the scanned handwritten notes, and the output is the converted handwritten notes. This allows the contents of the handwritten notes to be reflected in the sales documents.

[0128] Step 7: Check and print the final document

[0129] The user checks the generated sales documents on the terminal and makes corrections as necessary. The corrected documents are sent to the server, and the final confirmed sales documents are output in PDF or presentation format. The input is the sales documents corrected by the user, and the output is the final sales documents. For example, the user checks the presentation documents, edits them as necessary, and then saves the final version as a PDF.

[0130] (Application example 1)

[0131] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0132] The traditional process of creating sales documents requires a lot of time and effort, including collecting, cleaning, analyzing, and digitizing handwritten notes from past data and documents. It is also difficult to reflect this information in sales negotiations in real time. This results in lower efficiency in sales negotiations and makes it difficult to quickly provide appropriate proposals to customers.

[0133] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0134] In this invention, the server includes means for collecting past documents, photographic materials, and handwritten memo data, means for cleaning and analyzing the collected data, means for training a machine learning model based on the cleaned data, means for inputting information about new business partners from a user, means for analyzing the input information and automatically generating business negotiation materials, means for comparing the new business partners with past data and generating proposals, means for recognizing handwritten memo data as text and incorporating it into the business negotiation materials, means for checking and correcting the generated business negotiation materials and finally outputting the materials, means for providing the generated business negotiation materials to a user in real time, means for inputting new data through a user interface and immediately reflecting it in the business negotiation materials, and means for recognizing handwritten memo data using OCR technology and adding it to the business negotiation materials in real time, thereby enabling efficient and rapid generation of standardized business negotiation materials and real-time updates.

[0135] "Means of collection" refers to the methods and processes for obtaining past documents, photographic materials, handwritten memo data, etc. from the database and incorporating them into the system.

[0136] "Cleaning" is the process of removing unnecessary information from collected data and shaping it into a form suitable for analysis.

[0137] "Means for analysis" refers to a method for extracting and analyzing meaningful information from the cleaned data.

[0138] "Training" is the process of using cleaned and analyzed data to train a machine learning model.

[0139] "Means for input" refers to the interface and procedures for allowing users to input information about new business partners into the system.

[0140] The "means for automatic generation" is a process for automatically generating negotiation materials based on information input by the user.

[0141] The "means of generation" is a method of comparing information on new business partners with past data and deriving proposal content.

[0142] The "recognition method" is a method of converting handwritten note data into text data using OCR technology.

[0143] "Means for checking and correcting" refers to a procedure by which the user checks the generated business negotiation materials and revise them as necessary.

[0144] The "means of output" refers to the method of providing the final confirmed and revised business documents to the user in PDF or presentation format.

[0145] "Means for providing in real time" refers to a process for instantly presenting the generated business negotiation materials to the user.

[0146] A "means for reflecting" is a mechanism for immediately incorporating new data entered through the user interface into the sales materials.

[0147] "Means of recognizing using OCR technology" refers to a method of converting handwritten note data into digital text using optical character recognition technology.

[0148] "Means of adding in real time" refers to the process of immediately including the converted text data in the sales documents.

[0149] The present invention relates to a system for automatically generating business negotiation materials, and is implemented as follows: The system is mainly composed of three entities: a server, a terminal, and a user.

[0150] Data collection and preprocessing

[0151] The first thing the server does is collect past documents, photographs, and handwritten notes from the database. The server retrieves past business negotiation documents and cleans the collected data. This cleaning process involves removing unnecessary HTML tags and line breaks and normalizing the text data. The Python Pandas library can be used in this process.

[0152] Data training

[0153] The server trains a machine learning model based on the cleaned historical data. Specifically, it performs TF-IDF vectorization and then uses algorithms such as logistic regression to train the model. The Python scikit-learn library can be used. This model is used to predict key elements of sales documents and proposal content.

[0154] Entering new business partner data

[0155] The user inputs information about the potential business partner (e.g., company name, industry, and needs) through the interface from a terminal. The input data is sent to the server and analyzed.

[0156] Generate sales documents

[0157] The server analyzes the data of new business partners and automatically generates sales documents based on that data. The data of new business partners is converted back into TF-IDF vectors and input into the trained model to predict the proposal content. The prediction results are formatted as sales documents according to a template.

[0158] Proposal generation

[0159] The server compares the data of new business partners with past data and generates the most suitable proposals. These proposals are based on past success stories and learning data, allowing for an effective approach to new business partners.

[0160] Analysis of handwritten notes

[0161] The user scans their handwritten notes onto their device and sends them to the server. The server uses OCR technology to extract text from the handwritten notes and cleans them. A Python OCR library (such as Tesseract OCR) can be used. This allows the contents of the handwritten notes to be incorporated into the sales documents.

[0162] Real-time updates and provision of business documents

[0163] The generated sales documents are provided to the user in real time, and new data entered through the user interface is immediately reflected in the sales documents.

[0164] Check and print the final materials

[0165] The user can then review the generated sales documents on their device and make any necessary corrections. Finally, the server outputs the reviewed sales documents in PDF or presentation format and provides them to the user. This series of processes ensures that high-quality sales documents are generated efficiently.

[0166] Prompt Sentence Examples

[0167] Collect, clean, and format your past sales pitch data.

[0168] Next, enter the new contact information and generate the appropriate documents.

[0169] If you have handwritten note images, use OCR technology to convert them into text and include them in the generated document.

[0170] Finally, output the document in PDF format.

[0171] Specific examples

[0172] For example, if the client is a medical device manufacturer, the user inputs the information into the device. The server then analyzes the input data and references relevant past sales documents, photographs, and handwritten notes to automatically generate the optimal proposal and materials. The generated sales documents are provided in real time, and the user can modify them as needed and finally output them in PDF format. This process allows for the rapid and efficient provision of standardized sales documents.

[0173] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0174] Step 1:

[0175] Data collection

[0176] The server retrieves past business negotiation materials, photographic materials, and handwritten memo data from the database. Database connection information and a data collection query are used as input. Specifically, an SQL query is executed to retrieve data. For example, the command "SELECT FROM past_materials" is executed to retrieve past business negotiation materials. The output is the raw data of the retrieved materials and memo data.

[0177] Step 2:

[0178] Cleaning the data

[0179] The server removes unnecessary HTML tags and line breaks from the acquired data and normalizes the text data. It uses the collected raw data as input. Specifically, it uses Pandas and a regular expression library to format the data. For example, it performs processing such as "df['text'] = df['text'].str.replace(r'<[^>]>', '')". The output is cleaned text data.

[0180] Step 3:

[0181] Data training

[0182] The server performs TF-IDF vectorization on the cleaned past data and trains a Logistic Regression model. The cleaned text data is used as input. Specifically, the scikit-learn library is used to perform vectorization and model training. Each process is performed using "tfidf = TfidfVectorizer()" and "model = LogisticRegression()". The output is a trained model.

[0183] Step 4:

[0184] Entering new business partner data

[0185] The user enters information about the new business contact (company name, industry, and needs) through the terminal interface. The input is data manually entered by the user. Specifically, form input and text fields are used. The input new business contact information data is obtained as the output.

[0186] Step 5:

[0187] Analysis of new business contact data

[0188] The server analyzes the entered new client data and converts it back into a TF-IDF vector. The input is the new client information entered by the user. Specifically, the data is converted using the TF-IDF vectorizer defined earlier. Set "X_new = tfidf.transform([new_client_data])". The output is the vectorized new client data.

[0189] Step 6:

[0190] Automatic generation of sales documents

[0191] The server inputs the vectorized new business negotiation data into the trained model and predicts the proposal content. The vectorized new business negotiation data and the trained model are used as input. Specifically, the processing is performed as follows: "prediction = model.predict(X_new)". The output is the sales negotiation document data containing the predicted proposal content.

[0192] Step 7:

[0193] Proposal generation and formatting

[0194] The server formats the sales negotiation materials according to a template based on the predicted proposal content. The predicted proposal content data is used as input. Specifically, the proposal content is applied to the sales negotiation material template to generate a document. The formatted sales negotiation material data is obtained as output.

[0195] Step 8:

[0196] Analysis of handwritten notes

[0197] The user scans handwritten notes onto the device. The server receives the data and extracts text from the handwritten notes using OCR technology. The input is the image data of the handwritten notes. Specifically, the text is extracted using an OCR library (such as Tesseract OCR). The output is the handwritten note data converted into text.

[0198] Step 9:

[0199] Integrate handwritten notes into sales documents

[0200] The server cleans the converted handwritten notes and integrates them into the business negotiation materials. The input is the converted handwritten notes. Specifically, the cleaned data is added to the business negotiation materials, e.g., "formatted_material += ocr_text". The output is the integrated business negotiation materials.

[0201] Step 10:

[0202] Real-time provision of business negotiation materials

[0203] The server provides the generated sales negotiation materials to the user in real time. Every time new data is entered through the user interface, it is reflected in the sales negotiation materials. New user data is used as input. Specifically, data synchronization is performed using WebSocket or real-time communication technology. The latest sales negotiation material data is provided as output in real time.

[0204] Step 11:

[0205] Final confirmation and printing of business documents

[0206] The user checks the generated sales document on their device and makes any necessary corrections. The server then outputs the final, checked document in PDF or presentation format and provides it to the user. The final, checked sales document data is used as input. Specifically, the document is output using a PDF generation library (such as ReportLab). The final sales document in PDF format is obtained as output.

[0207] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0208] This invention relates to a system for automatically generating business negotiation materials, and furthermore, a system for adjusting proposal content by combining an emotion engine that recognizes the user's emotions. The system is mainly composed of four entities: a server, a terminal, a user, and an emotion engine.

[0209] Data collection and preprocessing

[0210] The first thing the server does is collect past business materials, photo materials, and handwritten memo data from the database. For example, the server uses an SQL query to execute a command such as "SELECT FROM past_materials" to retrieve this data. The retrieved data is saved as business materials data, image data, and handwritten memo data, respectively.

[0211] The server then cleans the collected data. For example, unnecessary HTML tags and line breaks are removed from past sales documents and text data is normalized. Image recognition algorithms are used to trim unnecessary parts of photo materials, and noise is removed and text is emphasized from handwritten notes.

[0212] Data training

[0213] The server trains a machine learning model based on the cleaned historical data. Specifically, it performs TF-IDF vectorization to convert text data into numerical vectors, and then trains the model using algorithms such as logistic regression. This model is used to predict key elements of sales documents and proposal content.

[0214] Entering new business partner data

[0215] The user inputs information about the new business partner (e.g., company name, industry, needs) from the terminal through a dedicated interface. The input information is sent to the server as new business partner data.

[0216] Generate sales documents

[0217] The server receives and analyzes the new business negotiation data sent by the user. Specifically, it converts the new data into text format and performs TF-IDF vectorization. After that, it uses a machine learning model to automatically generate optimal business negotiation materials based on the analysis results.

[0218] Proposal generation

[0219] The server compares the data of the new business partner with past data and generates the most suitable proposal. For example, based on past success stories and proposals, the server incorporates content to propose the optimal solutions and services to the new business partner into the business negotiation materials.

[0220] Analysis of handwritten notes

[0221] The user scans their handwritten notes into their device and sends the data to the server, which then uses OCR technology to extract text from the notes, cleans it, and formats it into a format suitable for analysis.

[0222] Use of emotion engine

[0223] The emotion engine analyzes user interactions and recognizes user emotions in real time. For example, facial expressions and voices are collected by a camera or microphone when the user uses an input interface, and the emotion engine analyzes them to extract emotional data.

[0224] The emotion data recognized by the emotion engine is used by the server to adjust the content of suggestions. For example, if the user is tired, the server can maximize the effectiveness of suggestions by summarizing the content of the materials more succinctly or using expressions that encourage relaxation.

[0225] Check and print the final materials

[0226] The user checks the generated business negotiation materials on the terminal and makes corrections as necessary. Checks the contents of the business negotiation materials and makes corrections if there is missing information or if there are any parts that need to be corrected.

[0227] Finally, the server outputs the confirmed business documents in PDF or presentation format, allowing the user to use the documents in actual business negotiations.

[0228] Specific examples

[0229] For example, if a new business partner is a technology company seeking cloud solutions, the user inputs that information into their device. The server then analyzes the input signal and references relevant past sales documents, photos, and handwritten notes to automatically generate the optimal proposal content and materials. An emotion engine also analyzes the user's interactions. If the user is excited, it can incorporate responses into the proposal materials that further emphasize the proposal and increase the dynamism of the presentation. Finally, the user reviews the generated sales documents, makes any necessary corrections, and the server outputs them in PDF format. This process ensures that standardized sales documents are provided quickly and efficiently.

[0230] The processing flow will be explained below.

[0231] Step 1:

[0232] The server collects past business materials, photo materials, and handwritten memo data from the database. Specifically, it executes a command such as "SELECT FROM past_materials" using an SQL query, and saves this data as business materials data, image data, and handwritten memo data, respectively.

[0233] Step 2:

[0234] The server cleans the collected data. For example, it removes unnecessary HTML tags and line breaks from past sales documents and normalizes the text data. It also uses image recognition algorithms to trim unnecessary parts of photographs, and removes noise and emphasizes characters from handwritten notes.

[0235] Step 3:

[0236] The server trains a machine learning model based on the cleaned data. Specifically, it performs TF-IDF vectorization to convert the text data into numerical vectors, and then trains the model using algorithms such as logistic regression. This model is used to predict important elements of sales documents and proposal content.

[0237] Step 4:

[0238] The user uses a terminal to input information about the new business partner (e.g., company name, industry, needs) through a dedicated interface. The input information is sent to the server as new business partner data.

[0239] Step 5:

[0240] The server receives and analyzes the new business negotiation data sent by the user. Specifically, it converts the new data into text format and performs TF-IDF vectorization. After that, it uses a machine learning model to automatically generate optimal business negotiation materials based on the analysis results.

[0241] Step 6:

[0242] The server compares the data of the new business partner with past data and generates the most suitable proposal. For example, based on past success stories and proposals, the server incorporates content to propose the optimal solutions and services to the new business partner into the business negotiation materials.

[0243] Step 7:

[0244] The user scans their handwritten notes into their device and sends the data to the server, which then uses OCR technology to extract text from the notes, cleans it, and formats it into a format suitable for analysis.

[0245] Step 8:

[0246] When a user inputs emotion data through the interface, the emotion engine analyzes the user's facial expressions and voice. The emotion engine analyzes the interaction and extracts the user's emotion data. For example, it can recognize the user's emotion in real time using a camera or microphone.

[0247] Step 9:

[0248] The server receives the user's emotional data recognized by the emotion engine and adjusts the content of the proposal. Specifically, if the user is excited, the proposal content is emphasized, while if the user is tired, the materials are summarized more succinctly. This allows the server to provide optimal business negotiation materials according to the user's psychological state.

[0249] Step 10:

[0250] The user checks the generated sales negotiation materials on the terminal and makes corrections as necessary. Checks the contents of the sales negotiation materials and makes corrections if there is missing information or any parts that need to be corrected.

[0251] Step 11:

[0252] The server then outputs the final confirmed business documents in PDF or presentation format and provides them to the user, who can then use the documents in actual business negotiations.

[0253] Example 2

[0254] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0255] The current process of creating sales documents requires manual input and data organization, which consumes a lot of time and effort. In addition, the proposals are often suboptimal because they do not take into account the user's emotional state.

[0256] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0257] In this invention, the server includes means for collecting past documents, image data, and handwritten information data, means for cleaning and analyzing the collected data, and means for training a machine learning model based on the cleaned data. This enables efficient data collection and preprocessing. The server also includes means for a user to input information about a new business partner, means for analyzing the input information and automatically generating business negotiation materials, means for comparing the data about the new business partner with past data and generating proposal content, means for recognizing handwritten information data as character data and incorporating it into the business negotiation materials, means for checking and correcting the generated business negotiation materials and finally outputting the materials, and means for analyzing the user's emotions and adjusting the proposal content based on the emotions. This enables automatic generation of business negotiation materials and adjustment of the proposal content based on the user's emotions.

[0258] "Past documents" are previously created records including business documents, image data, and handwritten information data.

[0259] "Image data" is digital data that contains visual information such as photographic material.

[0260] "Handwritten information data" is data in digital form obtained from handwritten notes or documents.

[0261] "Cleaning" refers to removing unnecessary elements from data and preparing it in a format suitable for analysis.

[0262] A "machine learning model" is an algorithm that learns patterns and relationships in data and makes predictions and classifications based on new data.

[0263] "Information on new business partners" refers to attribute data on new customers or companies that are the subject of business negotiations, such as company name, industry, and needs.

[0264] "Business documents" refers to presentation materials and proposals used to conduct business negotiations.

[0265] "Comparison" refers to comparing new data with previous data to identify similarities and differences.

[0266] The "proposal content" is information that includes an outline of the solution or service to be provided to the business partner.

[0267] "Character data" is digital data that represents information in text format.

[0268] "Optical character recognition technology (OCR)" is a technology that converts handwritten or printed character data into digital text.

[0269] "Emotion analysis" refers to the process of identifying a user's emotional state from their facial expressions and voice.

[0270] "Adjustment" refers to modifying content or behavior to suit specific conditions or circumstances.

[0271] The present invention relates to a system that combines the automatic generation of business negotiation materials with the recognition of user emotions. This system is composed of four main components: a server, a terminal, a user, and an emotion engine.

[0272] Data collection and preprocessing

[0273] The server first collects past sales documents, image data, and handwritten information data from the database. It retrieves the data using an SQL query (e.g., "SELECT FROM past_materials") and saves it in a folder corresponding to each data type. After this process, the server cleans the collected data. Specifically, it uses regular expressions to remove unnecessary HTML tags and line breaks from the text data and normalizes the text. For image data, Python's Pillow library is used to trim unnecessary parts, and for handwritten information, an OCR tool (e.g., Tesseract) is used to extract text and remove noise.

[0274] Data training

[0275] The server trains a machine learning model based on the cleaned data. It converts the text data into TF-IDF vectors, then uses the scikit-learn library to build a logistic regression model and learns the key elements for generating sales documents.

[0276] Entering new business partner data

[0277] The user inputs information about the new business contact (company name, industry, needs, etc.) through a dedicated interface on the terminal. The terminal receives the user's input using GUI components (e.g., text boxes, drop-down lists) and sends it to the server.

[0278] Generate sales documents

[0279] The server receives and analyzes the data of new business partners sent by the user. Specifically, it converts the new data into text format and converts it into TF-IDF vectors. It then uses the trained model to automatically generate optimal business documents.

[0280] Proposal generation

[0281] The server compares the data of the new business partner with past data and generates the optimal proposal for the new business partner based on past success stories and proposal content. This process may use natural language generation tools (e.g., GPT-3 (registered trademark)).

[0282] Analysis of handwritten notes

[0283] Users scan their handwritten notes onto the device, which uses OCR software to extract the text from the handwritten notes, and then send this data to a server that cleans and formats it for analysis.

[0284] Use of emotion engine

[0285] The emotion engine analyzes the user's interactions. The user's facial expressions and voice are collected by the device's camera and microphone, and the emotion engine analyzes them to extract emotional data. This data is sent to the server and used to adjust the content of the suggestions. For example, if the user is tired, the content of the materials can be summarized more concisely.

[0286] Check and print the final materials

[0287] The user checks the generated sales documents on the terminal. The user checks the documents in the GUI and makes any necessary corrections. Finally, the server outputs the checked sales documents in PDF or presentation format. The user downloads them and uses them in actual sales negotiations.

[0288] Specific examples

[0289] For example, if a new business partner is a technology company seeking a cloud solution, the user enters that information on their device. The server then analyzes the input data and automatically generates the optimal proposal by referencing relevant past business documents, image data, and handwritten notes. In addition, if the user expresses excitement, the emotion engine analyzes this and adds dynamism to further emphasize the proposal. The user can then review the final business document and make any necessary corrections, after which the server outputs it in PDF format.

[0290] Prompt Sentence Examples

[0291] "Please enter the company name, industry, and needs of your new business partner into the terminal."

[0292] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0293] Step 1: Data collection and preprocessing

[0294] The server collects past sales documents, image data, and handwritten information data from a database. The input is an SQL query (e.g., "SELECT FROM past_materials"), and the output is data saved in a folder corresponding to each data type. The server then cleans the collected data. Specifically, it uses regular expressions to remove unnecessary HTML tags and line breaks from the text data and normalizes the text. It also uses the Pillow library to trim unnecessary parts of the image data, and an OCR tool (e.g., Tesseract) to extract text and remove noise from the handwritten information data. This results in data in a format suitable for analysis.

[0295] Step 2: Train the data

[0296] The server trains a machine learning model using the cleaned data. The input is the cleaned text data, and the output is the trained model. First, the text data is converted into TF-IDF vectors, and then a logistic regression model is built using the scikit-learn library. This model learns the elements necessary for generating sales documents.

[0297] Step 3: Enter new customer data

[0298] The user inputs information about the new business contact (company name, industry, needs, etc.) through a dedicated interface on the terminal. The input is a GUI component (e.g., text box, drop-down list), and the output is the information about the new business contact that is sent to the server. The terminal receives this data and sends it to the server.

[0299] Step 4: Generate sales documents

[0300] The server receives and analyzes the data of new business partners sent by the user. The input is the information of new business partners, and the output is the generated business documents. Specifically, the new data is converted into text format and TF-IDF vectorized. The trained model is then used to automatically generate the optimal business documents.

[0301] Step 5: Generate proposals

[0302] The server compares the data of the new business partner with past data and generates an appropriate proposal. The input is the data of the new business partner, past success stories, and proposals, and the output is the proposal to be incorporated into the business materials. This process may use natural language generation tools (e.g., GPT-3).

[0303] Step 6: Analyze handwritten notes

[0304] A user scans their handwritten notes into a device, which then uses OCR software to extract text from them. The input is the scanned image of the handwritten notes, and the output is text data sent to a server, which cleans and formats the text data for analysis.

[0305] Step 7: Use the Emotion Engine

[0306] The emotion engine analyzes the user's interactions and recognizes their emotions. The input is the user's facial expressions and voice, and the output is emotional data. Data collected by the device's camera and microphone is analyzed by the emotion engine, and the extracted emotional data is sent to the server. The server adjusts the content of the suggestions based on this data. For example, if the system recognizes that the user is tired, it may make adjustments such as summarizing the contents of the materials more concisely.

[0307] Step 8: Check and print the final document

[0308] The user checks the sales negotiation materials generated on the terminal and makes corrections as necessary. The input is the generated sales negotiation materials, and the output is the final confirmed sales negotiation materials. The materials are displayed on the GUI, and missing information and corrections are made. Finally, the server outputs the confirmed sales negotiation materials in PDF or presentation format. The user downloads them and uses them in actual sales negotiations.

[0309] (Application example 2)

[0310] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0311] In an automatic generation system for sales documents, there is a demand for the ability to quickly and efficiently generate optimal proposals for new business partners based on past data, but conventional systems have had difficulty responding in real time or adjusting based on user emotions.In addition to sales documents, it is also important to make real-time situational assessments and take appropriate action in the manufacturing process, but there has been a lack of systems that can do this.

[0312] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting past documents, photographic materials, and handwritten memo data, means for cleaning and analyzing the collected data, means for training a machine learning model based on the cleaned data, means for inputting information about new business partners from a user, means for analyzing the input information and automatically generating business negotiation materials, means for comparing the new business partner with past data and generating proposal content, means for recognizing handwritten memo data as text and incorporating it into the business negotiation materials, means for assessing the situation in real time and taking appropriate action, means for recognizing the user's emotions and adjusting the proposal content, and means for confirming and correcting the generated business negotiation materials and finally outputting the materials. This makes it possible to realize a system that efficiently generates business negotiation materials and enables adjustments and real-time responses based on the user's emotions.

[0313] "Past materials" is a general term for information collected in the past, such as business negotiation materials, manufacturing records, photographic materials, and handwritten memo data.

[0314] "Means of collection" refers to the methods and devices used to obtain historical documents, photographic materials, and handwritten notes from databases and other sources.

[0315] "Cleaning and analyzing means" refers to a method or device that performs processing such as removing unnecessary parts from collected data, normalizing text data, or removing noise from image data.

[0316] "Means for training a machine learning model" refers to a method or apparatus for using cleaned data to train a model using a specific algorithm (e.g., logistic regression or TF-IDF vectorization).

[0317] "Means for inputting information about new business partners" refers to an interface or device that allows a user to input details about a new business partner (company name, industry, needs, etc.).

[0318] "Means for analyzing input information and automatically generating business negotiation materials" refers to a process or device that analyzes information input by a user and automatically creates optimal business negotiation materials based on that information.

[0319] "Means for generating proposals" refers to the process or device that compares data on new business partners with past data to create the most appropriate proposals.

[0320] The "OCR technology" in "recognizing handwritten memo data as text" refers to optical character recognition technology that converts handwritten character information into digital text data.

[0321] "Means for real-time situation assessment and appropriate response" refers to methods and devices that use collected data and machine learning models to instantly assess the current situation and take appropriate action.

[0322] "Means for recognizing user emotions and adjusting suggestions" refers to a process or device that analyzes the user's emotional state (e.g., information obtained from facial expressions and voice) and adjusts suggestions and interfaces based on that.

[0323] "Means for checking and correcting the generated sales documents and finally outputting the documents" refers to a process or device in which a user checks the automatically generated sales documents, corrects them as necessary, and finally outputs them in PDF or presentation format.

[0324] This invention relates to a system for automatically generating business negotiation materials and a system for optimizing manufacturing processes. By combining this system with an emotion engine that recognizes the user's emotions, it is possible to adjust proposal content and respond appropriately in real time.

[0325] Data collection and preprocessing

[0326] First, the server collects past business materials, photographs, and handwritten notes from the database. For example, it uses an SQL query to execute a command such as "SELECT FROM past_materials" to retrieve this data. This data includes business materials, photographs, and handwritten notes.

[0327] The collected data is cleaned by the server. Unnecessary HTML tags and line breaks are removed from past sales documents, and text data is normalized. Furthermore, image recognition algorithms are used to trim unnecessary parts of photographs, and noise is removed and text is emphasized from handwritten notes.

[0328] Data training

[0329] The server trains a machine learning model based on the cleaned data. Specifically, it performs TF-IDF vectorization to convert the text data into numerical vectors, and then trains the model using algorithms such as logistic regression. This model is used to predict key elements of sales documents and proposal content.

[0330] Entering new business partner data

[0331] The user inputs information about the new business partner (e.g., company name, industry, needs) from the terminal through a dedicated interface. The input information is sent to the server as new business partner data.

[0332] Generate sales documents

[0333] The server receives and analyzes the new business negotiation data sent by the user. Specifically, it converts the new data into text format and performs TF-IDF vectorization. After that, it uses a machine learning model to automatically generate optimal business negotiation materials based on the analysis results.

[0334] Proposal generation

[0335] The server compares the data of the new business partner with past data and generates the most suitable proposal. For example, based on past success stories and proposals, the server incorporates content to propose the optimal solutions and services to the new business partner into the business negotiation materials.

[0336] Analysis of handwritten notes

[0337] The user scans their handwritten notes into their device and sends the data to the server, which then uses OCR technology to extract text from the notes, cleans it, and formats it into a format suitable for analysis.

[0338] Use of emotion engine

[0339] The server analyzes the user's interactions using an emotion engine to recognize the user's emotions in real time. For example, the camera and microphone collect facial expressions and voices as the user uses the input interface, which the emotion engine then analyzes to extract emotional data. The emotional data recognized by the emotion engine is used by the server to adjust the content of suggestions. For example, if the user is tired, the server can maximize the effectiveness of suggestions by summarizing the content of the materials more succinctly or using expressions that encourage relaxation.

[0340] Check and print the final materials

[0341] The user checks the generated business documents on their device and makes corrections as necessary. They check the contents of the documents and make any necessary corrections or missing information. Finally, the server outputs the checked business documents in PDF or presentation format. This allows the user to use the documents in actual business negotiations.

[0342] Application to manufacturing processes

[0343] In the manufacturing process, the server collects and analyzes past manufacturing data and trains a machine learning model. This model is used to monitor the manufacturing process in real time and detect abnormalities early. If an abnormality is detected, the server proposes appropriate countermeasures and issues instructions to robots and human operators. The server also recognizes the emotions of workers and adjusts interface operations and information displays to improve work efficiency and safety.

[0344] Specific examples

[0345] For example, if a new business partner is a technology company seeking cloud solutions, the user inputs that information into their device. The server then analyzes the input signal and references relevant past sales documents, photos, and handwritten notes to automatically generate the optimal proposal content and materials. An emotion engine also analyzes the user's interactions. If the user is excited, it can incorporate responses into the proposal materials that further emphasize the proposal and increase the dynamism of the presentation. Finally, the user reviews the generated sales documents, makes any necessary corrections, and the server outputs them in PDF format. This process ensures that standardized sales documents are provided quickly and efficiently.

[0346] Prompt Sentence Examples

[0347] An example prompt for applying this system to a manufacturing process is:

[0348] "Build a system that detects anomalies in real time based on past manufacturing data and suggests optimal countermeasures."

[0349] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0350] Step 1:

[0351] Data collection

[0352] The server first collects past documents, photographs, and handwritten notes from the database. For example, it uses an SQL query to execute a command such as "SELECT FROM past_materials" to retrieve this data. This data includes business document data, image data, and handwritten notes. The input data is various past documents, and the output data is the set of collected documents.

[0353] Step 2:

[0354] Cleaning the data

[0355] The server cleans the collected data. For example, it removes unnecessary HTML tags and line breaks from past sales documents and normalizes the text data. It also uses image recognition algorithms to trim unnecessary parts of image data, and removes noise and emphasizes characters from handwritten memo data. The input data is the collected data, and the output data is the cleaned data.

[0356] Step 3:

[0357] Training a machine learning model

[0358] The server uses the cleaned data to train a machine learning model. Specifically, it performs TF-IDF vectorization to convert the text data into a numerical vector, and then trains the model using an algorithm such as logistic regression. The input data is the cleaned data, and the output data is the trained model.

[0359] Step 4:

[0360] Entering new business partner data

[0361] The user inputs information about the new business partner (company name, industry, needs) through a dedicated interface on the terminal. The input information is sent to the server as new business partner data. The input data is the new business partner information, and the output data is the data sent from the terminal to the server.

[0362] Step 5:

[0363] Automatic generation of sales documents

[0364] The server receives and analyzes the new business partner data sent by the user. Specifically, it converts the new data into text format and performs TF-IDF vectorization. It then uses a machine learning model to automatically generate optimal business partner documents based on the analysis results. The input data is the new business partner data, and the output data is the generated business partner documents.

[0365] Step 6:

[0366] Proposal generation

[0367] The server compares the new business partner data with past data and generates the most appropriate proposal. For example, based on past success stories and proposal content, the server incorporates content that proposes the optimal solutions and services for the new business partner into the business negotiation materials. The input data is the new business partner data and past data, and the output data is the generated proposal content.

[0368] Step 7:

[0369] Analysis of handwritten notes

[0370] Users scan their handwritten notes onto their devices and send the data to the server. The server uses OCR technology to extract text from the notes, cleans the text, and formats it into a format suitable for analysis. The input data is the handwritten note image, and the output data is the cleaned text data.

[0371] Step 8:

[0372] Use of emotion engine

[0373] The server uses an emotion engine to analyze user interactions and recognize the user's emotions in real time. For example, a camera or microphone can capture facial expressions and voices as the user uses the input interface, which the emotion engine then analyzes to extract emotion data. The input data is the voice and image data from the camera or microphone, and the output data is the extracted emotion data.

[0374] Step 9:

[0375] Check and print the final materials

[0376] The user checks the generated business meeting materials on their terminal and makes corrections as necessary. They check the contents of the business meeting materials and make any necessary corrections or missing information. Finally, the server outputs the checked business meeting materials in PDF or presentation format. The input data is the generated business meeting materials, and the output data is the final output business meeting materials.

[0377] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0378] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0379] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0380] [Second embodiment]

[0381] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0382] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0383] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0384] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0385] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0386] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0387] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0388] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0389] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0390] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0391] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0392] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0393] The present invention relates to a system for automatically generating business negotiation materials, and is implemented as follows: The system is mainly composed of three entities: a server, a terminal, and a user.

[0394] Data collection and preprocessing

[0395] The server first collects past sales materials, photographs, and handwritten notes from the database. For example, the server uses an SQL query to execute a command such as "SELECT FROM past_materials" to retrieve past sales materials. The collected data is then cleaned and formatted to make it suitable for analysis. For example, unnecessary HTML tags and line breaks are removed, and the text data is normalized.

[0396] Data training

[0397] The server trains a machine learning model based on the cleaned historical data. Specifically, it performs TF-IDF vectorization to convert text data into numerical vectors, and then trains the model using algorithms such as logistic regression. This model is used to predict key elements of sales documents and proposal content.

[0398] Entering new business partner data

[0399] The user inputs information about the potential business partner (e.g., company name, industry, and needs) through the interface from a terminal. The input data is sent to the server and analyzed.

[0400] Generate sales documents

[0401] The server analyzes the data of new business partners and automatically generates sales documents based on that data. The server converts the input text data back into TF-IDF vectors and inputs them into a trained model to predict the proposal content. The prediction results are formatted as sales documents according to a template.

[0402] Proposal generation

[0403] The server compares the data of new business partners with past data and generates the most suitable proposals. These proposals are based on past success stories and learning data, allowing for an effective approach to new business partners.

[0404] Analysis of handwritten notes

[0405] The user scans their handwritten notes onto their device and sends them to the server, which then uses OCR technology to extract and clean the text from the handwritten notes, thereby incorporating the contents of the handwritten notes into the business meeting materials.

[0406] Check and print the final materials

[0407] The user can then review the generated sales documents on their device and make any necessary corrections. Finally, the server outputs the reviewed sales documents in PDF or presentation format and provides them to the user. This series of processes ensures that high-quality sales documents are generated efficiently.

[0408] Specific examples

[0409] For example, if a new business partner is a technology company seeking cloud solutions, the user inputs that information into the device. The server then analyzes the input signal and references relevant past business documents, photos, and handwritten notes to automatically generate the optimal proposal and materials. Finally, the user can review the generated business documents and make any necessary corrections, and the server outputs them in PDF format. This process allows for the rapid and efficient provision of standardized business documents.

[0410] The processing flow will be explained below.

[0411] Step 1:

[0412] The server collects past business materials, photographic materials, and handwritten memo data from the database. Specifically, it retrieves this data by executing a command such as "SELECT FROM past_materials" using an SQL query. The retrieved data is saved as business materials data, image data, and handwritten memo data, respectively.

[0413] Step 2:

[0414] The server cleans the collected data. For example, it removes unnecessary HTML tags and line breaks from past sales documents and normalizes the text data. For photo materials, it uses image recognition algorithms to trim unnecessary parts, and removes noise and emphasizes characters from handwritten memo data.

[0415] Step 3:

[0416] The server trains a machine learning model based on the cleaned data. Specifically, it performs TF-IDF vectorization to convert the text data into a numerical vector. It then uses classification algorithms such as logistic regression to train the model on past sales documents, generating a model that can extract the elements necessary for sales documents.

[0417] Step 4:

[0418] The user uses a terminal to input information about the new business partner (e.g., company name, industry, needs) through a dedicated interface. The input information is sent to the server as new business partner data.

[0419] Step 5:

[0420] The server receives and analyzes the new business negotiation data sent by the user. Specifically, it converts the new data into text format and performs TF-IDF vectorization. After that, it uses a machine learning model to automatically generate optimal business negotiation materials based on the analysis results.

[0421] Step 6:

[0422] The server compares the data of the new business partner with past data and generates the most suitable proposal. For example, based on past success stories and proposals, the server incorporates content to propose the optimal solutions and services to the new business partner into the business negotiation materials.

[0423] Step 7:

[0424] The user scans their handwritten notes into their device and sends the data to the server, which then uses OCR technology to extract text from the notes, cleans it, and formats it into a format suitable for analysis.

[0425] Step 8:

[0426] The user checks the generated sales negotiation materials on the terminal and makes corrections as necessary. Checks the contents of the sales negotiation materials and makes corrections if there is missing information or any parts that need to be corrected.

[0427] Step 9:

[0428] The server then outputs the final confirmed business documents in PDF or presentation format and provides them to the user, who can then download them and use them in actual business negotiations.

[0429] Example 1

[0430] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0431] The traditional process of creating sales documents is often done manually, consuming a great deal of time and effort. It is also difficult to effectively utilize past success stories and data, making it difficult to make effective proposals to new business partners. Furthermore, the manual process of incorporating information from handwritten notes into sales documents is cumbersome and inaccurate. There is a need for a system that can improve this current situation and automatically generate efficient, high-quality sales documents.

[0432] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0433] In this invention, the server includes: means for retrieving documents from a database using an SQL query; means for cleaning and analyzing the collected data and deleting unnecessary HTML tags and line breaks; means for performing TF-IDF vectorization to convert the cleaned text data into a numerical vector; means for training a machine learning model using an algorithm such as logistic regression using the TF-IDF vectorized data; means for a user to input information about a new business partner through a terminal; means for analyzing the input information and automatically generating business partner data as business partner documents; means for comparing the input information with past business partner data to generate optimal proposals; means for scanning handwritten memo data entered by the user and converting the converted handwritten memo data into text using OCR technology; means for importing the converted handwritten memo data into business partner documents; means for a user to review and correct the generated business partner documents; and means for outputting the reviewed documents in PDF format or presentation format. This enables efficient automatic generation of high-quality business partner documents.

[0434] An "SQL query" is a syntax for retrieving data from a database.

[0435] A "database" is a system for storing structured data.

[0436] "Cleaning" is the process of removing unnecessary information from data and preparing it in a form suitable for analysis.

[0437] An "HTML tag" is a part of the markup language used to describe the structure of a web page.

[0438] "TF-IDF vectorization" is a method for converting text data into a numerical vector.

[0439] A "numeric vector" is a format in which text data is expressed as numbers.

[0440] A "machine learning model" is an algorithm that learns patterns and relationships from data and makes predictions and classifications.

[0441] "Logistic Regression" is a statistical model for performing binary classification based on given data.

[0442] An "interface" is a screen or input form that allows a user to interact with a system.

[0443] "OCR technology" stands for optical character recognition technology, which is a technology that extracts character data from images.

[0444] "PDF format" is an abbreviation for Portable Document Format, a file format for displaying documents and images in a fixed layout.

[0445] This invention is a system that collects past documents, photographs, and handwritten memo data, and automatically generates business negotiation materials based on these. The system is mainly composed of three entities: a server, a terminal, and a user.

[0446] Data collection and preprocessing

[0447] First, the server uses an SQL query to retrieve past sales documents, photographs, and handwritten notes from the database. An example of a specific SQL query is "SELECT FROM past_materials." The collected data is then cleaned by removing unnecessary HTML tags and line breaks, and normalizing the text data. This prepares the data for analysis.

[0448] Data training

[0449] The cleaned data is transformed into a TF-IDF vector by the server and converted into a numerical vector. A machine learning model is then trained using algorithms such as logistic regression. This creates a model for predicting important elements of sales documents and proposal content.

[0450] Entering new business partner data

[0451] The user uses the terminal to input information about the potential business partner (company name, industry, needs, etc.) through the interface. The input data is sent from the terminal to the server and used for analysis.

[0452] Generate sales documents

[0453] The server analyzes the data of new business partners and automatically generates sales documents based on that data. During this process, TF-IDF vectorization is performed again, and the data is input into a trained model to predict the proposal content. The prediction results are formatted as sales documents based on a template. A specific output may be data embedded in a PowerPoint template.

[0454] Proposal generation

[0455] The server compares the data of the new business partner with past data and generates the most suitable proposal. This proposal is based on past success stories and learning data, making it possible to approach the new business partner effectively.

[0456] Analysis of handwritten notes

[0457] The user scans their handwritten notes onto their device, which then sends them to a server where text data is extracted from the notes using OCR technology (e.g., Google Cloud Vision API), which then cleans the data and incorporates it into the sales documents.

[0458] Check and print the final materials

[0459] The user checks the generated sales documents on their device and makes any necessary corrections. The server then outputs the final, checked sales documents in PDF or presentation format and provides them to the user. This series of processes ensures that high-quality sales documents are generated efficiently.

[0460] Specific examples

[0461] For example, if a new business partner is a technology company seeking a cloud solution, the user inputs that information via their device. The server then analyzes the input signal and references relevant past business documents, photos, and handwritten notes to automatically generate the optimal proposal and materials. Finally, the user can review the generated business documents and make any necessary corrections, after which the server outputs them in PDF format. This process ensures that business documents are provided quickly and efficiently.

[0462] Prompt Sentence Examples

[0463] "Generate the perfect proposal for new business opportunities with technology companies looking for cloud solutions."

[0464] In this invention, by using a generative AI model, negotiation materials are generated automatically and with high accuracy.

[0465] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0466] Step 1: Data collection and preprocessing

[0467] The server first uses an SQL query to retrieve past sales documents, photographs, and handwritten notes from the database. Specifically, it executes the query "SELECT FROM past_materials." The input is various materials in the database, and the output is the data collected by the server. The server then cleans the collected data, removing unnecessary HTML tags and line breaks, and normalizing the text data to make it suitable for analysis.

[0468] Step 2: Train the data

[0469] The server converts the cleaned data into a TF-IDF vector and then into a numerical vector. Next, it trains a machine learning model using algorithms such as logistic regression. The input is the cleaned text data, and the output is a machine learning model that predicts the important elements of sales documents and proposal content. For example, a model can be trained using data from past sales negotiations to build a model that predicts "sales negotiation elements with a high probability of success."

[0470] Step 3: Enter new customer data

[0471] The user inputs information about the new business partner (company name, industry, needs, etc.) through a terminal. This data is sent to the server in JSON format. The input is specific information about the new business partner, and the output is data sent to the server for analysis. For example, the user might input information such as "technology company" and "needs cloud solutions."

[0472] Step 4: Generate sales documents

[0473] The server analyzes data from new business partners, performs TF-IDF vectorization, and then inputs it into a trained model to predict proposal content. The input is information about new business partners, and the output is automatically generated business negotiation materials. As a specific example, a "presentation material proposing the scalability of cloud infrastructure" is generated based on a template.

[0474] Step 5: Generate proposals

[0475] The server compares the data of the new business partner with data from past negotiations, and generates the optimal proposal based on past success stories and related materials. The input is the information of the new business partner and past data, and the output is the proposal. For example, by referring to past successful cases in the same industry, the optimal proposal for the new business partner can be created.

[0476] Step 6: Analyze handwritten notes

[0477] The user scans their handwritten notes onto their device and sends the data to the server. The server then uses OCR technology to extract text data from the handwritten notes and cleans them. The input is the scanned handwritten notes, and the output is the converted handwritten notes. This allows the contents of the handwritten notes to be reflected in the sales documents.

[0478] Step 7: Check and print the final document

[0479] The user checks the generated sales documents on the terminal and makes corrections as necessary. The corrected documents are sent to the server, and the final confirmed sales documents are output in PDF or presentation format. The input is the sales documents corrected by the user, and the output is the final sales documents. For example, the user checks the presentation documents, edits them as necessary, and then saves the final version as a PDF.

[0480] (Application example 1)

[0481] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0482] The traditional process of creating sales documents requires a lot of time and effort, including collecting, cleaning, analyzing, and digitizing handwritten notes from past data and documents. It is also difficult to reflect this information in sales negotiations in real time. This results in lower efficiency in sales negotiations and makes it difficult to quickly provide appropriate proposals to customers.

[0483] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0484] In this invention, the server includes means for collecting past documents, photographic materials, and handwritten memo data, means for cleaning and analyzing the collected data, means for training a machine learning model based on the cleaned data, means for inputting information about new business partners from a user, means for analyzing the input information and automatically generating business negotiation materials, means for comparing the new business partners with past data and generating proposals, means for recognizing handwritten memo data as text and incorporating it into the business negotiation materials, means for checking and correcting the generated business negotiation materials and finally outputting the materials, means for providing the generated business negotiation materials to a user in real time, means for inputting new data through a user interface and immediately reflecting it in the business negotiation materials, and means for recognizing handwritten memo data using OCR technology and adding it to the business negotiation materials in real time, thereby enabling efficient and rapid generation of standardized business negotiation materials and real-time updates.

[0485] "Means of collection" refers to the methods and processes for obtaining past documents, photographic materials, handwritten memo data, etc. from the database and incorporating them into the system.

[0486] "Cleaning" is the process of removing unnecessary information from collected data and shaping it into a form suitable for analysis.

[0487] "Means for analysis" refers to a method for extracting and analyzing meaningful information from the cleaned data.

[0488] "Training" is the process of using cleaned and analyzed data to train a machine learning model.

[0489] "Means for input" refers to the interface and procedures for allowing users to input information about new business partners into the system.

[0490] The "means for automatic generation" is a process for automatically generating negotiation materials based on information input by the user.

[0491] The "means of generation" is a method of comparing information on new business partners with past data and deriving proposal content.

[0492] The "recognition method" is a method of converting handwritten note data into text data using OCR technology.

[0493] "Means for checking and correcting" refers to a procedure by which the user checks the generated business negotiation materials and revise them as necessary.

[0494] The "means of output" refers to the method of providing the final confirmed and revised business documents to the user in PDF or presentation format.

[0495] "Means for providing in real time" refers to a process for instantly presenting the generated business negotiation materials to the user.

[0496] A "means for reflecting" is a mechanism for immediately incorporating new data entered through the user interface into the sales materials.

[0497] "Means of recognizing using OCR technology" refers to a method of converting handwritten note data into digital text using optical character recognition technology.

[0498] "Means of adding in real time" refers to the process of immediately including the converted text data in the sales documents.

[0499] The present invention relates to a system for automatically generating business negotiation materials, and is implemented as follows: The system is mainly composed of three entities: a server, a terminal, and a user.

[0500] Data collection and preprocessing

[0501] The first thing the server does is collect past documents, photographs, and handwritten notes from the database. The server retrieves past business negotiation documents and cleans the collected data. This cleaning process involves removing unnecessary HTML tags and line breaks and normalizing the text data. The Python Pandas library can be used in this process.

[0502] Data training

[0503] The server trains a machine learning model based on the cleaned historical data. Specifically, it performs TF-IDF vectorization and then uses algorithms such as logistic regression to train the model. The Python scikit-learn library can be used. This model is used to predict key elements of sales documents and proposal content.

[0504] Entering new business partner data

[0505] The user inputs information about the potential business partner (e.g., company name, industry, and needs) through the interface from a terminal. The input data is sent to the server and analyzed.

[0506] Generate sales documents

[0507] The server analyzes the data of new business partners and automatically generates sales documents based on that data. The data of new business partners is converted back into TF-IDF vectors and input into the trained model to predict the proposal content. The prediction results are formatted as sales documents according to a template.

[0508] Proposal generation

[0509] The server compares the data of new business partners with past data and generates the most suitable proposals. These proposals are based on past success stories and learning data, allowing for an effective approach to new business partners.

[0510] Analysis of handwritten notes

[0511] The user scans their handwritten notes onto their device and sends them to the server. The server uses OCR technology to extract text from the handwritten notes and cleans them. A Python OCR library (such as Tesseract OCR) can be used. This allows the contents of the handwritten notes to be incorporated into the sales documents.

[0512] Real-time updates and provision of business documents

[0513] The generated sales documents are provided to the user in real time, and new data entered through the user interface is immediately reflected in the sales documents.

[0514] Check and print the final materials

[0515] The user can then review the generated sales documents on their device and make any necessary corrections. Finally, the server outputs the reviewed sales documents in PDF or presentation format and provides them to the user. This series of processes ensures that high-quality sales documents are generated efficiently.

[0516] Prompt Sentence Examples

[0517] Collect, clean, and format your past sales pitch data.

[0518] Next, enter the new contact information and generate the appropriate documents.

[0519] If you have handwritten note images, use OCR technology to convert them into text and include them in the generated document.

[0520] Finally, output the document in PDF format.

[0521] Specific examples

[0522] For example, if the client is a medical device manufacturer, the user inputs the information into the device. The server then analyzes the input data and references relevant past sales documents, photographs, and handwritten notes to automatically generate the optimal proposal and materials. The generated sales documents are provided in real time, and the user can modify them as needed and finally output them in PDF format. This process allows for the rapid and efficient provision of standardized sales documents.

[0523] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0524] Step 1:

[0525] Data collection

[0526] The server retrieves past business negotiation materials, photographic materials, and handwritten memo data from the database. Database connection information and a data collection query are used as input. Specifically, an SQL query is executed to retrieve data. For example, the command "SELECT FROM past_materials" is executed to retrieve past business negotiation materials. The output is the raw data of the retrieved materials and memo data.

[0527] Step 2:

[0528] Cleaning the data

[0529] The server removes unnecessary HTML tags and line breaks from the acquired data and normalizes the text data. It uses the collected raw data as input. Specifically, it uses Pandas and a regular expression library to format the data. For example, it performs processing such as "df['text'] = df['text'].str.replace(r'<[^>]>', '')". The output is cleaned text data.

[0530] Step 3:

[0531] Data training

[0532] The server performs TF-IDF vectorization on the cleaned past data and trains a Logistic Regression model. The cleaned text data is used as input. Specifically, the scikit-learn library is used to perform vectorization and model training. Each process is performed using "tfidf = TfidfVectorizer()" and "model = LogisticRegression()". The output is a trained model.

[0533] Step 4:

[0534] Entering new business partner data

[0535] The user enters information about the new business contact (company name, industry, and needs) through the terminal interface. The input is data manually entered by the user. Specifically, form input and text fields are used. The input new business contact information data is obtained as the output.

[0536] Step 5:

[0537] Analysis of new business contact data

[0538] The server analyzes the entered new client data and converts it back into a TF-IDF vector. The input is the new client information entered by the user. Specifically, the data is converted using the TF-IDF vectorizer defined earlier. Set "X_new = tfidf.transform([new_client_data])". The output is the vectorized new client data.

[0539] Step 6:

[0540] Automatic generation of sales documents

[0541] The server inputs the vectorized new business negotiation data into the trained model and predicts the proposal content. The vectorized new business negotiation data and the trained model are used as input. Specifically, the processing is performed as follows: "prediction = model.predict(X_new)". The output is the sales negotiation document data containing the predicted proposal content.

[0542] Step 7:

[0543] Proposal generation and formatting

[0544] The server formats the sales negotiation materials according to a template based on the predicted proposal content. The predicted proposal content data is used as input. Specifically, the proposal content is applied to the sales negotiation material template to generate a document. The formatted sales negotiation material data is obtained as output.

[0545] Step 8:

[0546] Analysis of handwritten notes

[0547] The user scans handwritten notes onto the device. The server receives the data and extracts text from the handwritten notes using OCR technology. The input is the image data of the handwritten notes. Specifically, the text is extracted using an OCR library (such as Tesseract OCR). The output is the handwritten note data converted into text.

[0548] Step 9:

[0549] Integrate handwritten notes into sales documents

[0550] The server cleans the converted handwritten notes and integrates them into the business negotiation materials. The input is the converted handwritten notes. Specifically, the cleaned data is added to the business negotiation materials, e.g., "formatted_material += ocr_text". The output is the integrated business negotiation materials.

[0551] Step 10:

[0552] Real-time provision of business negotiation materials

[0553] The server provides the generated sales negotiation materials to the user in real time. Every time new data is entered through the user interface, it is reflected in the sales negotiation materials. New user data is used as input. Specifically, data synchronization is performed using WebSocket or real-time communication technology. The latest sales negotiation material data is provided as output in real time.

[0554] Step 11:

[0555] Final confirmation and printing of business documents

[0556] The user checks the generated sales document on their device and makes any necessary corrections. The server then outputs the final, checked document in PDF or presentation format and provides it to the user. The final, checked sales document data is used as input. Specifically, the document is output using a PDF generation library (such as ReportLab). The final sales document in PDF format is obtained as output.

[0557] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0558] This invention relates to a system for automatically generating business negotiation materials, and furthermore, a system for adjusting proposal content by combining an emotion engine that recognizes the user's emotions. The system is mainly composed of four entities: a server, a terminal, a user, and an emotion engine.

[0559] Data collection and preprocessing

[0560] The first thing the server does is collect past business materials, photo materials, and handwritten memo data from the database. For example, the server uses an SQL query to execute a command such as "SELECT FROM past_materials" to retrieve this data. The retrieved data is saved as business materials data, image data, and handwritten memo data, respectively.

[0561] The server then cleans the collected data. For example, unnecessary HTML tags and line breaks are removed from past sales documents and text data is normalized. Image recognition algorithms are used to trim unnecessary parts of photo materials, and noise is removed and text is emphasized from handwritten notes.

[0562] Data training

[0563] The server trains a machine learning model based on the cleaned historical data. Specifically, it performs TF-IDF vectorization to convert text data into numerical vectors, and then trains the model using algorithms such as logistic regression. This model is used to predict key elements of sales documents and proposal content.

[0564] Entering new business partner data

[0565] The user inputs information about the new business partner (e.g., company name, industry, needs) from the terminal through a dedicated interface. The input information is sent to the server as new business partner data.

[0566] Generate sales documents

[0567] The server receives and analyzes the new business negotiation data sent by the user. Specifically, it converts the new data into text format and performs TF-IDF vectorization. After that, it uses a machine learning model to automatically generate optimal business negotiation materials based on the analysis results.

[0568] Proposal generation

[0569] The server compares the data of the new business partner with past data and generates the most suitable proposal. For example, based on past success stories and proposals, the server incorporates content to propose the optimal solutions and services to the new business partner into the business negotiation materials.

[0570] Analysis of handwritten notes

[0571] The user scans their handwritten notes into their device and sends the data to the server, which then uses OCR technology to extract text from the notes, cleans it, and formats it into a format suitable for analysis.

[0572] Use of emotion engine

[0573] The emotion engine analyzes user interactions and recognizes user emotions in real time. For example, facial expressions and voices are collected by a camera or microphone when the user uses an input interface, and the emotion engine analyzes them to extract emotional data.

[0574] The emotion data recognized by the emotion engine is used by the server to adjust the content of suggestions. For example, if the user is tired, the server can maximize the effectiveness of suggestions by summarizing the content of the materials more succinctly or using expressions that encourage relaxation.

[0575] Check and print the final materials

[0576] The user checks the generated business negotiation materials on the terminal and makes corrections as necessary. Checks the contents of the business negotiation materials and makes corrections if there is missing information or if there are any parts that need to be corrected.

[0577] Finally, the server outputs the confirmed business documents in PDF or presentation format, allowing the user to use the documents in actual business negotiations.

[0578] Specific examples

[0579] For example, if a new business partner is a technology company seeking cloud solutions, the user inputs that information into their device. The server then analyzes the input signal and references relevant past sales documents, photos, and handwritten notes to automatically generate the optimal proposal content and materials. An emotion engine also analyzes the user's interactions. If the user is excited, it can incorporate responses into the proposal materials that further emphasize the proposal and increase the dynamism of the presentation. Finally, the user reviews the generated sales documents, makes any necessary corrections, and the server outputs them in PDF format. This process ensures that standardized sales documents are provided quickly and efficiently.

[0580] The processing flow will be explained below.

[0581] Step 1:

[0582] The server collects past business materials, photo materials, and handwritten memo data from the database. Specifically, it executes a command such as "SELECT FROM past_materials" using an SQL query, and saves this data as business materials data, image data, and handwritten memo data, respectively.

[0583] Step 2:

[0584] The server cleans the collected data. For example, it removes unnecessary HTML tags and line breaks from past sales documents and normalizes the text data. It also uses image recognition algorithms to trim unnecessary parts of photographs, and removes noise and emphasizes characters from handwritten notes.

[0585] Step 3:

[0586] The server trains a machine learning model based on the cleaned data. Specifically, it performs TF-IDF vectorization to convert the text data into numerical vectors, and then trains the model using algorithms such as logistic regression. This model is used to predict important elements of sales documents and proposal content.

[0587] Step 4:

[0588] The user uses a terminal to input information about the new business partner (e.g., company name, industry, needs) through a dedicated interface. The input information is sent to the server as new business partner data.

[0589] Step 5:

[0590] The server receives and analyzes the new business negotiation data sent by the user. Specifically, it converts the new data into text format and performs TF-IDF vectorization. After that, it uses a machine learning model to automatically generate optimal business negotiation materials based on the analysis results.

[0591] Step 6:

[0592] The server compares the data of the new business partner with past data and generates the most suitable proposal. For example, based on past success stories and proposals, the server incorporates content to propose the optimal solutions and services to the new business partner into the business negotiation materials.

[0593] Step 7:

[0594] The user scans their handwritten notes into their device and sends the data to the server, which then uses OCR technology to extract text from the notes, cleans it, and formats it into a format suitable for analysis.

[0595] Step 8:

[0596] When a user inputs emotion data through the interface, the emotion engine analyzes the user's facial expressions and voice. The emotion engine analyzes the interaction and extracts the user's emotion data. For example, it can recognize the user's emotion in real time using a camera or microphone.

[0597] Step 9:

[0598] The server receives the user's emotional data recognized by the emotion engine and adjusts the content of the proposal. Specifically, if the user is excited, the proposal content is emphasized, while if the user is tired, the materials are summarized more succinctly. This allows the server to provide optimal business negotiation materials according to the user's psychological state.

[0599] Step 10:

[0600] The user checks the generated sales negotiation materials on the terminal and makes corrections as necessary. Checks the contents of the sales negotiation materials and makes corrections if there is missing information or any parts that need to be corrected.

[0601] Step 11:

[0602] The server then outputs the final confirmed business documents in PDF or presentation format and provides them to the user, who can then use the documents in actual business negotiations.

[0603] Example 2

[0604] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0605] The current process of creating sales documents requires manual input and data organization, which consumes a lot of time and effort. In addition, the proposals are often suboptimal because they do not take into account the user's emotional state.

[0606] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0607] In this invention, the server includes means for collecting past documents, image data, and handwritten information data, means for cleaning and analyzing the collected data, and means for training a machine learning model based on the cleaned data. This enables efficient data collection and preprocessing. The server also includes means for a user to input information about a new business partner, means for analyzing the input information and automatically generating business negotiation materials, means for comparing the data about the new business partner with past data and generating proposal content, means for recognizing handwritten information data as character data and incorporating it into the business negotiation materials, means for checking and correcting the generated business negotiation materials and finally outputting the materials, and means for analyzing the user's emotions and adjusting the proposal content based on the emotions. This enables automatic generation of business negotiation materials and adjustment of the proposal content based on the user's emotions.

[0608] "Past documents" are previously created records including business documents, image data, and handwritten information data.

[0609] "Image data" is digital data that contains visual information such as photographic material.

[0610] "Handwritten information data" is data in digital form obtained from handwritten notes or documents.

[0611] "Cleaning" refers to removing unnecessary elements from data and preparing it in a format suitable for analysis.

[0612] A "machine learning model" is an algorithm that learns patterns and relationships in data and makes predictions and classifications based on new data.

[0613] "Information on new business partners" refers to attribute data on new customers or companies that are the subject of business negotiations, such as company name, industry, and needs.

[0614] "Business documents" refers to presentation materials and proposals used to conduct business negotiations.

[0615] "Comparison" refers to comparing new data with previous data to identify similarities and differences.

[0616] The "proposal content" is information that includes an outline of the solution or service to be provided to the business partner.

[0617] "Character data" is digital data that represents information in text format.

[0618] "Optical character recognition technology (OCR)" is a technology that converts handwritten or printed character data into digital text.

[0619] "Emotion analysis" refers to the process of identifying a user's emotional state from their facial expressions and voice.

[0620] "Adjustment" refers to modifying content or behavior to suit specific conditions or circumstances.

[0621] The present invention relates to a system that combines the automatic generation of business negotiation materials with the recognition of user emotions. This system is composed of four main components: a server, a terminal, a user, and an emotion engine.

[0622] Data collection and preprocessing

[0623] The server first collects past sales documents, image data, and handwritten information data from the database. It retrieves the data using an SQL query (e.g., "SELECT FROM past_materials") and saves it in a folder corresponding to each data type. After this process, the server cleans the collected data. Specifically, it uses regular expressions to remove unnecessary HTML tags and line breaks from the text data and normalizes the text. For image data, Python's Pillow library is used to trim unnecessary parts, and for handwritten information, an OCR tool (e.g., Tesseract) is used to extract text and remove noise.

[0624] Data training

[0625] The server trains a machine learning model based on the cleaned data. It converts the text data into TF-IDF vectors, then uses the scikit-learn library to build a logistic regression model and learns the key elements for generating sales documents.

[0626] Entering new business partner data

[0627] The user inputs information about the new business contact (company name, industry, needs, etc.) through a dedicated interface on the terminal. The terminal receives the user's input using GUI components (e.g., text boxes, drop-down lists) and sends it to the server.

[0628] Generate sales documents

[0629] The server receives and analyzes the data of new business partners sent by the user. Specifically, it converts the new data into text format and converts it into TF-IDF vectors. It then uses the trained model to automatically generate optimal business documents.

[0630] Proposal generation

[0631] The server compares the data of the new business partner with past data and generates the optimal proposal for the new business partner based on past success stories and proposal content. This process may use natural language generation tools (e.g., GPT-3).

[0632] Analysis of handwritten notes

[0633] Users scan their handwritten notes onto the device, which uses OCR software to extract the text from the handwritten notes, and then send this data to a server that cleans and formats it for analysis.

[0634] Use of emotion engine

[0635] The emotion engine analyzes the user's interactions. The user's facial expressions and voice are collected by the device's camera and microphone, and the emotion engine analyzes them to extract emotional data. This data is sent to the server and used to adjust the content of the suggestions. For example, if the user is tired, the content of the materials can be summarized more concisely.

[0636] Check and print the final materials

[0637] The user checks the generated sales documents on the terminal. The user checks the documents in the GUI and makes any necessary corrections. Finally, the server outputs the checked sales documents in PDF or presentation format. The user downloads them and uses them in actual sales negotiations.

[0638] Specific examples

[0639] For example, if a new business partner is a technology company seeking a cloud solution, the user enters that information on their device. The server then analyzes the input data and automatically generates the optimal proposal by referencing relevant past business documents, image data, and handwritten notes. In addition, if the user expresses excitement, the emotion engine analyzes this and adds dynamism to further emphasize the proposal. The user can then review the final business document and make any necessary corrections, after which the server outputs it in PDF format.

[0640] Prompt Sentence Examples

[0641] "Please enter the company name, industry, and needs of your new business partner into the terminal."

[0642] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0643] Step 1: Data collection and preprocessing

[0644] The server collects past sales documents, image data, and handwritten information data from a database. The input is an SQL query (e.g., "SELECT FROM past_materials"), and the output is data saved in a folder corresponding to each data type. The server then cleans the collected data. Specifically, it uses regular expressions to remove unnecessary HTML tags and line breaks from the text data and normalizes the text. It also uses the Pillow library to trim unnecessary parts of the image data, and an OCR tool (e.g., Tesseract) to extract text and remove noise from the handwritten information data. This results in data in a format suitable for analysis.

[0645] Step 2: Train the data

[0646] The server trains a machine learning model using the cleaned data. The input is the cleaned text data, and the output is the trained model. First, the text data is converted into TF-IDF vectors, and then a logistic regression model is built using the scikit-learn library. This model learns the elements necessary for generating sales documents.

[0647] Step 3: Enter new customer data

[0648] The user inputs information about the new business contact (company name, industry, needs, etc.) through a dedicated interface on the terminal. The input is a GUI component (e.g., text box, drop-down list), and the output is the information about the new business contact that is sent to the server. The terminal receives this data and sends it to the server.

[0649] Step 4: Generate sales documents

[0650] The server receives and analyzes the data of new business partners sent by the user. The input is the information of new business partners, and the output is the generated business documents. Specifically, the new data is converted into text format and TF-IDF vectorized. The trained model is then used to automatically generate the optimal business documents.

[0651] Step 5: Generate proposals

[0652] The server compares the data of the new business partner with past data and generates an appropriate proposal. The input is the data of the new business partner, past success stories, and proposals, and the output is the proposal to be incorporated into the business materials. This process may use natural language generation tools (e.g., GPT-3).

[0653] Step 6: Analyze handwritten notes

[0654] A user scans their handwritten notes into a device, which then uses OCR software to extract text from them. The input is the scanned image of the handwritten notes, and the output is text data sent to a server, which cleans and formats the text data for analysis.

[0655] Step 7: Use the Emotion Engine

[0656] The emotion engine analyzes the user's interactions and recognizes their emotions. The input is the user's facial expressions and voice, and the output is emotional data. Data collected by the device's camera and microphone is analyzed by the emotion engine, and the extracted emotional data is sent to the server. The server adjusts the content of the suggestions based on this data. For example, if the system recognizes that the user is tired, it may make adjustments such as summarizing the contents of the materials more concisely.

[0657] Step 8: Check and print the final document

[0658] The user checks the sales negotiation materials generated on the terminal and makes corrections as necessary. The input is the generated sales negotiation materials, and the output is the final confirmed sales negotiation materials. The materials are displayed on the GUI, and missing information and corrections are made. Finally, the server outputs the confirmed sales negotiation materials in PDF or presentation format. The user downloads them and uses them in actual sales negotiations.

[0659] (Application example 2)

[0660] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0661] In an automatic generation system for sales documents, there is a demand for the ability to quickly and efficiently generate optimal proposals for new business partners based on past data, but conventional systems have had difficulty responding in real time or adjusting based on user emotions.In addition to sales documents, it is also important to make real-time situational assessments and take appropriate action in the manufacturing process, but there has been a lack of systems that can do this.

[0662] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting past documents, photographic materials, and handwritten memo data, means for cleaning and analyzing the collected data, means for training a machine learning model based on the cleaned data, means for inputting information about new business partners from a user, means for analyzing the input information and automatically generating business negotiation materials, means for comparing the new business partner with past data and generating proposal content, means for recognizing handwritten memo data as text and incorporating it into the business negotiation materials, means for assessing the situation in real time and taking appropriate action, means for recognizing the user's emotions and adjusting the proposal content, and means for confirming and correcting the generated business negotiation materials and finally outputting the materials. This makes it possible to realize a system that efficiently generates business negotiation materials and enables adjustments and real-time responses based on the user's emotions.

[0663] "Past materials" is a general term for information collected in the past, such as business negotiation materials, manufacturing records, photographic materials, and handwritten memo data.

[0664] "Means of collection" refers to the methods and devices used to obtain historical documents, photographic materials, and handwritten notes from databases and other sources.

[0665] "Cleaning and analyzing means" refers to a method or device that performs processing such as removing unnecessary parts from collected data, normalizing text data, or removing noise from image data.

[0666] "Means for training a machine learning model" refers to a method or apparatus for using cleaned data to train a model using a specific algorithm (e.g., logistic regression or TF-IDF vectorization).

[0667] "Means for inputting information about new business partners" refers to an interface or device that allows a user to input details about a new business partner (company name, industry, needs, etc.).

[0668] "Means for analyzing input information and automatically generating business negotiation materials" refers to a process or device that analyzes information input by a user and automatically creates optimal business negotiation materials based on that information.

[0669] "Means for generating proposals" refers to the process or device that compares data on new business partners with past data to create the most appropriate proposals.

[0670] The "OCR technology" in "recognizing handwritten memo data as text" refers to optical character recognition technology that converts handwritten character information into digital text data.

[0671] "Means for real-time situation assessment and appropriate response" refers to methods and devices that use collected data and machine learning models to instantly assess the current situation and take appropriate action.

[0672] "Means for recognizing user emotions and adjusting suggestions" refers to a process or device that analyzes the user's emotional state (e.g., information obtained from facial expressions and voice) and adjusts suggestions and interfaces based on that.

[0673] "Means for checking and correcting the generated sales documents and finally outputting the documents" refers to a process or device in which a user checks the automatically generated sales documents, corrects them as necessary, and finally outputs them in PDF or presentation format.

[0674] This invention relates to a system for automatically generating business negotiation materials and a system for optimizing manufacturing processes. By combining this system with an emotion engine that recognizes the user's emotions, it is possible to adjust proposal content and respond appropriately in real time.

[0675] Data collection and preprocessing

[0676] First, the server collects past business materials, photographs, and handwritten notes from the database. For example, it uses an SQL query to execute a command such as "SELECT FROM past_materials" to retrieve this data. This data includes business materials, photographs, and handwritten notes.

[0677] The collected data is cleaned by the server. Unnecessary HTML tags and line breaks are removed from past sales documents, and text data is normalized. Furthermore, image recognition algorithms are used to trim unnecessary parts of photographs, and noise is removed and text is emphasized from handwritten notes.

[0678] Data training

[0679] The server trains a machine learning model based on the cleaned data. Specifically, it performs TF-IDF vectorization to convert the text data into numerical vectors, and then trains the model using algorithms such as logistic regression. This model is used to predict key elements of sales documents and proposal content.

[0680] Entering new business partner data

[0681] The user inputs information about the new business partner (e.g., company name, industry, needs) from the terminal through a dedicated interface. The input information is sent to the server as new business partner data.

[0682] Generate sales documents

[0683] The server receives and analyzes the new business negotiation data sent by the user. Specifically, it converts the new data into text format and performs TF-IDF vectorization. After that, it uses a machine learning model to automatically generate optimal business negotiation materials based on the analysis results.

[0684] Proposal generation

[0685] The server compares the data of the new business partner with past data and generates the most suitable proposal. For example, based on past success stories and proposals, the server incorporates content to propose the optimal solutions and services to the new business partner into the business negotiation materials.

[0686] Analysis of handwritten notes

[0687] The user scans their handwritten notes into their device and sends the data to the server, which then uses OCR technology to extract text from the notes, cleans it, and formats it into a format suitable for analysis.

[0688] Use of emotion engine

[0689] The server analyzes the user's interactions using an emotion engine to recognize the user's emotions in real time. For example, the camera and microphone collect facial expressions and voices as the user uses the input interface, which the emotion engine then analyzes to extract emotional data. The emotional data recognized by the emotion engine is used by the server to adjust the content of suggestions. For example, if the user is tired, the server can maximize the effectiveness of suggestions by summarizing the content of the materials more succinctly or using expressions that encourage relaxation.

[0690] Check and print the final materials

[0691] The user checks the generated business documents on their device and makes corrections as necessary. They check the contents of the documents and make any necessary corrections or missing information. Finally, the server outputs the checked business documents in PDF or presentation format. This allows the user to use the documents in actual business negotiations.

[0692] Application to manufacturing processes

[0693] In the manufacturing process, the server collects and analyzes past manufacturing data and trains a machine learning model. This model is used to monitor the manufacturing process in real time and detect abnormalities early. If an abnormality is detected, the server proposes appropriate countermeasures and issues instructions to robots and human operators. The server also recognizes the emotions of workers and adjusts interface operations and information displays to improve work efficiency and safety.

[0694] Specific examples

[0695] For example, if a new business partner is a technology company seeking cloud solutions, the user inputs that information into their device. The server then analyzes the input signal and references relevant past sales documents, photos, and handwritten notes to automatically generate the optimal proposal content and materials. An emotion engine also analyzes the user's interactions. If the user is excited, it can incorporate responses into the proposal materials that further emphasize the proposal and increase the dynamism of the presentation. Finally, the user reviews the generated sales documents, makes any necessary corrections, and the server outputs them in PDF format. This process ensures that standardized sales documents are provided quickly and efficiently.

[0696] Prompt Sentence Examples

[0697] An example prompt for applying this system to a manufacturing process is:

[0698] "Build a system that detects anomalies in real time based on past manufacturing data and suggests optimal countermeasures."

[0699] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0700] Step 1:

[0701] Data collection

[0702] The server first collects past documents, photographs, and handwritten notes from the database. For example, it uses an SQL query to execute a command such as "SELECT FROM past_materials" to retrieve this data. This data includes business document data, image data, and handwritten notes. The input data is various past documents, and the output data is the set of collected documents.

[0703] Step 2:

[0704] Cleaning the data

[0705] The server cleans the collected data. For example, it removes unnecessary HTML tags and line breaks from past sales documents and normalizes the text data. It also uses image recognition algorithms to trim unnecessary parts of image data, and removes noise and emphasizes characters from handwritten memo data. The input data is the collected data, and the output data is the cleaned data.

[0706] Step 3:

[0707] Training a machine learning model

[0708] The server uses the cleaned data to train a machine learning model. Specifically, it performs TF-IDF vectorization to convert the text data into a numerical vector, and then trains the model using an algorithm such as logistic regression. The input data is the cleaned data, and the output data is the trained model.

[0709] Step 4:

[0710] Entering new business partner data

[0711] The user inputs information about the new business partner (company name, industry, needs) through a dedicated interface on the terminal. The input information is sent to the server as new business partner data. The input data is the new business partner information, and the output data is the data sent from the terminal to the server.

[0712] Step 5:

[0713] Automatic generation of sales documents

[0714] The server receives and analyzes the new business partner data sent by the user. Specifically, it converts the new data into text format and performs TF-IDF vectorization. It then uses a machine learning model to automatically generate optimal business partner documents based on the analysis results. The input data is the new business partner data, and the output data is the generated business partner documents.

[0715] Step 6:

[0716] Proposal generation

[0717] The server compares the new business partner data with past data and generates the most appropriate proposal. For example, based on past success stories and proposal content, the server incorporates content that proposes the optimal solutions and services for the new business partner into the business negotiation materials. The input data is the new business partner data and past data, and the output data is the generated proposal content.

[0718] Step 7:

[0719] Analysis of handwritten notes

[0720] Users scan their handwritten notes onto their devices and send the data to the server. The server uses OCR technology to extract text from the notes, cleans the text, and formats it into a format suitable for analysis. The input data is the handwritten note image, and the output data is the cleaned text data.

[0721] Step 8:

[0722] Use of emotion engine

[0723] The server uses an emotion engine to analyze user interactions and recognize the user's emotions in real time. For example, a camera or microphone can capture facial expressions and voices as the user uses the input interface, which the emotion engine then analyzes to extract emotion data. The input data is the voice and image data from the camera or microphone, and the output data is the extracted emotion data.

[0724] Step 9:

[0725] Check and print the final materials

[0726] The user checks the generated business meeting materials on their terminal and makes corrections as necessary. They check the contents of the business meeting materials and make any necessary corrections or missing information. Finally, the server outputs the checked business meeting materials in PDF or presentation format. The input data is the generated business meeting materials, and the output data is the final output business meeting materials.

[0727] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0728] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0729] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0730] [Third embodiment]

[0731] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0732] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0733] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0734] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0735] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0736] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0737] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0738] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0739] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0740] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0741] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0742] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0743] The present invention relates to a system for automatically generating business negotiation materials, and is implemented as follows: The system is mainly composed of three entities: a server, a terminal, and a user.

[0744] Data collection and preprocessing

[0745] The server first collects past sales materials, photographs, and handwritten notes from the database. For example, the server uses an SQL query to execute a command such as "SELECT FROM past_materials" to retrieve past sales materials. The collected data is then cleaned and formatted to make it suitable for analysis. For example, unnecessary HTML tags and line breaks are removed, and the text data is normalized.

[0746] Data training

[0747] The server trains a machine learning model based on the cleaned historical data. Specifically, it performs TF-IDF vectorization to convert text data into numerical vectors, and then trains the model using algorithms such as logistic regression. This model is used to predict key elements of sales documents and proposal content.

[0748] Entering new business partner data

[0749] The user inputs information about the potential business partner (e.g., company name, industry, and needs) through the interface from a terminal. The input data is sent to the server and analyzed.

[0750] Generate sales documents

[0751] The server analyzes the data of new business partners and automatically generates sales documents based on that data. The server converts the input text data back into TF-IDF vectors and inputs them into a trained model to predict the proposal content. The prediction results are formatted as sales documents according to a template.

[0752] Proposal generation

[0753] The server compares the data of new business partners with past data and generates the most suitable proposals. These proposals are based on past success stories and learning data, allowing for an effective approach to new business partners.

[0754] Analysis of handwritten notes

[0755] The user scans their handwritten notes onto their device and sends them to the server, which then uses OCR technology to extract and clean the text from the handwritten notes, thereby incorporating the contents of the handwritten notes into the business meeting materials.

[0756] Check and print the final materials

[0757] The user can then review the generated sales documents on their device and make any necessary corrections. Finally, the server outputs the reviewed sales documents in PDF or presentation format and provides them to the user. This series of processes ensures that high-quality sales documents are generated efficiently.

[0758] Specific examples

[0759] For example, if a new business partner is a technology company seeking cloud solutions, the user inputs that information into the device. The server then analyzes the input signal and references relevant past business documents, photos, and handwritten notes to automatically generate the optimal proposal and materials. Finally, the user can review the generated business documents and make any necessary corrections, and the server outputs them in PDF format. This process allows for the rapid and efficient provision of standardized business documents.

[0760] The processing flow will be explained below.

[0761] Step 1:

[0762] The server collects past business materials, photographic materials, and handwritten memo data from the database. Specifically, it retrieves this data by executing a command such as "SELECT FROM past_materials" using an SQL query. The retrieved data is saved as business materials data, image data, and handwritten memo data, respectively.

[0763] Step 2:

[0764] The server cleans the collected data. For example, it removes unnecessary HTML tags and line breaks from past sales documents and normalizes the text data. For photo materials, it uses image recognition algorithms to trim unnecessary parts, and removes noise and emphasizes characters from handwritten memo data.

[0765] Step 3:

[0766] The server trains a machine learning model based on the cleaned data. Specifically, it performs TF-IDF vectorization to convert the text data into a numerical vector. It then uses classification algorithms such as logistic regression to train the model on past sales documents, generating a model that can extract the elements necessary for sales documents.

[0767] Step 4:

[0768] The user uses a terminal to input information about the new business partner (e.g., company name, industry, needs) through a dedicated interface. The input information is sent to the server as new business partner data.

[0769] Step 5:

[0770] The server receives and analyzes the new business negotiation data sent by the user. Specifically, it converts the new data into text format and performs TF-IDF vectorization. After that, it uses a machine learning model to automatically generate optimal business negotiation materials based on the analysis results.

[0771] Step 6:

[0772] The server compares the data of the new business partner with past data and generates the most suitable proposal. For example, based on past success stories and proposals, the server incorporates content to propose the optimal solutions and services to the new business partner into the business negotiation materials.

[0773] Step 7:

[0774] The user scans their handwritten notes into their device and sends the data to the server, which then uses OCR technology to extract text from the notes, cleans it, and formats it into a format suitable for analysis.

[0775] Step 8:

[0776] The user checks the generated sales negotiation materials on the terminal and makes corrections as necessary. Checks the contents of the sales negotiation materials and makes corrections if there is missing information or any parts that need to be corrected.

[0777] Step 9:

[0778] The server then outputs the final confirmed business documents in PDF or presentation format and provides them to the user, who can then download them and use them in actual business negotiations.

[0779] Example 1

[0780] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0781] The traditional process of creating sales documents is often done manually, consuming a great deal of time and effort. It is also difficult to effectively utilize past success stories and data, making it difficult to make effective proposals to new business partners. Furthermore, the manual process of incorporating information from handwritten notes into sales documents is cumbersome and inaccurate. There is a need for a system that can improve this current situation and automatically generate efficient, high-quality sales documents.

[0782] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0783] In this invention, the server includes: means for retrieving documents from a database using an SQL query; means for cleaning and analyzing the collected data and deleting unnecessary HTML tags and line breaks; means for performing TF-IDF vectorization to convert the cleaned text data into a numerical vector; means for training a machine learning model using an algorithm such as logistic regression using the TF-IDF vectorized data; means for a user to input information about a new business partner through a terminal; means for analyzing the input information and automatically generating business partner data as business partner documents; means for comparing the input information with past business partner data to generate optimal proposals; means for scanning handwritten memo data entered by the user and converting the converted handwritten memo data into text using OCR technology; means for importing the converted handwritten memo data into business partner documents; means for a user to review and correct the generated business partner documents; and means for outputting the reviewed documents in PDF format or presentation format. This enables efficient automatic generation of high-quality business partner documents.

[0784] An "SQL query" is a syntax for retrieving data from a database.

[0785] A "database" is a system for storing structured data.

[0786] "Cleaning" is the process of removing unnecessary information from data and preparing it in a form suitable for analysis.

[0787] An "HTML tag" is a part of the markup language used to describe the structure of a web page.

[0788] "TF-IDF vectorization" is a method for converting text data into a numerical vector.

[0789] A "numeric vector" is a format in which text data is expressed as numbers.

[0790] A "machine learning model" is an algorithm that learns patterns and relationships from data and makes predictions and classifications.

[0791] "Logistic Regression" is a statistical model for performing binary classification based on given data.

[0792] An "interface" is a screen or input form that allows a user to interact with a system.

[0793] "OCR technology" stands for optical character recognition technology, which is a technology that extracts character data from images.

[0794] "PDF format" is an abbreviation for Portable Document Format, a file format for displaying documents and images in a fixed layout.

[0795] This invention is a system that collects past documents, photographs, and handwritten memo data, and automatically generates business negotiation materials based on these. The system is mainly composed of three entities: a server, a terminal, and a user.

[0796] Data collection and preprocessing

[0797] First, the server uses an SQL query to retrieve past sales documents, photographs, and handwritten notes from the database. An example of a specific SQL query is "SELECT FROM past_materials." The collected data is then cleaned by removing unnecessary HTML tags and line breaks, and normalizing the text data. This prepares the data for analysis.

[0798] Data training

[0799] The cleaned data is transformed into a TF-IDF vector by the server and converted into a numerical vector. A machine learning model is then trained using algorithms such as logistic regression. This creates a model for predicting important elements of sales documents and proposal content.

[0800] Entering new business partner data

[0801] The user uses the terminal to input information about the potential business partner (company name, industry, needs, etc.) through the interface. The input data is sent from the terminal to the server and used for analysis.

[0802] Generate sales documents

[0803] The server analyzes the data of new business partners and automatically generates sales documents based on that data. During this process, TF-IDF vectorization is performed again, and the data is input into a trained model to predict the proposal content. The prediction results are formatted as sales documents based on a template. A specific output may be data embedded in a PowerPoint template.

[0804] Proposal generation

[0805] The server compares the data of the new business partner with past data and generates the most suitable proposal. This proposal is based on past success stories and learning data, making it possible to approach the new business partner effectively.

[0806] Analysis of handwritten notes

[0807] The user scans their handwritten notes onto their device, which then sends them to a server where text data is extracted from the notes using OCR technology (e.g., Google Cloud Vision API), which then cleans the data and incorporates it into the sales documents.

[0808] Check and print the final materials

[0809] The user checks the generated sales documents on their device and makes any necessary corrections. The server then outputs the final, checked sales documents in PDF or presentation format and provides them to the user. This series of processes ensures that high-quality sales documents are generated efficiently.

[0810] Specific examples

[0811] For example, if a new business partner is a technology company seeking a cloud solution, the user inputs that information via their device. The server then analyzes the input signal and references relevant past business documents, photos, and handwritten notes to automatically generate the optimal proposal and materials. Finally, the user can review the generated business documents and make any necessary corrections, after which the server outputs them in PDF format. This process ensures that business documents are provided quickly and efficiently.

[0812] Prompt Sentence Examples

[0813] "Generate the perfect proposal for new business opportunities with technology companies looking for cloud solutions."

[0814] In this invention, by using a generative AI model, negotiation materials are generated automatically and with high accuracy.

[0815] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0816] Step 1: Data collection and preprocessing

[0817] The server first uses an SQL query to retrieve past sales documents, photographs, and handwritten notes from the database. Specifically, it executes the query "SELECT FROM past_materials." The input is various materials in the database, and the output is the data collected by the server. The server then cleans the collected data, removing unnecessary HTML tags and line breaks, and normalizing the text data to make it suitable for analysis.

[0818] Step 2: Train the data

[0819] The server converts the cleaned data into a TF-IDF vector and then into a numerical vector. Next, it trains a machine learning model using algorithms such as logistic regression. The input is the cleaned text data, and the output is a machine learning model that predicts the important elements of sales documents and proposal content. For example, a model can be trained using data from past sales negotiations to build a model that predicts "sales negotiation elements with a high probability of success."

[0820] Step 3: Enter new customer data

[0821] The user inputs information about the new business partner (company name, industry, needs, etc.) through a terminal. This data is sent to the server in JSON format. The input is specific information about the new business partner, and the output is data sent to the server for analysis. For example, the user might input information such as "technology company" and "needs cloud solutions."

[0822] Step 4: Generate sales documents

[0823] The server analyzes data from new business partners, performs TF-IDF vectorization, and then inputs it into a trained model to predict proposal content. The input is information about new business partners, and the output is automatically generated business negotiation materials. As a specific example, a "presentation material proposing the scalability of cloud infrastructure" is generated based on a template.

[0824] Step 5: Generate proposals

[0825] The server compares the data of the new business partner with data from past negotiations, and generates the optimal proposal based on past success stories and related materials. The input is the information of the new business partner and past data, and the output is the proposal. For example, by referring to past successful cases in the same industry, the optimal proposal for the new business partner can be created.

[0826] Step 6: Analyze handwritten notes

[0827] The user scans their handwritten notes onto their device and sends the data to the server. The server then uses OCR technology to extract text data from the handwritten notes and cleans them. The input is the scanned handwritten notes, and the output is the converted handwritten notes. This allows the contents of the handwritten notes to be reflected in the sales documents.

[0828] Step 7: Check and print the final document

[0829] The user checks the generated sales documents on the terminal and makes corrections as necessary. The corrected documents are sent to the server, and the final confirmed sales documents are output in PDF or presentation format. The input is the sales documents corrected by the user, and the output is the final sales documents. For example, the user checks the presentation documents, edits them as necessary, and then saves the final version as a PDF.

[0830] (Application example 1)

[0831] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0832] The traditional process of creating sales documents requires a lot of time and effort, including collecting, cleaning, analyzing, and digitizing handwritten notes from past data and documents. It is also difficult to reflect this information in sales negotiations in real time. This results in lower efficiency in sales negotiations and makes it difficult to quickly provide appropriate proposals to customers.

[0833] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0834] In this invention, the server includes means for collecting past documents, photographic materials, and handwritten memo data, means for cleaning and analyzing the collected data, means for training a machine learning model based on the cleaned data, means for inputting information about new business partners from a user, means for analyzing the input information and automatically generating business negotiation materials, means for comparing the new business partners with past data and generating proposals, means for recognizing handwritten memo data as text and incorporating it into the business negotiation materials, means for checking and correcting the generated business negotiation materials and finally outputting the materials, means for providing the generated business negotiation materials to a user in real time, means for inputting new data through a user interface and immediately reflecting it in the business negotiation materials, and means for recognizing handwritten memo data using OCR technology and adding it to the business negotiation materials in real time, thereby enabling efficient and rapid generation of standardized business negotiation materials and real-time updates.

[0835] "Means of collection" refers to the methods and processes for obtaining past documents, photographic materials, handwritten memo data, etc. from the database and incorporating them into the system.

[0836] "Cleaning" is the process of removing unnecessary information from collected data and shaping it into a form suitable for analysis.

[0837] "Means for analysis" refers to a method for extracting and analyzing meaningful information from the cleaned data.

[0838] "Training" is the process of using cleaned and analyzed data to train a machine learning model.

[0839] "Means for input" refers to the interface and procedures for allowing users to input information about new business partners into the system.

[0840] The "means for automatic generation" is a process for automatically generating negotiation materials based on information input by the user.

[0841] The "means of generation" is a method of comparing information on new business partners with past data and deriving proposal content.

[0842] The "recognition method" is a method of converting handwritten note data into text data using OCR technology.

[0843] "Means for checking and correcting" refers to a procedure by which the user checks the generated business negotiation materials and revise them as necessary.

[0844] The "means of output" refers to the method of providing the final confirmed and revised business documents to the user in PDF or presentation format.

[0845] "Means for providing in real time" refers to a process for instantly presenting the generated business negotiation materials to the user.

[0846] A "means for reflecting" is a mechanism for immediately incorporating new data entered through the user interface into the sales materials.

[0847] "Means of recognizing using OCR technology" refers to a method of converting handwritten note data into digital text using optical character recognition technology.

[0848] "Means of adding in real time" refers to the process of immediately including the converted text data in the sales documents.

[0849] The present invention relates to a system for automatically generating business negotiation materials, and is implemented as follows: The system is mainly composed of three entities: a server, a terminal, and a user.

[0850] Data collection and preprocessing

[0851] The first thing the server does is collect past documents, photographs, and handwritten notes from the database. The server retrieves past business negotiation documents and cleans the collected data. This cleaning process involves removing unnecessary HTML tags and line breaks and normalizing the text data. The Python Pandas library can be used in this process.

[0852] Data training

[0853] The server trains a machine learning model based on the cleaned historical data. Specifically, it performs TF-IDF vectorization and then uses algorithms such as logistic regression to train the model. The Python scikit-learn library can be used. This model is used to predict key elements of sales documents and proposal content.

[0854] Entering new business partner data

[0855] The user inputs information about the potential business partner (e.g., company name, industry, and needs) through the interface from a terminal. The input data is sent to the server and analyzed.

[0856] Generate sales documents

[0857] The server analyzes the data of new business partners and automatically generates sales documents based on that data. The data of new business partners is converted back into TF-IDF vectors and input into the trained model to predict the proposal content. The prediction results are formatted as sales documents according to a template.

[0858] Proposal generation

[0859] The server compares the data of new business partners with past data and generates the most suitable proposals. These proposals are based on past success stories and learning data, allowing for an effective approach to new business partners.

[0860] Analysis of handwritten notes

[0861] The user scans their handwritten notes onto their device and sends them to the server. The server uses OCR technology to extract text from the handwritten notes and cleans them. A Python OCR library (such as Tesseract OCR) can be used. This allows the contents of the handwritten notes to be incorporated into the sales documents.

[0862] Real-time updates and provision of business documents

[0863] The generated sales documents are provided to the user in real time, and new data entered through the user interface is immediately reflected in the sales documents.

[0864] Check and print the final materials

[0865] The user can then review the generated sales documents on their device and make any necessary corrections. Finally, the server outputs the reviewed sales documents in PDF or presentation format and provides them to the user. This series of processes ensures that high-quality sales documents are generated efficiently.

[0866] Prompt Sentence Examples

[0867] Collect, clean, and format your past sales pitch data.

[0868] Next, enter the new contact information and generate the appropriate documents.

[0869] If you have handwritten note images, use OCR technology to convert them into text and include them in the generated document.

[0870] Finally, output the document in PDF format.

[0871] Specific examples

[0872] For example, if the client is a medical device manufacturer, the user inputs the information into the device. The server then analyzes the input data and references relevant past sales documents, photographs, and handwritten notes to automatically generate the optimal proposal and materials. The generated sales documents are provided in real time, and the user can modify them as needed and finally output them in PDF format. This process allows for the rapid and efficient provision of standardized sales documents.

[0873] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0874] Step 1:

[0875] Data collection

[0876] The server retrieves past business negotiation materials, photographic materials, and handwritten memo data from the database. Database connection information and a data collection query are used as input. Specifically, an SQL query is executed to retrieve data. For example, the command "SELECT FROM past_materials" is executed to retrieve past business negotiation materials. The output is the raw data of the retrieved materials and memo data.

[0877] Step 2:

[0878] Cleaning the data

[0879] The server removes unnecessary HTML tags and line breaks from the acquired data and normalizes the text data. It uses the collected raw data as input. Specifically, it uses Pandas and a regular expression library to format the data. For example, it performs processing such as "df['text'] = df['text'].str.replace(r'<[^>]>', '')". The output is cleaned text data.

[0880] Step 3:

[0881] Data training

[0882] The server performs TF-IDF vectorization on the cleaned past data and trains a Logistic Regression model. The cleaned text data is used as input. Specifically, the scikit-learn library is used to perform vectorization and model training. Each process is performed using "tfidf = TfidfVectorizer()" and "model = LogisticRegression()". The output is a trained model.

[0883] Step 4:

[0884] Entering new business partner data

[0885] The user enters information about the new business contact (company name, industry, and needs) through the terminal interface. The input is data manually entered by the user. Specifically, form input and text fields are used. The input new business contact information data is obtained as the output.

[0886] Step 5:

[0887] Analysis of new business contact data

[0888] The server analyzes the entered new client data and converts it back into a TF-IDF vector. The input is the new client information entered by the user. Specifically, the data is converted using the TF-IDF vectorizer defined earlier. Set "X_new = tfidf.transform([new_client_data])". The output is the vectorized new client data.

[0889] Step 6:

[0890] Automatic generation of sales documents

[0891] The server inputs the vectorized new business negotiation data into the trained model and predicts the proposal content. The vectorized new business negotiation data and the trained model are used as input. Specifically, the processing is performed as follows: "prediction = model.predict(X_new)". The output is the sales negotiation document data containing the predicted proposal content.

[0892] Step 7:

[0893] Proposal generation and formatting

[0894] The server formats the sales negotiation materials according to a template based on the predicted proposal content. The predicted proposal content data is used as input. Specifically, the proposal content is applied to the sales negotiation material template to generate a document. The formatted sales negotiation material data is obtained as output.

[0895] Step 8:

[0896] Analysis of handwritten notes

[0897] The user scans handwritten notes onto the device. The server receives the data and extracts text from the handwritten notes using OCR technology. The input is the image data of the handwritten notes. Specifically, the text is extracted using an OCR library (such as Tesseract OCR). The output is the handwritten note data converted into text.

[0898] Step 9:

[0899] Integrate handwritten notes into sales documents

[0900] The server cleans the converted handwritten notes and integrates them into the business negotiation materials. The input is the converted handwritten notes. Specifically, the cleaned data is added to the business negotiation materials, e.g., "formatted_material += ocr_text". The output is the integrated business negotiation materials.

[0901] Step 10:

[0902] Real-time provision of business negotiation materials

[0903] The server provides the generated sales negotiation materials to the user in real time. Every time new data is entered through the user interface, it is reflected in the sales negotiation materials. New user data is used as input. Specifically, data synchronization is performed using WebSocket or real-time communication technology. The latest sales negotiation material data is provided as output in real time.

[0904] Step 11:

[0905] Final confirmation and printing of business documents

[0906] The user checks the generated sales document on their device and makes any necessary corrections. The server then outputs the final, checked document in PDF or presentation format and provides it to the user. The final, checked sales document data is used as input. Specifically, the document is output using a PDF generation library (such as ReportLab). The final sales document in PDF format is obtained as output.

[0907] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0908] This invention relates to a system for automatically generating business negotiation materials, and furthermore, a system for adjusting proposal content by combining an emotion engine that recognizes the user's emotions. The system is mainly composed of four entities: a server, a terminal, a user, and an emotion engine.

[0909] Data collection and preprocessing

[0910] The first thing the server does is collect past business materials, photo materials, and handwritten memo data from the database. For example, the server uses an SQL query to execute a command such as "SELECT FROM past_materials" to retrieve this data. The retrieved data is saved as business materials data, image data, and handwritten memo data, respectively.

[0911] The server then cleans the collected data. For example, unnecessary HTML tags and line breaks are removed from past sales documents and text data is normalized. Image recognition algorithms are used to trim unnecessary parts of photo materials, and noise is removed and text is emphasized from handwritten notes.

[0912] Data training

[0913] The server trains a machine learning model based on the cleaned historical data. Specifically, it performs TF-IDF vectorization to convert text data into numerical vectors, and then trains the model using algorithms such as logistic regression. This model is used to predict key elements of sales documents and proposal content.

[0914] Entering new business partner data

[0915] The user inputs information about the new business partner (e.g., company name, industry, needs) from the terminal through a dedicated interface. The input information is sent to the server as new business partner data.

[0916] Generate sales documents

[0917] The server receives and analyzes the new business negotiation data sent by the user. Specifically, it converts the new data into text format and performs TF-IDF vectorization. After that, it uses a machine learning model to automatically generate optimal business negotiation materials based on the analysis results.

[0918] Proposal generation

[0919] The server compares the data of the new business partner with past data and generates the most suitable proposal. For example, based on past success stories and proposals, the server incorporates content to propose the optimal solutions and services to the new business partner into the business negotiation materials.

[0920] Analysis of handwritten notes

[0921] The user scans their handwritten notes into their device and sends the data to the server, which then uses OCR technology to extract text from the notes, cleans it, and formats it into a format suitable for analysis.

[0922] Use of emotion engine

[0923] The emotion engine analyzes user interactions and recognizes user emotions in real time. For example, facial expressions and voices are collected by a camera or microphone when the user uses an input interface, and the emotion engine analyzes them to extract emotional data.

[0924] The emotion data recognized by the emotion engine is used by the server to adjust the content of suggestions. For example, if the user is tired, the server can maximize the effectiveness of suggestions by summarizing the content of the materials more succinctly or using expressions that encourage relaxation.

[0925] Check and print the final materials

[0926] The user checks the generated business negotiation materials on the terminal and makes corrections as necessary. Checks the contents of the business negotiation materials and makes corrections if there is missing information or if there are any parts that need to be corrected.

[0927] Finally, the server outputs the confirmed business documents in PDF or presentation format, allowing the user to use the documents in actual business negotiations.

[0928] Specific examples

[0929] For example, if a new business partner is a technology company seeking cloud solutions, the user inputs that information into their device. The server then analyzes the input signal and references relevant past sales documents, photos, and handwritten notes to automatically generate the optimal proposal content and materials. An emotion engine also analyzes the user's interactions. If the user is excited, it can incorporate responses into the proposal materials that further emphasize the proposal and increase the dynamism of the presentation. Finally, the user reviews the generated sales documents, makes any necessary corrections, and the server outputs them in PDF format. This process ensures that standardized sales documents are provided quickly and efficiently.

[0930] The processing flow will be explained below.

[0931] Step 1:

[0932] The server collects past business materials, photo materials, and handwritten memo data from the database. Specifically, it executes a command such as "SELECT FROM past_materials" using an SQL query, and saves this data as business materials data, image data, and handwritten memo data, respectively.

[0933] Step 2:

[0934] The server cleans the collected data. For example, it removes unnecessary HTML tags and line breaks from past sales documents and normalizes the text data. It also uses image recognition algorithms to trim unnecessary parts of photographs, and removes noise and emphasizes characters from handwritten notes.

[0935] Step 3:

[0936] The server trains a machine learning model based on the cleaned data. Specifically, it performs TF-IDF vectorization to convert the text data into numerical vectors, and then trains the model using algorithms such as logistic regression. This model is used to predict important elements of sales documents and proposal content.

[0937] Step 4:

[0938] The user uses a terminal to input information about the new business partner (e.g., company name, industry, needs) through a dedicated interface. The input information is sent to the server as new business partner data.

[0939] Step 5:

[0940] The server receives and analyzes the new business negotiation data sent by the user. Specifically, it converts the new data into text format and performs TF-IDF vectorization. After that, it uses a machine learning model to automatically generate optimal business negotiation materials based on the analysis results.

[0941] Step 6:

[0942] The server compares the data of the new business partner with past data and generates the most suitable proposal. For example, based on past success stories and proposals, the server incorporates content to propose the optimal solutions and services to the new business partner into the business negotiation materials.

[0943] Step 7:

[0944] The user scans their handwritten notes into their device and sends the data to the server, which then uses OCR technology to extract text from the notes, cleans it, and formats it into a format suitable for analysis.

[0945] Step 8:

[0946] When a user inputs emotion data through the interface, the emotion engine analyzes the user's facial expressions and voice. The emotion engine analyzes the interaction and extracts the user's emotion data. For example, it can recognize the user's emotion in real time using a camera or microphone.

[0947] Step 9:

[0948] The server receives the user's emotional data recognized by the emotion engine and adjusts the content of the proposal. Specifically, if the user is excited, the proposal content is emphasized, while if the user is tired, the materials are summarized more succinctly. This allows the server to provide optimal business negotiation materials according to the user's psychological state.

[0949] Step 10:

[0950] The user checks the generated sales negotiation materials on the terminal and makes corrections as necessary. Checks the contents of the sales negotiation materials and makes corrections if there is missing information or any parts that need to be corrected.

[0951] Step 11:

[0952] The server then outputs the final confirmed business documents in PDF or presentation format and provides them to the user, who can then use the documents in actual business negotiations.

[0953] Example 2

[0954] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0955] The current process of creating sales documents requires manual input and data organization, which consumes a lot of time and effort. In addition, the proposals are often suboptimal because they do not take into account the user's emotional state.

[0956] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0957] In this invention, the server includes means for collecting past documents, image data, and handwritten information data, means for cleaning and analyzing the collected data, and means for training a machine learning model based on the cleaned data. This enables efficient data collection and preprocessing. The server also includes means for a user to input information about a new business partner, means for analyzing the input information and automatically generating business negotiation materials, means for comparing the data about the new business partner with past data and generating proposal content, means for recognizing handwritten information data as character data and incorporating it into the business negotiation materials, means for checking and correcting the generated business negotiation materials and finally outputting the materials, and means for analyzing the user's emotions and adjusting the proposal content based on the emotions. This enables automatic generation of business negotiation materials and adjustment of the proposal content based on the user's emotions.

[0958] "Past documents" are previously created records including business documents, image data, and handwritten information data.

[0959] "Image data" is digital data that contains visual information such as photographic material.

[0960] "Handwritten information data" is data in digital form obtained from handwritten notes or documents.

[0961] "Cleaning" refers to removing unnecessary elements from data and preparing it in a format suitable for analysis.

[0962] A "machine learning model" is an algorithm that learns patterns and relationships in data and makes predictions and classifications based on new data.

[0963] "Information on new business partners" refers to attribute data on new customers or companies that are the subject of business negotiations, such as company name, industry, and needs.

[0964] "Business documents" refers to presentation materials and proposals used to conduct business negotiations.

[0965] "Comparison" refers to comparing new data with previous data to identify similarities and differences.

[0966] The "proposal content" is information that includes an outline of the solution or service to be provided to the business partner.

[0967] "Character data" is digital data that represents information in text format.

[0968] "Optical character recognition technology (OCR)" is a technology that converts handwritten or printed character data into digital text.

[0969] "Emotion analysis" refers to the process of identifying a user's emotional state from their facial expressions and voice.

[0970] "Adjustment" refers to modifying content or behavior to suit specific conditions or circumstances.

[0971] The present invention relates to a system that combines the automatic generation of business negotiation materials with the recognition of user emotions. This system is composed of four main components: a server, a terminal, a user, and an emotion engine.

[0972] Data collection and preprocessing

[0973] The server first collects past sales documents, image data, and handwritten information data from the database. It retrieves the data using an SQL query (e.g., "SELECT FROM past_materials") and saves it in a folder corresponding to each data type. After this process, the server cleans the collected data. Specifically, it uses regular expressions to remove unnecessary HTML tags and line breaks from the text data and normalizes the text. For image data, Python's Pillow library is used to trim unnecessary parts, and for handwritten information, an OCR tool (e.g., Tesseract) is used to extract text and remove noise.

[0974] Data training

[0975] The server trains a machine learning model based on the cleaned data. It converts the text data into TF-IDF vectors, then uses the scikit-learn library to build a logistic regression model and learns the key elements for generating sales documents.

[0976] Entering new business partner data

[0977] The user inputs information about the new business contact (company name, industry, needs, etc.) through a dedicated interface on the terminal. The terminal receives the user's input using GUI components (e.g., text boxes, drop-down lists) and sends it to the server.

[0978] Generate sales documents

[0979] The server receives and analyzes the data of new business partners sent by the user. Specifically, it converts the new data into text format and converts it into TF-IDF vectors. It then uses the trained model to automatically generate optimal business documents.

[0980] Proposal generation

[0981] The server compares the data of the new business partner with past data and generates the optimal proposal for the new business partner based on past success stories and proposal content. This process may use natural language generation tools (e.g., GPT-3).

[0982] Analysis of handwritten notes

[0983] Users scan their handwritten notes onto the device, which uses OCR software to extract the text from the handwritten notes, and then send this data to a server that cleans and formats it for analysis.

[0984] Use of emotion engine

[0985] The emotion engine analyzes the user's interactions. The user's facial expressions and voice are collected by the device's camera and microphone, and the emotion engine analyzes them to extract emotional data. This data is sent to the server and used to adjust the content of the suggestions. For example, if the user is tired, the content of the materials can be summarized more concisely.

[0986] Check and print the final materials

[0987] The user checks the generated sales documents on the terminal. The user checks the documents in the GUI and makes any necessary corrections. Finally, the server outputs the checked sales documents in PDF or presentation format. The user downloads them and uses them in actual sales negotiations.

[0988] Specific examples

[0989] For example, if a new business partner is a technology company seeking a cloud solution, the user enters that information on their device. The server then analyzes the input data and automatically generates the optimal proposal by referencing relevant past business documents, image data, and handwritten notes. In addition, if the user expresses excitement, the emotion engine analyzes this and adds dynamism to further emphasize the proposal. The user can then review the final business document and make any necessary corrections, after which the server outputs it in PDF format.

[0990] Prompt Sentence Examples

[0991] "Please enter the company name, industry, and needs of your new business partner into the terminal."

[0992] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0993] Step 1: Data collection and preprocessing

[0994] The server collects past sales documents, image data, and handwritten information data from a database. The input is an SQL query (e.g., "SELECT FROM past_materials"), and the output is data saved in a folder corresponding to each data type. The server then cleans the collected data. Specifically, it uses regular expressions to remove unnecessary HTML tags and line breaks from the text data and normalizes the text. It also uses the Pillow library to trim unnecessary parts of the image data, and an OCR tool (e.g., Tesseract) to extract text and remove noise from the handwritten information data. This results in data in a format suitable for analysis.

[0995] Step 2: Train the data

[0996] The server trains a machine learning model using the cleaned data. The input is the cleaned text data, and the output is the trained model. First, the text data is converted into TF-IDF vectors, and then a logistic regression model is built using the scikit-learn library. This model learns the elements necessary for generating sales documents.

[0997] Step 3: Enter new customer data

[0998] The user inputs information about the new business contact (company name, industry, needs, etc.) through a dedicated interface on the terminal. The input is a GUI component (e.g., text box, drop-down list), and the output is the information about the new business contact that is sent to the server. The terminal receives this data and sends it to the server.

[0999] Step 4: Generate sales documents

[1000] The server receives and analyzes the data of new business partners sent by the user. The input is the information of new business partners, and the output is the generated business documents. Specifically, the new data is converted into text format and TF-IDF vectorized. The trained model is then used to automatically generate the optimal business documents.

[1001] Step 5: Generate proposals

[1002] The server compares the data of the new business partner with past data and generates an appropriate proposal. The input is the data of the new business partner, past success stories, and proposals, and the output is the proposal to be incorporated into the business materials. This process may use natural language generation tools (e.g., GPT-3).

[1003] Step 6: Analyze handwritten notes

[1004] A user scans their handwritten notes into a device, which then uses OCR software to extract text from them. The input is the scanned image of the handwritten notes, and the output is text data sent to a server, which cleans and formats the text data for analysis.

[1005] Step 7: Use the Emotion Engine

[1006] The emotion engine analyzes the user's interactions and recognizes their emotions. The input is the user's facial expressions and voice, and the output is emotional data. Data collected by the device's camera and microphone is analyzed by the emotion engine, and the extracted emotional data is sent to the server. The server adjusts the content of the suggestions based on this data. For example, if the system recognizes that the user is tired, it may make adjustments such as summarizing the contents of the materials more concisely.

[1007] Step 8: Check and print the final document

[1008] The user checks the sales negotiation materials generated on the terminal and makes corrections as necessary. The input is the generated sales negotiation materials, and the output is the final confirmed sales negotiation materials. The materials are displayed on the GUI, and missing information and corrections are made. Finally, the server outputs the confirmed sales negotiation materials in PDF or presentation format. The user downloads them and uses them in actual sales negotiations.

[1009] (Application example 2)

[1010] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1011] In an automatic generation system for sales documents, there is a demand for the ability to quickly and efficiently generate optimal proposals for new business partners based on past data, but conventional systems have had difficulty responding in real time or adjusting based on user emotions.In addition to sales documents, it is also important to make real-time situational assessments and take appropriate action in the manufacturing process, but there has been a lack of systems that can do this.

[1012] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting past documents, photographic materials, and handwritten memo data, means for cleaning and analyzing the collected data, means for training a machine learning model based on the cleaned data, means for inputting information about new business partners from a user, means for analyzing the input information and automatically generating business negotiation materials, means for comparing the new business partner with past data and generating proposal content, means for recognizing handwritten memo data as text and incorporating it into the business negotiation materials, means for assessing the situation in real time and taking appropriate action, means for recognizing the user's emotions and adjusting the proposal content, and means for confirming and correcting the generated business negotiation materials and finally outputting the materials. This makes it possible to realize a system that efficiently generates business negotiation materials and enables adjustments and real-time responses based on the user's emotions.

[1013] "Past materials" is a general term for information collected in the past, such as business negotiation materials, manufacturing records, photographic materials, and handwritten memo data.

[1014] "Means of collection" refers to the methods and devices used to obtain historical documents, photographic materials, and handwritten notes from databases and other sources.

[1015] "Cleaning and analyzing means" refers to a method or device that performs processing such as removing unnecessary parts from collected data, normalizing text data, or removing noise from image data.

[1016] "Means for training a machine learning model" refers to a method or apparatus for using cleaned data to train a model using a specific algorithm (e.g., logistic regression or TF-IDF vectorization).

[1017] "Means for inputting information about new business partners" refers to an interface or device that allows a user to input details about a new business partner (company name, industry, needs, etc.).

[1018] "Means for analyzing input information and automatically generating business negotiation materials" refers to a process or device that analyzes information input by a user and automatically creates optimal business negotiation materials based on that information.

[1019] "Means for generating proposals" refers to the process or device that compares data on new business partners with past data to create the most appropriate proposals.

[1020] The "OCR technology" in "recognizing handwritten memo data as text" refers to optical character recognition technology that converts handwritten character information into digital text data.

[1021] "Means for real-time situation assessment and appropriate response" refers to methods and devices that use collected data and machine learning models to instantly assess the current situation and take appropriate action.

[1022] "Means for recognizing user emotions and adjusting suggestions" refers to a process or device that analyzes the user's emotional state (e.g., information obtained from facial expressions and voice) and adjusts suggestions and interfaces based on that.

[1023] "Means for checking and correcting the generated sales documents and finally outputting the documents" refers to a process or device in which a user checks the automatically generated sales documents, corrects them as necessary, and finally outputs them in PDF or presentation format.

[1024] This invention relates to a system for automatically generating business negotiation materials and a system for optimizing manufacturing processes. By combining this system with an emotion engine that recognizes the user's emotions, it is possible to adjust proposal content and respond appropriately in real time.

[1025] Data collection and preprocessing

[1026] First, the server collects past business materials, photographs, and handwritten notes from the database. For example, it uses an SQL query to execute a command such as "SELECT FROM past_materials" to retrieve this data. This data includes business materials, photographs, and handwritten notes.

[1027] The collected data is cleaned by the server. Unnecessary HTML tags and line breaks are removed from past sales documents, and text data is normalized. Furthermore, image recognition algorithms are used to trim unnecessary parts of photographs, and noise is removed and text is emphasized from handwritten notes.

[1028] Data training

[1029] The server trains a machine learning model based on the cleaned data. Specifically, it performs TF-IDF vectorization to convert the text data into numerical vectors, and then trains the model using algorithms such as logistic regression. This model is used to predict key elements of sales documents and proposal content.

[1030] Entering new business partner data

[1031] The user inputs information about the new business partner (e.g., company name, industry, needs) from the terminal through a dedicated interface. The input information is sent to the server as new business partner data.

[1032] Generate sales documents

[1033] The server receives and analyzes the new business negotiation data sent by the user. Specifically, it converts the new data into text format and performs TF-IDF vectorization. After that, it uses a machine learning model to automatically generate optimal business negotiation materials based on the analysis results.

[1034] Proposal generation

[1035] The server compares the data of the new business partner with past data and generates the most suitable proposal. For example, based on past success stories and proposals, the server incorporates content to propose the optimal solutions and services to the new business partner into the business negotiation materials.

[1036] Analysis of handwritten notes

[1037] The user scans their handwritten notes into their device and sends the data to the server, which then uses OCR technology to extract text from the notes, cleans it, and formats it into a format suitable for analysis.

[1038] Use of emotion engine

[1039] The server analyzes the user's interactions using an emotion engine to recognize the user's emotions in real time. For example, the camera and microphone collect facial expressions and voices as the user uses the input interface, which the emotion engine then analyzes to extract emotional data. The emotional data recognized by the emotion engine is used by the server to adjust the content of suggestions. For example, if the user is tired, the server can maximize the effectiveness of suggestions by summarizing the content of the materials more succinctly or using expressions that encourage relaxation.

[1040] Check and print the final materials

[1041] The user checks the generated business documents on their device and makes corrections as necessary. They check the contents of the documents and make any necessary corrections or missing information. Finally, the server outputs the checked business documents in PDF or presentation format. This allows the user to use the documents in actual business negotiations.

[1042] Application to manufacturing processes

[1043] In the manufacturing process, the server collects and analyzes past manufacturing data and trains a machine learning model. This model is used to monitor the manufacturing process in real time and detect abnormalities early. If an abnormality is detected, the server proposes appropriate countermeasures and issues instructions to robots and human operators. The server also recognizes the emotions of workers and adjusts interface operations and information displays to improve work efficiency and safety.

[1044] Specific examples

[1045] For example, if a new business partner is a technology company seeking cloud solutions, the user inputs that information into their device. The server then analyzes the input signal and references relevant past sales documents, photos, and handwritten notes to automatically generate the optimal proposal content and materials. An emotion engine also analyzes the user's interactions. If the user is excited, it can incorporate responses into the proposal materials that further emphasize the proposal and increase the dynamism of the presentation. Finally, the user reviews the generated sales documents, makes any necessary corrections, and the server outputs them in PDF format. This process ensures that standardized sales documents are provided quickly and efficiently.

[1046] Prompt Sentence Examples

[1047] An example prompt for applying this system to a manufacturing process is:

[1048] "Build a system that detects anomalies in real time based on past manufacturing data and suggests optimal countermeasures."

[1049] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1050] Step 1:

[1051] Data collection

[1052] The server first collects past documents, photographs, and handwritten notes from the database. For example, it uses an SQL query to execute a command such as "SELECT FROM past_materials" to retrieve this data. This data includes business document data, image data, and handwritten notes. The input data is various past documents, and the output data is the set of collected documents.

[1053] Step 2:

[1054] Cleaning the data

[1055] The server cleans the collected data. For example, it removes unnecessary HTML tags and line breaks from past sales documents and normalizes the text data. It also uses image recognition algorithms to trim unnecessary parts of image data, and removes noise and emphasizes characters from handwritten memo data. The input data is the collected data, and the output data is the cleaned data.

[1056] Step 3:

[1057] Training a machine learning model

[1058] The server uses the cleaned data to train a machine learning model. Specifically, it performs TF-IDF vectorization to convert the text data into a numerical vector, and then trains the model using an algorithm such as logistic regression. The input data is the cleaned data, and the output data is the trained model.

[1059] Step 4:

[1060] Entering new business partner data

[1061] The user inputs information about the new business partner (company name, industry, needs) through a dedicated interface on the terminal. The input information is sent to the server as new business partner data. The input data is the new business partner information, and the output data is the data sent from the terminal to the server.

[1062] Step 5:

[1063] Automatic generation of sales documents

[1064] The server receives and analyzes the new business partner data sent by the user. Specifically, it converts the new data into text format and performs TF-IDF vectorization. It then uses a machine learning model to automatically generate optimal business partner documents based on the analysis results. The input data is the new business partner data, and the output data is the generated business partner documents.

[1065] Step 6:

[1066] Proposal generation

[1067] The server compares the new business partner data with past data and generates the most appropriate proposal. For example, based on past success stories and proposal content, the server incorporates content that proposes the optimal solutions and services for the new business partner into the business negotiation materials. The input data is the new business partner data and past data, and the output data is the generated proposal content.

[1068] Step 7:

[1069] Analysis of handwritten notes

[1070] Users scan their handwritten notes onto their devices and send the data to the server. The server uses OCR technology to extract text from the notes, cleans the text, and formats it into a format suitable for analysis. The input data is the handwritten note image, and the output data is the cleaned text data.

[1071] Step 8:

[1072] Use of emotion engine

[1073] The server uses an emotion engine to analyze user interactions and recognize the user's emotions in real time. For example, a camera or microphone can capture facial expressions and voices as the user uses the input interface, which the emotion engine then analyzes to extract emotion data. The input data is the voice and image data from the camera or microphone, and the output data is the extracted emotion data.

[1074] Step 9:

[1075] Check and print the final materials

[1076] The user checks the generated business meeting materials on their terminal and makes corrections as necessary. They check the contents of the business meeting materials and make any necessary corrections or missing information. Finally, the server outputs the checked business meeting materials in PDF or presentation format. The input data is the generated business meeting materials, and the output data is the final output business meeting materials.

[1077] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1078] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1079] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1080] [Fourth embodiment]

[1081] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1082] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1083] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1084] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1085] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1086] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1087] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1088] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1089] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1090] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1091] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1092] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1093] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1094] The present invention relates to a system for automatically generating business negotiation materials, and is implemented as follows: The system is mainly composed of three entities: a server, a terminal, and a user.

[1095] Data collection and preprocessing

[1096] The server first collects past sales materials, photographs, and handwritten notes from the database. For example, the server uses an SQL query to execute a command such as "SELECT FROM past_materials" to retrieve past sales materials. The collected data is then cleaned and formatted to make it suitable for analysis. For example, unnecessary HTML tags and line breaks are removed, and the text data is normalized.

[1097] Data training

[1098] The server trains a machine learning model based on the cleaned historical data. Specifically, it performs TF-IDF vectorization to convert text data into numerical vectors, and then trains the model using algorithms such as logistic regression. This model is used to predict key elements of sales documents and proposal content.

[1099] Entering new business partner data

[1100] The user inputs information about the potential business partner (e.g., company name, industry, and needs) through the interface from a terminal. The input data is sent to the server and analyzed.

[1101] Generate sales documents

[1102] The server analyzes the data of new business partners and automatically generates sales documents based on that data. The server converts the input text data back into TF-IDF vectors and inputs them into a trained model to predict the proposal content. The prediction results are formatted as sales documents according to a template.

[1103] Proposal generation

[1104] The server compares the data of new business partners with past data and generates the most suitable proposals. These proposals are based on past success stories and learning data, allowing for an effective approach to new business partners.

[1105] Analysis of handwritten notes

[1106] The user scans their handwritten notes onto their device and sends them to the server, which then uses OCR technology to extract and clean the text from the handwritten notes, thereby incorporating the contents of the handwritten notes into the business meeting materials.

[1107] Check and print the final materials

[1108] The user can then review the generated sales documents on their device and make any necessary corrections. Finally, the server outputs the reviewed sales documents in PDF or presentation format and provides them to the user. This series of processes ensures that high-quality sales documents are generated efficiently.

[1109] Specific examples

[1110] For example, if a new business partner is a technology company seeking cloud solutions, the user inputs that information into the device. The server then analyzes the input signal and references relevant past business documents, photos, and handwritten notes to automatically generate the optimal proposal and materials. Finally, the user can review the generated business documents and make any necessary corrections, and the server outputs them in PDF format. This process allows for the rapid and efficient provision of standardized business documents.

[1111] The processing flow will be explained below.

[1112] Step 1:

[1113] The server collects past business materials, photographic materials, and handwritten memo data from the database. Specifically, it retrieves this data by executing a command such as "SELECT FROM past_materials" using an SQL query. The retrieved data is saved as business materials data, image data, and handwritten memo data, respectively.

[1114] Step 2:

[1115] The server cleans the collected data. For example, it removes unnecessary HTML tags and line breaks from past sales documents and normalizes the text data. For photo materials, it uses image recognition algorithms to trim unnecessary parts, and removes noise and emphasizes characters from handwritten memo data.

[1116] Step 3:

[1117] The server trains a machine learning model based on the cleaned data. Specifically, it performs TF-IDF vectorization to convert the text data into a numerical vector. It then uses classification algorithms such as logistic regression to train the model on past sales documents, generating a model that can extract the elements necessary for sales documents.

[1118] Step 4:

[1119] The user uses a terminal to input information about the new business partner (e.g., company name, industry, needs) through a dedicated interface. The input information is sent to the server as new business partner data.

[1120] Step 5:

[1121] The server receives and analyzes the new business negotiation data sent by the user. Specifically, it converts the new data into text format and performs TF-IDF vectorization. After that, it uses a machine learning model to automatically generate optimal business negotiation materials based on the analysis results.

[1122] Step 6:

[1123] The server compares the data of the new business partner with past data and generates the most suitable proposal. For example, based on past success stories and proposals, the server incorporates content to propose the optimal solutions and services to the new business partner into the business negotiation materials.

[1124] Step 7:

[1125] The user scans their handwritten notes into their device and sends the data to the server, which then uses OCR technology to extract text from the notes, cleans it, and formats it into a format suitable for analysis.

[1126] Step 8:

[1127] The user checks the generated sales negotiation materials on the terminal and makes corrections as necessary. Checks the contents of the sales negotiation materials and makes corrections if there is missing information or any parts that need to be corrected.

[1128] Step 9:

[1129] The server then outputs the final confirmed business documents in PDF or presentation format and provides them to the user, who can then download them and use them in actual business negotiations.

[1130] Example 1

[1131] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1132] The traditional process of creating sales documents is often done manually, consuming a great deal of time and effort. It is also difficult to effectively utilize past success stories and data, making it difficult to make effective proposals to new business partners. Furthermore, the manual process of incorporating information from handwritten notes into sales documents is cumbersome and inaccurate. There is a need for a system that can improve this current situation and automatically generate efficient, high-quality sales documents.

[1133] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1134] In this invention, the server includes: means for retrieving documents from a database using an SQL query; means for cleaning and analyzing the collected data and deleting unnecessary HTML tags and line breaks; means for performing TF-IDF vectorization to convert the cleaned text data into a numerical vector; means for training a machine learning model using an algorithm such as logistic regression using the TF-IDF vectorized data; means for a user to input information about a new business partner through a terminal; means for analyzing the input information and automatically generating business partner data as business partner documents; means for comparing the input information with past business partner data to generate optimal proposals; means for scanning handwritten memo data entered by the user and converting the converted handwritten memo data into text using OCR technology; means for importing the converted handwritten memo data into business partner documents; means for a user to review and correct the generated business partner documents; and means for outputting the reviewed documents in PDF format or presentation format. This enables efficient automatic generation of high-quality business partner documents.

[1135] An "SQL query" is a syntax for retrieving data from a database.

[1136] A "database" is a system for storing structured data.

[1137] "Cleaning" is the process of removing unnecessary information from data and preparing it in a form suitable for analysis.

[1138] An "HTML tag" is a part of the markup language used to describe the structure of a web page.

[1139] "TF-IDF vectorization" is a method for converting text data into a numerical vector.

[1140] A "numeric vector" is a format in which text data is expressed as numbers.

[1141] A "machine learning model" is an algorithm that learns patterns and relationships from data and makes predictions and classifications.

[1142] "Logistic Regression" is a statistical model for performing binary classification based on given data.

[1143] An "interface" is a screen or input form that allows a user to interact with a system.

[1144] "OCR technology" stands for optical character recognition technology, which is a technology that extracts character data from images.

[1145] "PDF format" is an abbreviation for Portable Document Format, a file format for displaying documents and images in a fixed layout.

[1146] This invention is a system that collects past documents, photographs, and handwritten memo data, and automatically generates business negotiation materials based on these. The system is mainly composed of three entities: a server, a terminal, and a user.

[1147] Data collection and preprocessing

[1148] First, the server uses an SQL query to retrieve past sales documents, photographs, and handwritten notes from the database. An example of a specific SQL query is "SELECT FROM past_materials." The collected data is then cleaned by removing unnecessary HTML tags and line breaks, and normalizing the text data. This prepares the data for analysis.

[1149] Data training

[1150] The cleaned data is transformed into a TF-IDF vector by the server and converted into a numerical vector. A machine learning model is then trained using algorithms such as logistic regression. This creates a model for predicting important elements of sales documents and proposal content.

[1151] Entering new business partner data

[1152] The user uses the terminal to input information about the potential business partner (company name, industry, needs, etc.) through the interface. The input data is sent from the terminal to the server and used for analysis.

[1153] Generate sales documents

[1154] The server analyzes the data of new business partners and automatically generates sales documents based on that data. During this process, TF-IDF vectorization is performed again, and the data is input into a trained model to predict the proposal content. The prediction results are formatted as sales documents based on a template. A specific output may be data embedded in a PowerPoint template.

[1155] Proposal generation

[1156] The server compares the data of the new business partner with past data and generates the most suitable proposal. This proposal is based on past success stories and learning data, making it possible to approach the new business partner effectively.

[1157] Analysis of handwritten notes

[1158] The user scans their handwritten notes onto their device, which then sends them to a server where text data is extracted from the notes using OCR technology (e.g., Google Cloud Vision API), which then cleans the data and incorporates it into the sales documents.

[1159] Check and print the final materials

[1160] The user checks the generated sales documents on their device and makes any necessary corrections. The server then outputs the final, checked sales documents in PDF or presentation format and provides them to the user. This series of processes ensures that high-quality sales documents are generated efficiently.

[1161] Specific examples

[1162] For example, if a new business partner is a technology company seeking a cloud solution, the user inputs that information via their device. The server then analyzes the input signal and references relevant past business documents, photos, and handwritten notes to automatically generate the optimal proposal and materials. Finally, the user can review the generated business documents and make any necessary corrections, after which the server outputs them in PDF format. This process ensures that business documents are provided quickly and efficiently.

[1163] Prompt Sentence Examples

[1164] "Generate the perfect proposal for new business opportunities with technology companies looking for cloud solutions."

[1165] In this invention, by using a generative AI model, negotiation materials are generated automatically and with high accuracy.

[1166] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1167] Step 1: Data collection and preprocessing

[1168] The server first uses an SQL query to retrieve past sales documents, photographs, and handwritten notes from the database. Specifically, it executes the query "SELECT FROM past_materials." The input is various materials in the database, and the output is the data collected by the server. The server then cleans the collected data, removing unnecessary HTML tags and line breaks, and normalizing the text data to make it suitable for analysis.

[1169] Step 2: Train the data

[1170] The server converts the cleaned data into a TF-IDF vector and then into a numerical vector. Next, it trains a machine learning model using algorithms such as logistic regression. The input is the cleaned text data, and the output is a machine learning model that predicts the important elements of sales documents and proposal content. For example, a model can be trained using data from past sales negotiations to build a model that predicts "sales negotiation elements with a high probability of success."

[1171] Step 3: Enter new customer data

[1172] The user inputs information about the new business partner (company name, industry, needs, etc.) through a terminal. This data is sent to the server in JSON format. The input is specific information about the new business partner, and the output is data sent to the server for analysis. For example, the user might input information such as "technology company" and "needs cloud solutions."

[1173] Step 4: Generate sales documents

[1174] The server analyzes data from new business partners, performs TF-IDF vectorization, and then inputs it into a trained model to predict proposal content. The input is information about new business partners, and the output is automatically generated business negotiation materials. As a specific example, a "presentation material proposing the scalability of cloud infrastructure" is generated based on a template.

[1175] Step 5: Generate proposals

[1176] The server compares the data of the new business partner with data from past negotiations, and generates the optimal proposal based on past success stories and related materials. The input is the information of the new business partner and past data, and the output is the proposal. For example, by referring to past successful cases in the same industry, the optimal proposal for the new business partner can be created.

[1177] Step 6: Analyze handwritten notes

[1178] The user scans their handwritten notes onto their device and sends the data to the server. The server then uses OCR technology to extract text data from the handwritten notes and cleans them. The input is the scanned handwritten notes, and the output is the converted handwritten notes. This allows the contents of the handwritten notes to be reflected in the sales documents.

[1179] Step 7: Check and print the final document

[1180] The user checks the generated sales documents on the terminal and makes corrections as necessary. The corrected documents are sent to the server, and the final confirmed sales documents are output in PDF or presentation format. The input is the sales documents corrected by the user, and the output is the final sales documents. For example, the user checks the presentation documents, edits them as necessary, and then saves the final version as a PDF.

[1181] (Application example 1)

[1182] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1183] The traditional process of creating sales documents requires a lot of time and effort, including collecting, cleaning, analyzing, and digitizing handwritten notes from past data and documents. It is also difficult to reflect this information in sales negotiations in real time. This results in lower efficiency in sales negotiations and makes it difficult to quickly provide appropriate proposals to customers.

[1184] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1185] In this invention, the server includes means for collecting past documents, photographic materials, and handwritten memo data, means for cleaning and analyzing the collected data, means for training a machine learning model based on the cleaned data, means for inputting information about new business partners from a user, means for analyzing the input information and automatically generating business negotiation materials, means for comparing the new business partners with past data and generating proposals, means for recognizing handwritten memo data as text and incorporating it into the business negotiation materials, means for checking and correcting the generated business negotiation materials and finally outputting the materials, means for providing the generated business negotiation materials to a user in real time, means for inputting new data through a user interface and immediately reflecting it in the business negotiation materials, and means for recognizing handwritten memo data using OCR technology and adding it to the business negotiation materials in real time, thereby enabling efficient and rapid generation of standardized business negotiation materials and real-time updates.

[1186] "Means of collection" refers to the methods and processes for obtaining past documents, photographic materials, handwritten memo data, etc. from the database and incorporating them into the system.

[1187] "Cleaning" is the process of removing unnecessary information from collected data and shaping it into a form suitable for analysis.

[1188] "Means for analysis" refers to a method for extracting and analyzing meaningful information from the cleaned data.

[1189] "Training" is the process of using cleaned and analyzed data to train a machine learning model.

[1190] "Means for input" refers to the interface and procedures for allowing users to input information about new business partners into the system.

[1191] The "means for automatic generation" is a process for automatically generating negotiation materials based on information input by the user.

[1192] The "means of generation" is a method of comparing information on new business partners with past data and deriving proposal content.

[1193] The "recognition method" is a method of converting handwritten note data into text data using OCR technology.

[1194] "Means for checking and correcting" refers to a procedure by which the user checks the generated business negotiation materials and revise them as necessary.

[1195] The "means of output" refers to the method of providing the final confirmed and revised business documents to the user in PDF or presentation format.

[1196] "Means for providing in real time" refers to a process for instantly presenting the generated business negotiation materials to the user.

[1197] A "means for reflecting" is a mechanism for immediately incorporating new data entered through the user interface into the sales materials.

[1198] "Means of recognizing using OCR technology" refers to a method of converting handwritten note data into digital text using optical character recognition technology.

[1199] "Means of adding in real time" refers to the process of immediately including the converted text data in the sales documents.

[1200] The present invention relates to a system for automatically generating business negotiation materials, and is implemented as follows: The system is mainly composed of three entities: a server, a terminal, and a user.

[1201] Data collection and preprocessing

[1202] The first thing the server does is collect past documents, photographs, and handwritten notes from the database. The server retrieves past business negotiation documents and cleans the collected data. This cleaning process involves removing unnecessary HTML tags and line breaks and normalizing the text data. The Python Pandas library can be used in this process.

[1203] Data training

[1204] The server trains a machine learning model based on the cleaned historical data. Specifically, it performs TF-IDF vectorization and then uses algorithms such as logistic regression to train the model. The Python scikit-learn library can be used. This model is used to predict key elements of sales documents and proposal content.

[1205] Entering new business partner data

[1206] The user inputs information about the potential business partner (e.g., company name, industry, and needs) through the interface from a terminal. The input data is sent to the server and analyzed.

[1207] Generate sales documents

[1208] The server analyzes the data of new business partners and automatically generates sales documents based on that data. The data of new business partners is converted back into TF-IDF vectors and input into the trained model to predict the proposal content. The prediction results are formatted as sales documents according to a template.

[1209] Proposal generation

[1210] The server compares the data of new business partners with past data and generates the most suitable proposals. These proposals are based on past success stories and learning data, allowing for an effective approach to new business partners.

[1211] Analysis of handwritten notes

[1212] The user scans their handwritten notes onto their device and sends them to the server. The server uses OCR technology to extract text from the handwritten notes and cleans them. A Python OCR library (such as Tesseract OCR) can be used. This allows the contents of the handwritten notes to be incorporated into the sales documents.

[1213] Real-time updates and provision of business documents

[1214] The generated sales documents are provided to the user in real time, and new data entered through the user interface is immediately reflected in the sales documents.

[1215] Check and print the final materials

[1216] The user can then review the generated sales documents on their device and make any necessary corrections. Finally, the server outputs the reviewed sales documents in PDF or presentation format and provides them to the user. This series of processes ensures that high-quality sales documents are generated efficiently.

[1217] Prompt Sentence Examples

[1218] Collect, clean, and format your past sales pitch data.

[1219] Next, enter the new contact information and generate the appropriate documents.

[1220] If you have handwritten note images, use OCR technology to convert them into text and include them in the generated document.

[1221] Finally, output the document in PDF format.

[1222] Specific examples

[1223] For example, if the client is a medical device manufacturer, the user inputs the information into the device. The server then analyzes the input data and references relevant past sales documents, photographs, and handwritten notes to automatically generate the optimal proposal and materials. The generated sales documents are provided in real time, and the user can modify them as needed and finally output them in PDF format. This process allows for the rapid and efficient provision of standardized sales documents.

[1224] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1225] Step 1:

[1226] Data collection

[1227] The server retrieves past business negotiation materials, photographic materials, and handwritten memo data from the database. Database connection information and a data collection query are used as input. Specifically, an SQL query is executed to retrieve data. For example, the command "SELECT FROM past_materials" is executed to retrieve past business negotiation materials. The output is the raw data of the retrieved materials and memo data.

[1228] Step 2:

[1229] Cleaning the data

[1230] The server removes unnecessary HTML tags and line breaks from the acquired data and normalizes the text data. It uses the collected raw data as input. Specifically, it uses Pandas and a regular expression library to format the data. For example, it performs processing such as "df['text'] = df['text'].str.replace(r'<[^>]>', '')". The output is cleaned text data.

[1231] Step 3:

[1232] Data training

[1233] The server performs TF-IDF vectorization on the cleaned past data and trains a Logistic Regression model. The cleaned text data is used as input. Specifically, the scikit-learn library is used to perform vectorization and model training. Each process is performed using "tfidf = TfidfVectorizer()" and "model = LogisticRegression()". The output is a trained model.

[1234] Step 4:

[1235] Entering new business partner data

[1236] The user enters information about the new business contact (company name, industry, and needs) through the terminal interface. The input is data manually entered by the user. Specifically, form input and text fields are used. The input new business contact information data is obtained as the output.

[1237] Step 5:

[1238] Analysis of new business contact data

[1239] The server analyzes the entered new client data and converts it back into a TF-IDF vector. The input is the new client information entered by the user. Specifically, the data is converted using the TF-IDF vectorizer defined earlier. Set "X_new = tfidf.transform([new_client_data])". The output is the vectorized new client data.

[1240] Step 6:

[1241] Automatic generation of sales documents

[1242] The server inputs the vectorized new business negotiation data into the trained model and predicts the proposal content. The vectorized new business negotiation data and the trained model are used as input. Specifically, the processing is performed as follows: "prediction = model.predict(X_new)". The output is the sales negotiation document data containing the predicted proposal content.

[1243] Step 7:

[1244] Proposal generation and formatting

[1245] The server formats the sales negotiation materials according to a template based on the predicted proposal content. The predicted proposal content data is used as input. Specifically, the proposal content is applied to the sales negotiation material template to generate a document. The formatted sales negotiation material data is obtained as output.

[1246] Step 8:

[1247] Analysis of handwritten notes

[1248] The user scans handwritten notes onto the device. The server receives the data and extracts text from the handwritten notes using OCR technology. The input is the image data of the handwritten notes. Specifically, the text is extracted using an OCR library (such as Tesseract OCR). The output is the handwritten note data converted into text.

[1249] Step 9:

[1250] Integrate handwritten notes into sales documents

[1251] The server cleans the converted handwritten notes and integrates them into the business negotiation materials. The input is the converted handwritten notes. Specifically, the cleaned data is added to the business negotiation materials, e.g., "formatted_material += ocr_text". The output is the integrated business negotiation materials.

[1252] Step 10:

[1253] Real-time provision of business negotiation materials

[1254] The server provides the generated sales negotiation materials to the user in real time. Every time new data is entered through the user interface, it is reflected in the sales negotiation materials. New user data is used as input. Specifically, data synchronization is performed using WebSocket or real-time communication technology. The latest sales negotiation material data is provided as output in real time.

[1255] Step 11:

[1256] Final confirmation and printing of business documents

[1257] The user checks the generated sales document on their device and makes any necessary corrections. The server then outputs the final, checked document in PDF or presentation format and provides it to the user. The final, checked sales document data is used as input. Specifically, the document is output using a PDF generation library (such as ReportLab). The final sales document in PDF format is obtained as output.

[1258] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1259] This invention relates to a system for automatically generating business negotiation materials, and furthermore, a system for adjusting proposal content by combining an emotion engine that recognizes the user's emotions. The system is mainly composed of four entities: a server, a terminal, a user, and an emotion engine.

[1260] Data collection and preprocessing

[1261] The first thing the server does is collect past business materials, photo materials, and handwritten memo data from the database. For example, the server uses an SQL query to execute a command such as "SELECT FROM past_materials" to retrieve this data. The retrieved data is saved as business materials data, image data, and handwritten memo data, respectively.

[1262] The server then cleans the collected data. For example, unnecessary HTML tags and line breaks are removed from past sales documents and text data is normalized. Image recognition algorithms are used to trim unnecessary parts of photo materials, and noise is removed and text is emphasized from handwritten notes.

[1263] Data training

[1264] The server trains a machine learning model based on the cleaned historical data. Specifically, it performs TF-IDF vectorization to convert text data into numerical vectors, and then trains the model using algorithms such as logistic regression. This model is used to predict key elements of sales documents and proposal content.

[1265] Entering new business partner data

[1266] The user inputs information about the new business partner (e.g., company name, industry, needs) from the terminal through a dedicated interface. The input information is sent to the server as new business partner data.

[1267] Generate sales documents

[1268] The server receives and analyzes the new business negotiation data sent by the user. Specifically, it converts the new data into text format and performs TF-IDF vectorization. After that, it uses a machine learning model to automatically generate optimal business negotiation materials based on the analysis results.

[1269] Proposal generation

[1270] The server compares the data of the new business partner with past data and generates the most suitable proposal. For example, based on past success stories and proposals, the server incorporates content to propose the optimal solutions and services to the new business partner into the business negotiation materials.

[1271] Analysis of handwritten notes

[1272] The user scans their handwritten notes into their device and sends the data to the server, which then uses OCR technology to extract text from the notes, cleans it, and formats it into a format suitable for analysis.

[1273] Use of emotion engine

[1274] The emotion engine analyzes user interactions and recognizes user emotions in real time. For example, facial expressions and voices are collected by a camera or microphone when the user uses an input interface, and the emotion engine analyzes them to extract emotional data.

[1275] The emotion data recognized by the emotion engine is used by the server to adjust the content of suggestions. For example, if the user is tired, the server can maximize the effectiveness of suggestions by summarizing the content of the materials more succinctly or using expressions that encourage relaxation.

[1276] Check and print the final materials

[1277] The user checks the generated business negotiation materials on the terminal and makes corrections as necessary. Checks the contents of the business negotiation materials and makes corrections if there is missing information or if there are any parts that need to be corrected.

[1278] Finally, the server outputs the confirmed business documents in PDF or presentation format, allowing the user to use the documents in actual business negotiations.

[1279] Specific examples

[1280] For example, if a new business partner is a technology company seeking cloud solutions, the user inputs that information into their device. The server then analyzes the input signal and references relevant past sales documents, photos, and handwritten notes to automatically generate the optimal proposal content and materials. An emotion engine also analyzes the user's interactions. If the user is excited, it can incorporate responses into the proposal materials that further emphasize the proposal and increase the dynamism of the presentation. Finally, the user reviews the generated sales documents, makes any necessary corrections, and the server outputs them in PDF format. This process ensures that standardized sales documents are provided quickly and efficiently.

[1281] The processing flow will be explained below.

[1282] Step 1:

[1283] The server collects past business materials, photo materials, and handwritten memo data from the database. Specifically, it executes a command such as "SELECT FROM past_materials" using an SQL query, and saves this data as business materials data, image data, and handwritten memo data, respectively.

[1284] Step 2:

[1285] The server cleans the collected data. For example, it removes unnecessary HTML tags and line breaks from past sales documents and normalizes the text data. It also uses image recognition algorithms to trim unnecessary parts of photographs, and removes noise and emphasizes characters from handwritten notes.

[1286] Step 3:

[1287] The server trains a machine learning model based on the cleaned data. Specifically, it performs TF-IDF vectorization to convert the text data into numerical vectors, and then trains the model using algorithms such as logistic regression. This model is used to predict important elements of sales documents and proposal content.

[1288] Step 4:

[1289] The user uses a terminal to input information about the new business partner (e.g., company name, industry, needs) through a dedicated interface. The input information is sent to the server as new business partner data.

[1290] Step 5:

[1291] The server receives and analyzes the new business negotiation data sent by the user. Specifically, it converts the new data into text format and performs TF-IDF vectorization. After that, it uses a machine learning model to automatically generate optimal business negotiation materials based on the analysis results.

[1292] Step 6:

[1293] The server compares the data of the new business partner with past data and generates the most suitable proposal. For example, based on past success stories and proposals, the server incorporates content to propose the optimal solutions and services to the new business partner into the business negotiation materials.

[1294] Step 7:

[1295] The user scans their handwritten notes into their device and sends the data to the server, which then uses OCR technology to extract text from the notes, cleans it, and formats it into a format suitable for analysis.

[1296] Step 8:

[1297] When a user inputs emotion data through the interface, the emotion engine analyzes the user's facial expressions and voice. The emotion engine analyzes the interaction and extracts the user's emotion data. For example, it can recognize the user's emotion in real time using a camera or microphone.

[1298] Step 9:

[1299] The server receives the user's emotional data recognized by the emotion engine and adjusts the content of the proposal. Specifically, if the user is excited, the proposal content is emphasized, while if the user is tired, the materials are summarized more succinctly. This allows the server to provide optimal business negotiation materials according to the user's psychological state.

[1300] Step 10:

[1301] The user checks the generated sales negotiation materials on the terminal and makes corrections as necessary. Checks the contents of the sales negotiation materials and makes corrections if there is missing information or any parts that need to be corrected.

[1302] Step 11:

[1303] The server then outputs the final confirmed business documents in PDF or presentation format and provides them to the user, who can then use the documents in actual business negotiations.

[1304] Example 2

[1305] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1306] The current process of creating sales documents requires manual input and data organization, which consumes a lot of time and effort. In addition, the proposals are often suboptimal because they do not take into account the user's emotional state.

[1307] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1308] In this invention, the server includes means for collecting past documents, image data, and handwritten information data, means for cleaning and analyzing the collected data, and means for training a machine learning model based on the cleaned data. This enables efficient data collection and preprocessing. The server also includes means for a user to input information about a new business partner, means for analyzing the input information and automatically generating business negotiation materials, means for comparing the data about the new business partner with past data and generating proposal content, means for recognizing handwritten information data as character data and incorporating it into the business negotiation materials, means for checking and correcting the generated business negotiation materials and finally outputting the materials, and means for analyzing the user's emotions and adjusting the proposal content based on the emotions. This enables automatic generation of business negotiation materials and adjustment of the proposal content based on the user's emotions.

[1309] "Past documents" are previously created records including business documents, image data, and handwritten information data.

[1310] "Image data" is digital data that contains visual information such as photographic material.

[1311] "Handwritten information data" is data in digital form obtained from handwritten notes or documents.

[1312] "Cleaning" refers to removing unnecessary elements from data and preparing it in a format suitable for analysis.

[1313] A "machine learning model" is an algorithm that learns patterns and relationships in data and makes predictions and classifications based on new data.

[1314] "Information on new business partners" refers to attribute data on new customers or companies that are the subject of business negotiations, such as company name, industry, and needs.

[1315] "Business documents" refers to presentation materials and proposals used to conduct business negotiations.

[1316] "Comparison" refers to comparing new data with previous data to identify similarities and differences.

[1317] The "proposal content" is information that includes an outline of the solution or service to be provided to the business partner.

[1318] "Character data" is digital data that represents information in text format.

[1319] "Optical character recognition technology (OCR)" is a technology that converts handwritten or printed character data into digital text.

[1320] "Emotion analysis" refers to the process of identifying a user's emotional state from their facial expressions and voice.

[1321] "Adjustment" refers to modifying content or behavior to suit specific conditions or circumstances.

[1322] The present invention relates to a system that combines the automatic generation of business negotiation materials with the recognition of user emotions. This system is composed of four main components: a server, a terminal, a user, and an emotion engine.

[1323] Data collection and preprocessing

[1324] The server first collects past sales documents, image data, and handwritten information data from the database. It retrieves the data using an SQL query (e.g., "SELECT FROM past_materials") and saves it in a folder corresponding to each data type. After this process, the server cleans the collected data. Specifically, it uses regular expressions to remove unnecessary HTML tags and line breaks from the text data and normalizes the text. For image data, Python's Pillow library is used to trim unnecessary parts, and for handwritten information, an OCR tool (e.g., Tesseract) is used to extract text and remove noise.

[1325] Data training

[1326] The server trains a machine learning model based on the cleaned data. It converts the text data into TF-IDF vectors, then uses the scikit-learn library to build a logistic regression model and learns the key elements for generating sales documents.

[1327] Entering new business partner data

[1328] The user inputs information about the new business contact (company name, industry, needs, etc.) through a dedicated interface on the terminal. The terminal receives the user's input using GUI components (e.g., text boxes, drop-down lists) and sends it to the server.

[1329] Generate sales documents

[1330] The server receives and analyzes the data of new business partners sent by the user. Specifically, it converts the new data into text format and converts it into TF-IDF vectors. It then uses the trained model to automatically generate optimal business documents.

[1331] Proposal generation

[1332] The server compares the data of the new business partner with past data and generates the optimal proposal for the new business partner based on past success stories and proposal content. This process may use natural language generation tools (e.g., GPT-3).

[1333] Analysis of handwritten notes

[1334] Users scan their handwritten notes onto the device, which uses OCR software to extract the text from the handwritten notes, and then send this data to a server that cleans and formats it for analysis.

[1335] Use of emotion engine

[1336] The emotion engine analyzes the user's interactions. The user's facial expressions and voice are collected by the device's camera and microphone, and the emotion engine analyzes them to extract emotional data. This data is sent to the server and used to adjust the content of the suggestions. For example, if the user is tired, the content of the materials can be summarized more concisely.

[1337] Check and print the final materials

[1338] The user checks the generated sales documents on the terminal. The user checks the documents in the GUI and makes any necessary corrections. Finally, the server outputs the checked sales documents in PDF or presentation format. The user downloads them and uses them in actual sales negotiations.

[1339] Specific examples

[1340] For example, if a new business partner is a technology company seeking a cloud solution, the user enters that information on their device. The server then analyzes the input data and automatically generates the optimal proposal by referencing relevant past business documents, image data, and handwritten notes. In addition, if the user expresses excitement, the emotion engine analyzes this and adds dynamism to further emphasize the proposal. The user can then review the final business document and make any necessary corrections, after which the server outputs it in PDF format.

[1341] Prompt Sentence Examples

[1342] "Please enter the company name, industry, and needs of your new business partner into the terminal."

[1343] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1344] Step 1: Data collection and preprocessing

[1345] The server collects past sales documents, image data, and handwritten information data from a database. The input is an SQL query (e.g., "SELECT FROM past_materials"), and the output is data saved in a folder corresponding to each data type. The server then cleans the collected data. Specifically, it uses regular expressions to remove unnecessary HTML tags and line breaks from the text data and normalizes the text. It also uses the Pillow library to trim unnecessary parts of the image data, and an OCR tool (e.g., Tesseract) to extract text and remove noise from the handwritten information data. This results in data in a format suitable for analysis.

[1346] Step 2: Train the data

[1347] The server trains a machine learning model using the cleaned data. The input is the cleaned text data, and the output is the trained model. First, the text data is converted into TF-IDF vectors, and then a logistic regression model is built using the scikit-learn library. This model learns the elements necessary for generating sales documents.

[1348] Step 3: Enter new customer data

[1349] The user inputs information about the new business contact (company name, industry, needs, etc.) through a dedicated interface on the terminal. The input is a GUI component (e.g., text box, drop-down list), and the output is the information about the new business contact that is sent to the server. The terminal receives this data and sends it to the server.

[1350] Step 4: Generate sales documents

[1351] The server receives and analyzes the data of new business partners sent by the user. The input is the information of new business partners, and the output is the generated business documents. Specifically, the new data is converted into text format and TF-IDF vectorized. The trained model is then used to automatically generate the optimal business documents.

[1352] Step 5: Generate proposals

[1353] The server compares the data of the new business partner with past data and generates an appropriate proposal. The input is the data of the new business partner, past success stories, and proposals, and the output is the proposal to be incorporated into the business materials. This process may use natural language generation tools (e.g., GPT-3).

[1354] Step 6: Analyze handwritten notes

[1355] A user scans their handwritten notes into a device, which then uses OCR software to extract text from them. The input is the scanned image of the handwritten notes, and the output is text data sent to a server, which cleans and formats the text data for analysis.

[1356] Step 7: Use the Emotion Engine

[1357] The emotion engine analyzes the user's interactions and recognizes their emotions. The input is the user's facial expressions and voice, and the output is emotional data. Data collected by the device's camera and microphone is analyzed by the emotion engine, and the extracted emotional data is sent to the server. The server adjusts the content of the suggestions based on this data. For example, if the system recognizes that the user is tired, it may make adjustments such as summarizing the contents of the materials more concisely.

[1358] Step 8: Check and print the final document

[1359] The user checks the sales negotiation materials generated on the terminal and makes corrections as necessary. The input is the generated sales negotiation materials, and the output is the final confirmed sales negotiation materials. The materials are displayed on the GUI, and missing information and corrections are made. Finally, the server outputs the confirmed sales negotiation materials in PDF or presentation format. The user downloads them and uses them in actual sales negotiations.

[1360] (Application example 2)

[1361] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1362] In an automatic generation system for sales documents, there is a demand for the ability to quickly and efficiently generate optimal proposals for new business partners based on past data, but conventional systems have had difficulty responding in real time or adjusting based on user emotions.In addition to sales documents, it is also important to make real-time situational assessments and take appropriate action in the manufacturing process, but there has been a lack of systems that can do this.

[1363] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting past documents, photographic materials, and handwritten memo data, means for cleaning and analyzing the collected data, means for training a machine learning model based on the cleaned data, means for inputting information about new business partners from a user, means for analyzing the input information and automatically generating business negotiation materials, means for comparing the new business partner with past data and generating proposal content, means for recognizing handwritten memo data as text and incorporating it into the business negotiation materials, means for assessing the situation in real time and taking appropriate action, means for recognizing the user's emotions and adjusting the proposal content, and means for confirming and correcting the generated business negotiation materials and finally outputting the materials. This makes it possible to realize a system that efficiently generates business negotiation materials and enables adjustments and real-time responses based on the user's emotions.

[1364] "Past materials" is a general term for information collected in the past, such as business negotiation materials, manufacturing records, photographic materials, and handwritten memo data.

[1365] "Means of collection" refers to the methods and devices used to obtain historical documents, photographic materials, and handwritten notes from databases and other sources.

[1366] "Cleaning and analyzing means" refers to a method or device that performs processing such as removing unnecessary parts from collected data, normalizing text data, or removing noise from image data.

[1367] "Means for training a machine learning model" refers to a method or apparatus for using cleaned data to train a model using a specific algorithm (e.g., logistic regression or TF-IDF vectorization).

[1368] "Means for inputting information about new business partners" refers to an interface or device that allows a user to input details about a new business partner (company name, industry, needs, etc.).

[1369] "Means for analyzing input information and automatically generating business negotiation materials" refers to a process or device that analyzes information input by a user and automatically creates optimal business negotiation materials based on that information.

[1370] "Means for generating proposals" refers to the process or device that compares data on new business partners with past data to create the most appropriate proposals.

[1371] The "OCR technology" in "recognizing handwritten memo data as text" refers to optical character recognition technology that converts handwritten character information into digital text data.

[1372] "Means for real-time situation assessment and appropriate response" refers to methods and devices that use collected data and machine learning models to instantly assess the current situation and take appropriate action.

[1373] "Means for recognizing user emotions and adjusting suggestions" refers to a process or device that analyzes the user's emotional state (e.g., information obtained from facial expressions and voice) and adjusts suggestions and interfaces based on that.

[1374] "Means for checking and correcting the generated sales documents and finally outputting the documents" refers to a process or device in which a user checks the automatically generated sales documents, corrects them as necessary, and finally outputs them in PDF or presentation format.

[1375] This invention relates to a system for automatically generating business negotiation materials and a system for optimizing manufacturing processes. By combining this system with an emotion engine that recognizes the user's emotions, it is possible to adjust proposal content and respond appropriately in real time.

[1376] Data collection and preprocessing

[1377] First, the server collects past business materials, photographs, and handwritten notes from the database. For example, it uses an SQL query to execute a command such as "SELECT FROM past_materials" to retrieve this data. This data includes business materials, photographs, and handwritten notes.

[1378] The collected data is cleaned by the server. Unnecessary HTML tags and line breaks are removed from past sales documents, and text data is normalized. Furthermore, image recognition algorithms are used to trim unnecessary parts of photographs, and noise is removed and text is emphasized from handwritten notes.

[1379] Data training

[1380] The server trains a machine learning model based on the cleaned data. Specifically, it performs TF-IDF vectorization to convert the text data into numerical vectors, and then trains the model using algorithms such as logistic regression. This model is used to predict key elements of sales documents and proposal content.

[1381] Entering new business partner data

[1382] The user inputs information about the new business partner (e.g., company name, industry, needs) from the terminal through a dedicated interface. The input information is sent to the server as new business partner data.

[1383] Generate sales documents

[1384] The server receives and analyzes the new business negotiation data sent by the user. Specifically, it converts the new data into text format and performs TF-IDF vectorization. After that, it uses a machine learning model to automatically generate optimal business negotiation materials based on the analysis results.

[1385] Proposal generation

[1386] The server compares the data of the new business partner with past data and generates the most suitable proposal. For example, based on past success stories and proposals, the server incorporates content to propose the optimal solutions and services to the new business partner into the business negotiation materials.

[1387] Analysis of handwritten notes

[1388] The user scans their handwritten notes into their device and sends the data to the server, which then uses OCR technology to extract text from the notes, cleans it, and formats it into a format suitable for analysis.

[1389] Use of emotion engine

[1390] The server analyzes the user's interactions using an emotion engine to recognize the user's emotions in real time. For example, the camera and microphone collect facial expressions and voices as the user uses the input interface, which the emotion engine then analyzes to extract emotional data. The emotional data recognized by the emotion engine is used by the server to adjust the content of suggestions. For example, if the user is tired, the server can maximize the effectiveness of suggestions by summarizing the content of the materials more succinctly or using expressions that encourage relaxation.

[1391] Check and print the final materials

[1392] The user checks the generated business documents on their device and makes corrections as necessary. They check the contents of the documents and make any necessary corrections or missing information. Finally, the server outputs the checked business documents in PDF or presentation format. This allows the user to use the documents in actual business negotiations.

[1393] Application to manufacturing processes

[1394] In the manufacturing process, the server collects and analyzes past manufacturing data and trains a machine learning model. This model is used to monitor the manufacturing process in real time and detect abnormalities early. If an abnormality is detected, the server proposes appropriate countermeasures and issues instructions to robots and human operators. The server also recognizes the emotions of workers and adjusts interface operations and information displays to improve work efficiency and safety.

[1395] Specific examples

[1396] For example, if a new business partner is a technology company seeking cloud solutions, the user inputs that information into their device. The server then analyzes the input signal and references relevant past sales documents, photos, and handwritten notes to automatically generate the optimal proposal content and materials. An emotion engine also analyzes the user's interactions. If the user is excited, it can incorporate responses into the proposal materials that further emphasize the proposal and increase the dynamism of the presentation. Finally, the user reviews the generated sales documents, makes any necessary corrections, and the server outputs them in PDF format. This process ensures that standardized sales documents are provided quickly and efficiently.

[1397] Prompt Sentence Examples

[1398] An example prompt for applying this system to a manufacturing process is:

[1399] "Build a system that detects anomalies in real time based on past manufacturing data and suggests optimal countermeasures."

[1400] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1401] Step 1:

[1402] Data collection

[1403] The server first collects past documents, photographs, and handwritten notes from the database. For example, it uses an SQL query to execute a command such as "SELECT FROM past_materials" to retrieve this data. This data includes business document data, image data, and handwritten notes. The input data is various past documents, and the output data is the set of collected documents.

[1404] Step 2:

[1405] Cleaning the data

[1406] The server cleans the collected data. For example, it removes unnecessary HTML tags and line breaks from past sales documents and normalizes the text data. It also uses image recognition algorithms to trim unnecessary parts of image data, and removes noise and emphasizes characters from handwritten memo data. The input data is the collected data, and the output data is the cleaned data.

[1407] Step 3:

[1408] Training a machine learning model

[1409] The server uses the cleaned data to train a machine learning model. Specifically, it performs TF-IDF vectorization to convert the text data into a numerical vector, and then trains the model using an algorithm such as logistic regression. The input data is the cleaned data, and the output data is the trained model.

[1410] Step 4:

[1411] Entering new business partner data

[1412] The user inputs information about the new business partner (company name, industry, needs) through a dedicated interface on the terminal. The input information is sent to the server as new business partner data. The input data is the new business partner information, and the output data is the data sent from the terminal to the server.

[1413] Step 5:

[1414] Automatic generation of sales documents

[1415] The server receives and analyzes the new business partner data sent by the user. Specifically, it converts the new data into text format and performs TF-IDF vectorization. It then uses a machine learning model to automatically generate optimal business partner documents based on the analysis results. The input data is the new business partner data, and the output data is the generated business partner documents.

[1416] Step 6:

[1417] Proposal generation

[1418] The server compares the new business partner data with past data and generates the most appropriate proposal. For example, based on past success stories and proposal content, the server incorporates content that proposes the optimal solutions and services for the new business partner into the business negotiation materials. The input data is the new business partner data and past data, and the output data is the generated proposal content.

[1419] Step 7:

[1420] Analysis of handwritten notes

[1421] Users scan their handwritten notes onto their devices and send the data to the server. The server uses OCR technology to extract text from the notes, cleans the text, and formats it into a format suitable for analysis. The input data is the handwritten note image, and the output data is the cleaned text data.

[1422] Step 8:

[1423] Use of emotion engine

[1424] The server uses an emotion engine to analyze user interactions and recognize the user's emotions in real time. For example, a camera or microphone can capture facial expressions and voices as the user uses the input interface, which the emotion engine then analyzes to extract emotion data. The input data is the voice and image data from the camera or microphone, and the output data is the extracted emotion data.

[1425] Step 9:

[1426] Check and print the final materials

[1427] The user checks the generated business meeting materials on their terminal and makes corrections as necessary. They check the contents of the business meeting materials and make any necessary corrections or missing information. Finally, the server outputs the checked business meeting materials in PDF or presentation format. The input data is the generated business meeting materials, and the output data is the final output business meeting materials.

[1428] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1429] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1430] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1431] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1432] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1433] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1434] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1435] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1436] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1437] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1438] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1439] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1440] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1441] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1442] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1443] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1444] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different ty...

Claims

1. A means of collecting historical documents, photographic materials, and handwritten notes; a means of cleaning and analyzing the collected data; A means of training a machine learning model based on the cleaned data; and A means for a user to input information about a new business partner; A means for analyzing input information and automatically generating business negotiation materials; A means to compare new business partners with past data and generate proposals; A means to recognize handwritten memo data as text and incorporate it into business negotiation materials; A means for checking and correcting the generated business negotiation materials and finally outputting the materials; A system including:

2. The system according to claim 1 , wherein a model for automatically generating proposals from past data is trained.

3. 2. The system according to claim 1, wherein handwritten memo data is recognized as text using OCR technology and is incorporated into business negotiation materials.

Images