system

Abstract

We provide the system. [Solution] A method for extracting information from past negotiation databases, A means of preprocessing the extracted information and using it as input for an analysis model, A means for learning from preprocessed data and providing a trained model for generating proposal materials, A means of automatically generating proposal materials tailored to specific customers, A means of providing an interface that allows the generated proposal document to be reviewed and edited, A means of collecting user feedback and using it to improve trained models, A system that includes this.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a persona chatbot control method performed by at least one processor, including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a 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

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the sales department of an enterprise, due to the fact that it takes a great deal of time and effort to create proposal documents and conduct data analysis, there is a problem that sales staff cannot allocate sufficient time to customer response and strategic activities. As a result, the efficiency of the entire sales activity decreases, and it becomes difficult to timely make competitive proposals.

Means for Solving the Problems

[0005] This invention provides a system that extracts information from a past negotiation database, preprocesses the extracted information to serve as input for an analysis model, and uses the preprocessed data to provide a trained model for generating proposal materials. Furthermore, it automatically generates proposal materials tailored to specific customers and provides an interface for users to review and edit these materials, thereby directly linking the generated proposal materials to actual sales activities. In addition, by collecting user feedback and using it to improve the trained model, a feedback loop is formed to improve the quality of proposal materials, enabling the rapid provision of optimal sales materials.

[0006] A "negotiation database" is a database used by companies to store and manage information about past business negotiations and deals.

[0007] "Preprocessing" refers to the data preparation process that transforms acquired data into a format usable by machine learning models, and includes tasks such as handling missing values ​​and normalizing the data.

[0008] An "analysis model" is a system that executes machine learning algorithms to create proposal documents based on the generated data.

[0009] A "proposal document" is a document that summarizes the content of a proposal made to a specific customer, and includes an overview of the product or service, its benefits, and examples of its use.

[0010] A "trained model" is a machine learning algorithm that has been refined based on training data and has the ability to make predictions and classifications based on new data.

[0011] An "interface" refers to the environment or means by which a user interacts with a computer or system, and includes screens and operating procedures for users to review and edit proposal documents.

[0012] "Feedback" refers to the opinions and evaluations that users provide to a system, and this information is used to help improve the system or model. [Brief explanation of the drawing]

[0013] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11] This is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] This is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] This is a sequence diagram showing the processing flow of the data processing system in Example 2, which incorporates an emotion engine. [Figure 14]It is a sequence diagram showing the processing flow of a data processing system in Application Example 2 when a sentiment engine is combined.

Embodiment for Implementing the Invention

[0014] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0015] First, the terms used in the following description will be explained.

[0016] In the following embodiments, the numbered processor (hereinafter simply referred to as "processor") may be one arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be one type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0017] In the following embodiments, the numbered RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0018] In the following embodiments, the numbered storage is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, etc.

[0019] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0020] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0021] [First Embodiment]

[0022] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0023] As shown in Figure 1, the 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.

[0024] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0025] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0026] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and 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.

[0027] 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 perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0028] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0029] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0030] As shown in Figure 2, in the data processing device 12, a specific processing 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" related to the technology of this 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 according to the specific processing program 56 executed on the RAM 30.

[0031] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0032] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0033] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0034] This invention is a system designed to improve the efficiency of proposal document creation in the sales department, and is implemented by the following means. First, the server extracts past sales negotiation data from the company's negotiation database. At this time, the server uses database queries to obtain necessary sales negotiation information (customer name, industry, negotiation details, etc.). This data forms the basis of a trained model for generating proposal documents.

[0035] Next, the server preprocesses the acquired sales opportunity data. In this process, the server performs data cleaning, such as imputing missing values, removing outliers, and normalizing the data. It also prepares the data into a format that the model can learn from, such as converting categorical data into numerical data.

[0036] The server uses pre-processed data to train a generative AI model. This trained model is designed to present optimal proposals for specific customers, and by referencing past success stories, it can generate more effective sales materials.

[0037] The generated proposal document is sent to the terminal, where the user can review it on the interface and edit it as needed. For example, the user can fine-tune the wording of the proposal document to suit a specific customer or to emphasize the product's features.

[0038] Finally, after the user completes the proposal document, the system collects feedback from the user. The user can point out which parts of the document were effective, and the server can incorporate this feedback data into the next learning cycle to improve the accuracy and quality of the generated AI model. This feedback loop continuously improves the efficiency and effectiveness of proposal document creation as a whole system.

[0039] The following describes the processing flow.

[0040] Step 1:

[0041] The server connects to the negotiation database and extracts past deal data. It issues SQL queries to collect data based on specific criteria (for example, deal history for the past 6 months).

[0042] Step 2:

[0043] The server preprocesses the extracted data. It sorts the data types and imputes any missing values ​​using appropriate methods. Furthermore, it encodes categorical variables into numerical values ​​and converts the data into a format that is easy for the model to handle.

[0044] Step 3:

[0045] The server trains the generative AI model using preprocessed data. Training is performed using an iterative learning algorithm, which learns features based on past successes.

[0046] Step 4:

[0047] The device automatically generates new proposal materials using a pre-trained generative AI model in response to user requests. These generated materials include personalized content for each customer.

[0048] Step 5:

[0049] Users review the proposal documents on their devices. If necessary, they edit the wording of the documents to tailor them to the specific needs of the client.

[0050] Step 6:

[0051] After the user gives final approval to the proposal, they enter feedback. The system collects this feedback and stores it on the server as data that can be used for future model improvements.

[0052] (Example 1)

[0053] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0054] In sales operations, efficiently and quickly creating proposal materials tailored to each customer is challenging, and improving the accuracy of these materials, especially by leveraging past success stories, requires significant time and effort. Furthermore, there is a need for a means to continuously improve the effectiveness of automatically generated proposal materials in actual business negotiations.

[0055] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0056] In this invention, the server includes means for extracting data from past negotiation information, means for preprocessing the extracted data and shaping it as input for an analysis model, and means for providing a trained model for proposal creation based on the preprocessed data. This enables a system that efficiently creates optimal proposals for each customer and includes a feedback process to improve their effectiveness.

[0057] "Past negotiation information" refers to documents and records related to business negotiations and transactions that have taken place in the past during a company's activities.

[0058] "Means of data extraction" refers to methods or devices for selecting and retrieving necessary information from databases or records.

[0059] "Preprocessing" refers to the process of preparing data to be analyzable, and specifically includes imputing missing values, removing outliers, and normalization.

[0060] "Model input for analysis" refers to data formatted in a way that machine learning models can use for learning and inference.

[0061] A "trained model" is a machine learning model that has been trained on existing data and is used to perform a specific task.

[0062] A "proposal" is a document outlining a plan or information submitted to a customer or business partner, and is used for business negotiations and contract signing.

[0063] "Feedback" refers to information provided to evaluate a system or process and to facilitate improvement.

[0064] "Methods for removing outliers" refer to methods or processes for identifying and eliminating inappropriate data points within a dataset.

[0065] "Means of converting categorical data into numerical format" refers to methods of replacing categorical data with numerical formats, enabling computational processing in machine learning.

[0066] This invention consists of a system that streamlines the creation of proposal materials in sales operations. Specifically, various processes are carried out primarily using a server, terminal, and user.

[0067] The server extracts necessary data from the company's negotiation information database. Using database queries, the server extracts important information related to past business negotiations, such as customer names, industries, and negotiation details. This data then forms the basis for proposal generation.

[0068] Next, the server preprocesses the extracted data. This preprocessing includes imputing missing values, removing outliers, and normalizing the data. It also converts categorical data into numerical format, making it suitable for machine learning models to learn from. This preprocessing improves the accuracy of the data.

[0069] The server uses the formatted data to train a generative AI model. The trained model is then used to provide customers with the most suitable proposals. This includes the ability to leverage past success stories to generate sales materials more effectively.

[0070] The generated proposal document is sent to the terminal and can be reviewed by the user through the interface. The user can review the document and edit it to suit a specific customer. Specifically, it is possible to adjust it to the customer's specialized terminology and emphasize the features of specific products.

[0071] For example, when a user creates a sales proposal for a new product, the server generates the initial proposal by retrieving data from past success stories and applying it to a learning model. An example of a prompt message as text might be: "Generate a proposal for a new product. The target market is small and medium-sized enterprises, and cost efficiency should be emphasized." This allows the system to provide the user with more specific and compelling materials, helping subsequent sales negotiations proceed smoothly.

[0072] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0073] Step 1:

[0074] The server queries a company's negotiation information database to extract data. The input database contains past business negotiation information. Specifically, it retrieves information such as customer name, industry, and negotiation details from the database. The output is negotiation data that forms the basis for creating proposal documents.

[0075] Step 2:

[0076] The server preprocesses the extracted sales opportunity data. The raw data obtained in the previous step is used as input. Specifically, it performs operations such as imputing missing values, removing outliers, and normalizing the data, converting categorical data into numerical format. The output is data converted into a format that can be processed by machine learning models.

[0077] Step 3:

[0078] The server trains a generative AI model using pre-processed data. The input is formatted data. Specifically, the model is trained based on past successful sales negotiations. The output is a trained model capable of generating optimal proposals for a particular customer.

[0079] Step 4:

[0080] The server generates proposal documents using a pre-trained model. The input consists of data and requirements related to a specific customer. Specifically, the generating AI model creates proposals tailored to the customer. The output is the automatically generated proposal document.

[0081] Step 5:

[0082] The terminal displays the generated proposal document to the user. The input is the proposal document sent from the server. Specifically, it visualizes the document on the user interface and makes it available for review and editing. The output is the document screen that the user views.

[0083] Step 6:

[0084] The user reviews the proposal document and edits it as needed. The input is the proposal document displayed on the terminal. Specifically, the user modifies the text to suit the customer's characteristics, emphasizing wording and product features. The output is the revised proposal document.

[0085] Step 7:

[0086] Users submit feedback on the materials to the system. The input consists of user ratings and opinions. Specifically, users input comments indicating which parts of the materials were effective. The output is feedback data used by the server for the next learning cycle.

[0087] (Application Example 1)

[0088] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0089] In modern household proposal activities, there is a need to efficiently create optimal proposal materials based on individual conditions and past cases. However, with current methods, the analysis of past cases and the generation of proposal materials are often done manually, requiring a tremendous amount of time and effort, making efficient proposal activities difficult. To solve this problem, there is a need for a system that utilizes household robots and smart devices to support efficiently optimized proposal activities.

[0090] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0091] In this invention, the server includes means for extracting data from a set of past negotiation information, means for preprocessing the extracted information and formatting it as input data for a data analysis model, and means for presenting a trained model for creating proposal materials using the preprocessed data. This enables efficient support for proposal activities within the home and automatic generation of optimal proposal materials based on past cases.

[0092] "Past negotiation information" refers to historical data of past transactions and business negotiations related to proposal activities.

[0093] A "data analysis model" refers to the algorithms and analytical methods used to analyze historical data and generate proposal documents.

[0094] A "pre-trained model" refers to a model that has been developed using machine learning based on past information and is intended for use in generating proposal documents.

[0095] "Household suggestions" refer to suggestions made in relation to activities and decisions within the household, and specifically include things like savings plans and education plans.

[0096] "Information display means" refers to devices or functions that display generated proposal documents so that users can review and modify them.

[0097] "User feedback" refers to opinions and evaluations regarding the suggestion materials that users provide to the system.

[0098] An "optimized proposal document" refers to a document that contains proposals that are appropriately tailored to a specific individual or situation.

[0099] To implement this invention, the system is configured as follows: The server extracts past negotiation information from the database and uses it as the basic data for proposal materials. SQL queries are used to extract the data, and the specific data to be extracted includes transaction history and customer information. This forms a dataset for generating proposal materials.

[0100] Next, the server preprocesses the extracted data and formats it into a suitable format for training the data analysis model. This preprocessing includes data cleaning, normalization, and numerical conversion of categorical data. This process is carried out using Python, utilizing data processing libraries such as Numpy and Pandas.

[0101] Using the formatted data, the server trains machine learning models using TENSORFLOW® or Keras. The trained models are then used to automatically generate personalized recommendations optimized for a specific user or household. These recommendations are intended for home use and may include, for example, savings plans or educational suggestions.

[0102] The terminal displays the generated proposal documents on the interface, allowing the user to review and edit them. The interface is created using HTML / CSS / JavaScript (registered trademark), providing an environment where users can intuitively manipulate the documents.

[0103] Finally, user feedback is collected and used to continuously improve the trained model. Feedback is collected in the form of questionnaires, and the analyzed data is reflected in the next model training.

[0104] One concrete example is the suggestion of energy-saving plans based on household electricity consumption data. The server uses AI to suggest ways to reduce the average daily consumption based on past electricity usage data. These suggestions are customized according to the user's preferences.

[0105] An example of a prompt for a generating AI model is, "Generate optimal energy-saving suggestions based on your household's electricity usage data for the past year."

[0106] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0107] Step 1:

[0108] The server extracts past negotiation information from the database. An SQL query is used as input, and the output is a dataset containing transaction history and customer information. This data forms the basis of the proposal document.

[0109] Step 2:

[0110] The server preprocesses the extracted data. This process uses Python, employing NumPy and Pandas for data cleaning, missing value imputation, and normalization. The input is the dataset, and the output is formatted data usable by the analysis model.

[0111] Step 3:

[0112] The server uses TensorFlow or Keras to train the formatted data. The input is pre-processed data, and the output is a trained generative AI model. This model is used to automatically generate proposal documents.

[0113] Step 4:

[0114] The server uses a pre-trained model to generate optimal recommendation materials tailored to a specific individual. The input consists of the individual's attributes and past usage data, while the output is customized recommendation materials.

[0115] Step 5:

[0116] The terminal displays the generated proposal document on the user interface. The user can review this document and edit its contents as needed. The displayed document is the generated proposal document, and its contents are updated based on user actions.

[0117] Step 6:

[0118] Users provide feedback on the proposed document. This feedback is entered in a questionnaire format and saved on the server as data for the next model training.

[0119] Step 7:

[0120] The server analyzes the collected feedback and uses it to improve the generated AI model. It uses user feedback as input data and obtains a learning model with improved accuracy as output.

[0121] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0122] This invention is a system that not only streamlines the creation of proposal materials in the sales department, but also has the function of recognizing the user's emotional state and optimizing the content of the proposal.

[0123] First, the server extracts past business negotiation data from the company's negotiation database and preprocesses it to prepare it for training. This data is then used to train a generative AI model, which generates proposal materials optimized for specific customers. This process allows for the creation of more persuasive proposal materials by learning patterns from past success stories.

[0124] Once the proposal document is generated, the terminal presents it to the user and provides an interface for reviewing and editing the proposal. At this stage, the user can adjust the document's content and add information tailored to specific customer needs.

[0125] Furthermore, this system incorporates an emotion engine that extracts emotional data from the user's voice and text input. For example, the emotion engine analyzes the emotions of the user based on the tone and content of comments they make while using the proposal document. The server then analyzes this emotional data and optimizes the proposal document, taking into account the user's current mental state.

[0126] For example, if a user is dissatisfied with the content of a proposal document, the emotion engine detects their stress level and adjusts the proposal to emphasize language and examples that are particularly well-received by customers. Conversely, when a positive user response is detected, additional information or suggestions are automatically inserted to maintain that positive tone.

[0127] Ultimately, the feedback users input into the system helps improve the trained model, allowing the entire system to evolve and provide higher-quality proposal materials for future sales meetings. This feedback loop continuously improves the quality of proposals and the efficiency of sales activities.

[0128] The following describes the processing flow.

[0129] Step 1:

[0130] The server extracts historical sales negotiation data from the negotiation database. To do this, it uses SQL queries to retrieve datasets that match specific time periods and criteria. The data includes customer names, industry, deal outcomes, and proposal details.

[0131] Step 2:

[0132] The server preprocesses the extracted data. It performs data cleaning, imputes missing values, and corrects outliers. It also encodes categorical data into numerical data and converts it into a format usable by the model.

[0133] Step 3:

[0134] The server uses pre-processed data to train a generative AI model. This model learns patterns based on past successes, forming the foundation for generating new proposal documents.

[0135] Step 4:

[0136] The device automatically generates new proposal documents using a generation AI model in response to user requests. These documents include customized content tailored to the specific customer's attributes and industry.

[0137] Step 5:

[0138] Users review the generated proposal documents on their devices and edit them as needed. They optimize the documents by adding vocabulary and details tailored to their business needs.

[0139] Step 6:

[0140] The device collects emotional data from the user's voice and input. An emotion engine analyzes this data to estimate the user's emotional state. This data is used to further refine the proposed materials.

[0141] Step 7:

[0142] The server optimizes the content of the proposal document based on sentiment data. For example, if a user expresses dissatisfaction, it automatically makes adjustments such as emphasizing more persuasive examples.

[0143] Step 8:

[0144] Once the user finally approves the proposal, their feedback is collected. The server analyzes this feedback and stores it to use for future model improvements.

[0145] (Example 2)

[0146] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0147] In modern sales activities, there is a need to streamline the creation of proposal materials in order to respond quickly and accurately to the diverse needs of customers. Furthermore, traditional methods make it difficult to optimize proposals while considering the emotional state of the user, making it a challenge to enhance their appeal to customers. Against this backdrop, there is a demand for a system that utilizes user emotional information to automatically generate optimal proposals for each customer.

[0148] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0149] In this invention, the server includes means for extracting information from a past commercial transaction database, means for preprocessing the extracted information and processing it as input for an analysis model, means for learning the preprocessed data and providing a trained model for generating proposal documents, means for automatically generating proposal documents suitable for specific customers, means for analyzing user sentiment information and optimizing the proposal documents, and means for collecting user feedback on the generated proposal documents and using it to improve the model. This makes it possible to effectively and efficiently generate optimal proposal documents tailored to individual customers.

[0150] A "commercial transaction database" is an information system that systematically stores information about past business negotiations and transactions, making it easy to search and extract that information.

[0151] "Preprocessing" refers to the process of preparing data into a format suitable for learning and analysis, and this includes denoising and normalizing the data.

[0152] An "analytical model" is a mathematical model that learns features from data and makes predictions and classifications based on new data.

[0153] A "trained model" is a model that has been trained based on past data and is in a state where it can make inferences on new data.

[0154] A "proposal document" is a document that contains proposals for products and services tailored to the individual needs of each customer, and is used in sales activities.

[0155] "Automatic generation" refers to a function where the system autonomously produces a certain output without user intervention.

[0156] "User emotional information" refers to data that indicates the emotional state of a user, extracted from their voice and text.

[0157] "Optimization" is the process of adjusting conditions and parameters to maximize their effectiveness for a specific purpose.

[0158] "Collecting feedback" is the activity of obtaining opinions and usage results from users to help improve the system.

[0159] The system of this invention operates collaboratively between a server and a terminal to streamline proposal document creation in the commercial sector. The server first extracts historical business negotiation data from the company's internal business transaction database. At this stage, information is retrieved using a database management system such as SQL. Next, the server preprocesses the extracted data to prepare it for use as input to an analysis model. Python libraries such as Pandas and NumPy are used for preprocessing, including data cleaning and normalization.

[0160] Next, the server trains a generative AI model based on this pre-processed data. This process uses deep learning libraries such as TensorFlow and PyTorch. The trained model forms the basis for generating proposal documents that are best suited to a specific customer. As an example of a prompt, the model is given the instruction, "Create a proposal for the latest product for customer A." This automatically generates a persuasive proposal tailored to the user.

[0161] The generated proposal document is presented to the user via a terminal. The terminal provides an interface for the user to review and edit the document. Through this interface, the user can adjust the document through drag-and-drop and direct editing, and add information that meets the customer's needs.

[0162] The system also features an emotion engine to analyze user sentiment. It detects emotions from the voice and text spoken by users while editing a proposed document, and the server optimizes the document based on that sentiment information. For example, if a user comments, "This content needs improvement," the server detects the negative emotion and restructures the document to make it more acceptable.

[0163] Ultimately, users provide feedback to the system, and this new data is used to retrain the model. The server uses this feedback to improve the model's accuracy for the next document generation. This feedback loop allows the system to improve over time, enabling the generation of higher-quality proposal documents.

[0164] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0165] Step 1:

[0166] The server extracts historical sales data from a company's business transaction database. SQL queries are executed as input, and the output includes datasets such as sales history and customer feedback. This data serves as the foundational information necessary for generating proposal documents.

[0167] Step 2:

[0168] The server performs preprocessing on the extracted data. The input is the raw data obtained in step 1, and the output is prepared into a format suitable for input to the analysis model through data cleaning and normalization. Python libraries such as Pandas and NumPy are used to impute missing values ​​and scale the data.

[0169] Step 3:

[0170] The server trains a generative AI model using preprocessed data. Preprocessed data is used as input, and a trained model usable for proposal document generation is obtained as output. TensorFlow or PyTorch is used to train the model by setting appropriate epoch counts and batch sizes.

[0171] Step 4:

[0172] The server uses a trained model to automatically generate proposal documents tailored to specific customers. The input is a prompt (e.g., "Create a proposal for the latest product for customer A"), and the output is a proposal document that meets that request. This document is optimized based on the specific customer's interests.

[0173] Step 5:

[0174] The terminal presents the generated proposal document to the user. The input is the proposal document sent from the server, and the output provides an interface that the user can view and edit. Through the interface, the user can visually review the document and perform drag-and-drop and text editing as needed.

[0175] Step 6:

[0176] The server acquires and analyzes user emotional information. Input includes user voice comments and text input, and output is an analysis result indicating the user's emotional state. An emotion engine is used to analyze tone and word choice to determine the user's mental state.

[0177] Step 7:

[0178] The server optimizes the proposal document based on sentiment information. The input is the sentiment information obtained in step 6 and the edited proposal document, and the output is the optimized proposal document. If a negative user reaction is detected, the wording is adjusted; if a positive reaction is received, the content is improved to maintain that tone.

[0179] Step 8:

[0180] Users provide feedback to the system. Inputs include usage results and opinions on the final proposed document, while output summarizes areas for improvement. This feedback is incorporated into subsequent model training sessions, contributing to overall system quality improvement.

[0181] (Application Example 2)

[0182] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0183] In sales activities, it is crucial to quickly and effectively prepare optimal proposal materials for each customer. However, efficiently and automatically generating proposal materials based on vast amounts of past negotiation data, and further optimizing them in real time according to the customer's emotional state, is difficult. This requires advanced data analysis and user interaction. Conventional systems struggle to integrate and achieve such complex processing.

[0184] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0185] In this invention, the server includes means for extracting information from past negotiation data, means for preprocessing the extracted information and processing it as input to an analysis model, means for learning the preprocessed data and providing a trained model for generating proposal materials, means for automatically generating proposal materials suitable for a specific customer, means for collecting emotional data from the user's visual and voice input using computer vision technology, means for analyzing the emotional data and optimizing the proposal materials, and means for collecting user feedback and using it to improve the trained model. This enables the rapid provision of individually optimized proposal materials for each customer and real-time adjustment of materials based on emotions.

[0186] "Negotiation data" refers to data that includes information about past business negotiations and transactions within a company.

[0187] "Preprocessing" refers to the preparatory work required to appropriately transform raw data into input for an analytical model.

[0188] A "trained model" is an algorithmic model that uses knowledge gained from learning from past data to make predictions or generate data on new data.

[0189] A "proposal document" is a document or presentation created to explain the appeal of a product or service to a customer and encourage them to make a purchase.

[0190] "Computer vision technology" is a technology that allows computers to extract and analyze information from images and videos.

[0191] "Emotional data" refers to information about a person's emotional state, extracted from their voice, facial expressions, and other data.

[0192] "Optimization" is the act of adjusting a system or process to its best possible state according to its purpose.

[0193] "Feedback" refers to information provided as input, such as evaluations and opinions on the system's output, which is used to improve performance.

[0194] Based on this invention, the following embodiments can be considered in order to actually streamline sales activities and build a system that provides optimal proposals to customers.

[0195] The server first extracts data on past negotiations from the database. The extracted data is preprocessed and appropriately adapted as input for the analysis model. The adapted data is used to train the generative AI model, generating a trained model. This automatically generates proposal materials tailored to specific customers.

[0196] As a terminal, store staff use smart glasses during business negotiations. The smart glasses display generated proposal materials, which users use to make proposals to customers. Furthermore, the camera and microphone built into the smart glasses record the customer's facial expressions and voice, and transmit this data to a server in real time. Using computer vision and voice analysis technologies, the server extracts and analyzes the customer's emotional data.

[0197] The server further optimizes the proposal materials using the extracted emotional data. This process ensures that information and proposals are presented in a way that aligns with the customer's emotions and interests. For example, when a customer smiles, relevant product and special offer information is displayed on the smart glasses, facilitating a smoother business negotiation.

[0198] The feedback users provide to the system is collected by the server and used to improve subsequent sales materials. This feedback loop allows the system to continuously evolve and improve the quality of proposals. In this way, optimized proposals for each customer can be delivered quickly, increasing the success rate of sales deals.

[0199] An example of a prompt message might be, "Please display additional information if the customer gives a positive response at the next business meeting."

[0200] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0201] Step 1:

[0202] The server extracts information from a database of past negotiations. This extraction process filters deal history and customer information related to the business domain to select the necessary datasets. The input is the entire deal history in the database, and the output is the selected deal data.

[0203] Step 2:

[0204] The server preprocesses the extracted sales opportunity data and transforms it appropriately as input for the generating AI model. This process includes data normalization, missing value imputation, and feature encoding. The input is selected sales opportunity data, and the output is in a data format that the machine learning model can understand.

[0205] Step 3:

[0206] The server trains a generative AI model using preprocessed data to create a trained model. This model learns from past successful patterns and generates predictions and suggestion materials for new data. The input is the training data, and the output is the trained generative AI model.

[0207] Step 4:

[0208] The server automatically generates proposal materials tailored to specific customers. It uses a generation AI model to construct customized materials based on customer attributes and past behavior. The input is customer information and a trained model, and the output is a specific proposal document.

[0209] Step 5:

[0210] The smart glasses, acting as a terminal, display proposal materials received from the server. Store staff, acting as users, interact with customers based on these materials. The input is the proposal materials, and the output is the staff's presentation to the customer.

[0211] Step 6:

[0212] The smart glasses use a built-in camera and microphone to record the customer's facial expressions and voice in real time and transmit them to a server. The input is the customer's facial expressions and voice information, and the output is the transmission of data to the server.

[0213] Step 7:

[0214] The server analyzes the received customer's facial expressions and voice data using computer vision and voice analysis technologies to extract emotional data. The input is the customer's visual and voice data, and the output is the analyzed emotional data.

[0215] Step 8:

[0216] The server optimizes proposal materials in real time based on sentiment data. It makes adjustments such as adding upsell suggestions based on positive emotions or emphasizing reassuring information based on negative emotions. The input is analyzed sentiment data, and the output is the optimized proposal material.

[0217] Step 9:

[0218] User feedback is sent from the smart glasses to the server and used to create proposals for future updates. The input consists of user evaluations and improvement requests, and the output is data that contributes to the continuous improvement of the system.

[0219] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating 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.

[0220] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0221] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0222] [Second Embodiment]

[0223] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0224] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0225] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0226] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0227] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0228] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0229] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0230] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0231] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0232] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0233] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0234] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. 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".

[0235] This invention is a system designed to improve the efficiency of proposal document creation in the sales department, and is implemented by the following means. First, the server extracts past sales negotiation data from the company's negotiation database. At this time, the server uses database queries to obtain necessary sales negotiation information (customer name, industry, negotiation details, etc.). This data forms the basis of a trained model for generating proposal documents.

[0236] Next, the server preprocesses the acquired sales opportunity data. In this process, the server performs data cleaning, such as imputing missing values, removing outliers, and normalizing the data. It also prepares the data into a format that the model can learn from, such as converting categorical data into numerical data.

[0237] The server uses pre-processed data to train a generative AI model. This trained model is designed to present optimal proposals for specific customers, and by referencing past success stories, it can generate more effective sales materials.

[0238] The generated proposal document is sent to the terminal, where the user can review it on the interface and edit it as needed. For example, the user can fine-tune the wording of the proposal document to suit a specific customer or to emphasize the product's features.

[0239] Finally, after the user completes the proposal document, the system collects feedback from the user. The user can point out which parts of the document were effective, and the server can incorporate this feedback data into the next learning cycle to improve the accuracy and quality of the generated AI model. This feedback loop continuously improves the efficiency and effectiveness of proposal document creation as a whole system.

[0240] The following describes the processing flow.

[0241] Step 1:

[0242] The server connects to the negotiation database and extracts past deal data. It issues SQL queries to collect data based on specific criteria (for example, deal history for the past 6 months).

[0243] Step 2:

[0244] The server preprocesses the extracted data. It sorts the data types and imputes any missing values ​​using appropriate methods. Furthermore, it encodes categorical variables into numerical values ​​and converts the data into a format that is easy for the model to handle.

[0245] Step 3:

[0246] The server trains the generative AI model using preprocessed data. Training is performed using an iterative learning algorithm, which learns features based on past successes.

[0247] Step 4:

[0248] The device automatically generates new proposal materials using a pre-trained generative AI model in response to user requests. These generated materials include personalized content for each customer.

[0249] Step 5:

[0250] Users review the proposal documents on their devices. If necessary, they edit the wording of the documents to tailor them to the specific needs of the client.

[0251] Step 6:

[0252] After the user gives final approval to the proposal, they enter feedback. The system collects this feedback and stores it on the server as data that can be used for future model improvements.

[0253] (Example 1)

[0254] Next, we will describe Example 1. 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."

[0255] In sales operations, efficiently and quickly creating proposal materials tailored to each customer is challenging, and improving the accuracy of these materials, especially by leveraging past success stories, requires significant time and effort. Furthermore, there is a need for a means to continuously improve the effectiveness of automatically generated proposal materials in actual business negotiations.

[0256] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0257] In this invention, the server includes means for extracting data from past negotiation information, means for preprocessing the extracted data and shaping it as input for an analysis model, and means for providing a trained model for proposal creation based on the preprocessed data. This enables a system that efficiently creates optimal proposals for each customer and includes a feedback process to improve their effectiveness.

[0258] "Past negotiation information" refers to documents and records related to business negotiations and transactions that have taken place in the past during a company's activities.

[0259] "Means of data extraction" refers to methods or devices for selecting and retrieving necessary information from databases or records.

[0260] "Preprocessing" refers to the process of preparing data to be analyzable, and specifically includes imputing missing values, removing outliers, and normalization.

[0261] "Model input for analysis" refers to data formatted in a way that machine learning models can use for learning and inference.

[0262] A "trained model" is a machine learning model that has been trained on existing data and is used to perform a specific task.

[0263] A "proposal" is a document outlining a plan or information submitted to a customer or business partner, and is used for business negotiations and contract signing.

[0264] "Feedback" refers to information provided to evaluate a system or process and to facilitate improvement.

[0265] "Methods for removing outliers" refer to methods or processes for identifying and eliminating inappropriate data points within a dataset.

[0266] "Means of converting categorical data into numerical format" refers to methods of replacing categorical data with numerical formats, enabling computational processing in machine learning.

[0267] This invention consists of a system that streamlines the creation of proposal materials in sales operations. Specifically, various processes are carried out primarily using a server, terminal, and user.

[0268] The server extracts necessary data from the company's negotiation information database. Using database queries, the server extracts important information related to past business negotiations, such as customer names, industries, and negotiation details. This data then forms the basis for proposal generation.

[0269] Next, the server preprocesses the extracted data. This preprocessing includes imputing missing values, removing outliers, and normalizing the data. It also converts categorical data into numerical format, making it suitable for machine learning models to learn from. This preprocessing improves the accuracy of the data.

[0270] The server uses the formatted data to train a generative AI model. The trained model is then used to provide customers with the most suitable proposals. This includes the ability to leverage past success stories to generate sales materials more effectively.

[0271] The generated proposal document is sent to the terminal and can be reviewed by the user through the interface. The user can review the document and edit it to suit a specific customer. Specifically, it is possible to adjust it to the customer's specialized terminology and emphasize the features of specific products.

[0272] For example, when a user creates a sales proposal for a new product, the server generates the initial proposal by retrieving data from past success stories and applying it to a learning model. An example of a prompt message as text might be: "Generate a proposal for a new product. The target market is small and medium-sized enterprises, and cost efficiency should be emphasized." This allows the system to provide the user with more specific and compelling materials, helping subsequent sales negotiations proceed smoothly.

[0273] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0274] Step 1:

[0275] The server queries a company's negotiation information database to extract data. The input database contains past business negotiation information. Specifically, it retrieves information such as customer name, industry, and negotiation details from the database. The output is negotiation data that forms the basis for creating proposal documents.

[0276] Step 2:

[0277] The server preprocesses the extracted sales opportunity data. The raw data obtained in the previous step is used as input. Specifically, it performs operations such as imputing missing values, removing outliers, and normalizing the data, converting categorical data into numerical format. The output is data converted into a format that can be processed by machine learning models.

[0278] Step 3:

[0279] The server trains a generative AI model using preprocessed data. The input is the formatted data. As a specific operation, the model is trained based on past negotiation success cases. The output is a trained model that can generate optimal proposals for specific customers.

[0280] Step 4:

[0281] The server generates proposal materials using the trained model. The input is data and requirements related to specific customers. As a specific operation, the generative AI model creates proposal content suitable for the customer. The output is automatically generated proposal materials.

[0282] Step 5:

[0283] The terminal displays the generated proposal materials to the user. The input is the proposal materials sent from the server. As a specific operation, the materials are visualized on the user interface and made viewable and editable. The output is the material screen for the user to view.

[0284] Step 6:

[0285] The user checks the proposal materials and edits them if necessary. The input is the proposal materials displayed on the terminal. As a specific operation, the text is modified according to the customer's characteristics, and the diction and product features are emphasized. The output is the modified proposal materials.

[0286] Step 7:

[0287] The user sends feedback on the materials to the system. The input is the user's evaluation and opinions. As a specific operation, comments are entered on which parts of the materials were effective. The output is feedback data for the server to utilize in the next learning cycle.

[0288] (Application Example 1)

[0289] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0290] In modern household proposal activities, there is a need to efficiently create optimal proposal materials based on individual conditions and past cases. However, with current methods, the analysis of past cases and the generation of proposal materials are often done manually, requiring a tremendous amount of time and effort, making efficient proposal activities difficult. To solve this problem, there is a need for a system that utilizes household robots and smart devices to support efficiently optimized proposal activities.

[0291] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0292] In this invention, the server includes means for extracting data from a set of past negotiation information, means for preprocessing the extracted information and formatting it as input data for a data analysis model, and means for presenting a trained model for creating proposal materials using the preprocessed data. This enables efficient support for proposal activities within the home and automatic generation of optimal proposal materials based on past cases.

[0293] "Past negotiation information" refers to historical data of past transactions and business negotiations related to proposal activities.

[0294] A "data analysis model" refers to the algorithms and analytical methods used to analyze historical data and generate proposal documents.

[0295] A "pre-trained model" refers to a model that has been developed using machine learning based on past information and is intended for use in generating proposal documents.

[0296] "Household suggestions" refer to suggestions made in relation to activities and decisions within the household, and specifically include things like savings plans and education plans.

[0297] "Information display means" refers to devices or functions that display generated proposal documents so that users can review and modify them.

[0298] "User feedback" refers to opinions and evaluations regarding the suggestion materials that users provide to the system.

[0299] An "optimized proposal document" refers to a document that contains proposals that are appropriately tailored to a specific individual or situation.

[0300] To implement this invention, the system is configured as follows: The server extracts past negotiation information from the database and uses it as the basic data for proposal materials. SQL queries are used to extract the data, and the specific data to be extracted includes transaction history and customer information. This forms a dataset for generating proposal materials.

[0301] Next, the server preprocesses the extracted data and formats it into a suitable format for training the data analysis model. This preprocessing includes data cleaning, normalization, and numerical conversion of categorical data. This process is carried out using Python, utilizing data processing libraries such as Numpy and Pandas.

[0302] Using the formatted data, the server trains machine learning models using TensorFlow or Keras. The trained models are then used to automatically generate personalized recommendations optimized for a specific user or household. These recommendations are intended for home use and may include, for example, savings plans or educational suggestions.

[0303] The terminal displays the generated proposal documents on an interface, allowing the user to review and edit them. The interface is created using HTML / CSS / JavaScript, providing an environment where users can intuitively manipulate the documents.

[0304] Finally, feedback from users is collected, and based on this, the pre-trained learning model is gradually improved. The feedback is collected in the form of a questionnaire, and the analyzed data is reflected in the next model training.

[0305] As a specific example, there is a proposal for a savings plan based on household electricity consumption data. The server uses the past electricity usage data to propose to the AI "how to reduce the average daily consumption". This proposal is customized according to the user's preferences.

[0306] An example of a prompt sentence for the generative AI model is "Please generate an optimal savings proposal based on the electricity usage data of the household for the past year."

[0307] The flow of the specific process in Application Example 1 will be described using FIG. 12.

[0308] Step 1:

[0309] The server extracts past negotiation information groups from the database. An SQL query is used as the input, and a dataset containing transaction history and customer information is obtained as the output. This data is the information that forms the basis of the proposal document.

[0310] Step 2:

[0311] The server preprocesses the extracted data. In this process, Python is used, and data cleaning, missing value imputation, and normalization are performed using Numpy and Pandas. The input is the dataset, and the output is the formatted data that can be used in the analysis model.

[0312] Step 3:

[0313] The server performs machine learning on the formatted data using TensorFlow or Keras. The input is the preprocessed data, and the output is the pre-trained generative AI model. This model is used for the automatic generation of the proposal document.

[0314] Step 4:

[0315] The server uses a pre-trained model to generate optimal recommendation materials tailored to a specific individual. The input consists of the individual's attributes and past usage data, while the output is customized recommendation materials.

[0316] Step 5:

[0317] The terminal displays the generated proposal document on the user interface. The user can review this document and edit its contents as needed. The displayed document is the generated proposal document, and its contents are updated based on user actions.

[0318] Step 6:

[0319] Users provide feedback on the proposed document. This feedback is entered in a questionnaire format and saved on the server as data for the next model training.

[0320] Step 7:

[0321] The server analyzes the collected feedback and uses it to improve the generated AI model. It uses user feedback as input data and obtains a learning model with improved accuracy as output.

[0322] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0323] This invention is a system that not only streamlines the creation of proposal materials in the sales department, but also has the function of recognizing the user's emotional state and optimizing the content of the proposal.

[0324] First, the server extracts past business negotiation data from the company's negotiation database and preprocesses it to prepare it for training. This data is then used to train a generative AI model, which generates proposal materials optimized for specific customers. This process allows for the creation of more persuasive proposal materials by learning patterns from past success stories.

[0325] Once the proposal document is generated, the terminal presents it to the user and provides an interface for reviewing and editing the proposal. At this stage, the user can adjust the document's content and add information tailored to specific customer needs.

[0326] Furthermore, this system incorporates an emotion engine that extracts emotional data from the user's voice and text input. For example, the emotion engine analyzes the emotions of the user based on the tone and content of comments they make while using the proposal document. The server then analyzes this emotional data and optimizes the proposal document, taking into account the user's current mental state.

[0327] For example, if a user is dissatisfied with the content of a proposal document, the emotion engine detects their stress level and adjusts the proposal to emphasize language and examples that are particularly well-received by customers. Conversely, when a positive user response is detected, additional information or suggestions are automatically inserted to maintain that positive tone.

[0328] Ultimately, the feedback users input into the system helps improve the trained model, allowing the entire system to evolve and provide higher-quality proposal materials for future sales meetings. This feedback loop continuously improves the quality of proposals and the efficiency of sales activities.

[0329] The following describes the processing flow.

[0330] Step 1:

[0331] The server extracts historical sales negotiation data from the negotiation database. To do this, it uses SQL queries to retrieve datasets that match specific time periods and criteria. The data includes customer names, industry, deal outcomes, and proposal details.

[0332] Step 2:

[0333] The server preprocesses the extracted data. It performs data cleaning, imputes missing values, and corrects outliers. It also encodes categorical data into numerical data and converts it into a format usable by the model.

[0334] Step 3:

[0335] The server uses pre-processed data to train a generative AI model. This model learns patterns based on past successes, forming the foundation for generating new proposal documents.

[0336] Step 4:

[0337] The device automatically generates new proposal documents using a generation AI model in response to user requests. These documents include customized content tailored to the specific customer's attributes and industry.

[0338] Step 5:

[0339] Users review the generated proposal documents on their devices and edit them as needed. They optimize the documents by adding vocabulary and details tailored to their business needs.

[0340] Step 6:

[0341] The device collects emotional data from the user's voice and input. An emotion engine analyzes this data to estimate the user's emotional state. This data is used to further refine the proposed materials.

[0342] Step 7:

[0343] The server optimizes the content of the proposal document based on sentiment data. For example, if a user expresses dissatisfaction, it automatically makes adjustments such as emphasizing more persuasive examples.

[0344] Step 8:

[0345] Once the user finally approves the proposal, their feedback is collected. The server analyzes this feedback and stores it to use for future model improvements.

[0346] (Example 2)

[0347] Next, we will describe Example 2. 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".

[0348] In modern sales activities, there is a need to streamline the creation of proposal materials in order to respond quickly and accurately to the diverse needs of customers. Furthermore, traditional methods make it difficult to optimize proposals while considering the emotional state of the user, making it a challenge to enhance their appeal to customers. Against this backdrop, there is a demand for a system that utilizes user emotional information to automatically generate optimal proposals for each customer.

[0349] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0350] In this invention, the server includes means for extracting information from a past commercial transaction database, means for preprocessing the extracted information and processing it as input for an analysis model, means for learning the preprocessed data and providing a trained model for generating proposal documents, means for automatically generating proposal documents suitable for specific customers, means for analyzing user sentiment information and optimizing the proposal documents, and means for collecting user feedback on the generated proposal documents and using it to improve the model. This makes it possible to effectively and efficiently generate optimal proposal documents tailored to individual customers.

[0351] A "commercial transaction database" is an information system that systematically stores information about past business negotiations and transactions, making it easy to search and extract that information.

[0352] "Preprocessing" refers to the process of preparing data into a format suitable for learning and analysis, and this includes denoising and normalizing the data.

[0353] An "analytical model" is a mathematical model that learns features from data and makes predictions and classifications based on new data.

[0354] A "trained model" is a model that has been trained based on past data and is in a state where it can make inferences on new data.

[0355] A "proposal document" is a document that contains proposals for products and services tailored to the individual needs of each customer, and is used in sales activities.

[0356] "Automatic generation" refers to a function where the system autonomously produces a certain output without user intervention.

[0357] "User emotional information" refers to data that indicates the emotional state of a user, extracted from their voice and text.

[0358] "Optimization" is the process of adjusting conditions and parameters to maximize their effectiveness for a specific purpose.

[0359] "Collecting feedback" is the activity of obtaining opinions and usage results from users to help improve the system.

[0360] The system of this invention operates collaboratively between a server and a terminal to streamline proposal document creation in the commercial sector. The server first extracts historical business negotiation data from the company's internal business transaction database. At this stage, information is retrieved using a database management system such as SQL. Next, the server preprocesses the extracted data to prepare it for use as input to an analysis model. Python libraries such as Pandas and NumPy are used for preprocessing, including data cleaning and normalization.

[0361] Next, the server trains a generative AI model based on this pre-processed data. This process uses deep learning libraries such as TensorFlow and PyTorch. The trained model forms the basis for generating proposal documents that are best suited to a specific customer. As an example of a prompt, the model is given the instruction, "Create a proposal for the latest product for customer A." This automatically generates a persuasive proposal tailored to the user.

[0362] The generated proposal document is presented to the user via a terminal. The terminal provides an interface for the user to review and edit the document. Through this interface, the user can adjust the document through drag-and-drop and direct editing, and add information that meets the customer's needs.

[0363] The system also features an emotion engine to analyze user sentiment. It detects emotions from the voice and text spoken by users while editing a proposed document, and the server optimizes the document based on that sentiment information. For example, if a user comments, "This content needs improvement," the server detects the negative emotion and restructures the document to make it more acceptable.

[0364] Ultimately, users provide feedback to the system, and this new data is used to retrain the model. The server uses this feedback to improve the model's accuracy for the next document generation. This feedback loop allows the system to improve over time, enabling the generation of higher-quality proposal documents.

[0365] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0366] Step 1:

[0367] The server extracts historical sales data from a company's business transaction database. SQL queries are executed as input, and the output includes datasets such as sales history and customer feedback. This data serves as the foundational information necessary for generating proposal documents.

[0368] Step 2:

[0369] The server performs preprocessing on the extracted data. The input is the raw data obtained in step 1, and the output is prepared into a format suitable for input to the analysis model through data cleaning and normalization. Python libraries such as Pandas and NumPy are used to impute missing values ​​and scale the data.

[0370] Step 3:

[0371] The server trains a generative AI model using preprocessed data. Preprocessed data is used as input, and a trained model usable for proposal document generation is obtained as output. TensorFlow or PyTorch is used to train the model by setting appropriate epoch counts and batch sizes.

[0372] Step 4:

[0373] The server uses a trained model to automatically generate proposal documents tailored to specific customers. The input is a prompt (e.g., "Create a proposal for the latest product for customer A"), and the output is a proposal document that meets that request. This document is optimized based on the specific customer's interests.

[0374] Step 5:

[0375] The terminal presents the generated proposal document to the user. The input is the proposal document sent from the server, and the output provides an interface that the user can view and edit. Through the interface, the user can visually review the document and perform drag-and-drop and text editing as needed.

[0376] Step 6:

[0377] The server acquires and analyzes user emotional information. Input includes user voice comments and text input, and output is an analysis result indicating the user's emotional state. An emotion engine is used to analyze tone and word choice to determine the user's mental state.

[0378] Step 7:

[0379] The server optimizes the proposal document based on sentiment information. The input is the sentiment information obtained in step 6 and the edited proposal document, and the output is the optimized proposal document. If a negative user reaction is detected, the wording is adjusted; if a positive reaction is received, the content is improved to maintain that tone.

[0380] Step 8:

[0381] Users provide feedback to the system. Inputs include usage results and opinions on the final proposed document, while output summarizes areas for improvement. This feedback is incorporated into subsequent model training sessions, contributing to overall system quality improvement.

[0382] (Application Example 2)

[0383] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0384] In sales activities, it is crucial to quickly and effectively prepare optimal proposal materials for each customer. However, efficiently and automatically generating proposal materials based on vast amounts of past negotiation data, and further optimizing them in real time according to the customer's emotional state, is difficult. This requires advanced data analysis and user interaction. Conventional systems struggle to integrate and achieve such complex processing.

[0385] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0386] In this invention, the server includes means for extracting information from past negotiation data, means for preprocessing the extracted information and processing it as input to an analysis model, means for learning the preprocessed data and providing a trained model for generating proposal materials, means for automatically generating proposal materials suitable for a specific customer, means for collecting emotional data from the user's visual and voice input using computer vision technology, means for analyzing the emotional data and optimizing the proposal materials, and means for collecting user feedback and using it to improve the trained model. This enables the rapid provision of individually optimized proposal materials for each customer and real-time adjustment of materials based on emotions.

[0387] "Negotiation data" refers to data that includes information about past business negotiations and transactions within a company.

[0388] "Preprocessing" refers to the preparatory work required to appropriately transform raw data into input for an analytical model.

[0389] A "trained model" is an algorithmic model that uses knowledge gained from learning from past data to make predictions or generate data on new data.

[0390] A "proposal document" is a document or presentation created to explain the appeal of a product or service to a customer and encourage them to make a purchase.

[0391] "Computer vision technology" is a technology that allows computers to extract and analyze information from images and videos.

[0392] "Emotional data" refers to information about a person's emotional state, extracted from their voice, facial expressions, and other data.

[0393] "Optimization" is the act of adjusting a system or process to its best possible state according to its purpose.

[0394] "Feedback" refers to information provided as input, such as evaluations and opinions on the system's output, which is used to improve performance.

[0395] Based on this invention, the following embodiments can be considered in order to actually streamline sales activities and build a system that provides optimal proposals to customers.

[0396] The server first extracts data on past negotiations from the database. The extracted data is preprocessed and appropriately adapted as input for the analysis model. The adapted data is used to train the generative AI model, generating a trained model. This automatically generates proposal materials tailored to specific customers.

[0397] As a terminal, store staff use smart glasses during business negotiations. The smart glasses display generated proposal materials, which users use to make proposals to customers. Furthermore, the camera and microphone built into the smart glasses record the customer's facial expressions and voice, and transmit this data to a server in real time. Using computer vision and voice analysis technologies, the server extracts and analyzes the customer's emotional data.

[0398] The server further optimizes the proposal materials using the extracted emotional data. This process ensures that information and proposals are presented in a way that aligns with the customer's emotions and interests. For example, when a customer smiles, relevant product and special offer information is displayed on the smart glasses, facilitating a smoother business negotiation.

[0399] The feedback users provide to the system is collected by the server and used to improve subsequent sales materials. This feedback loop allows the system to continuously evolve and improve the quality of proposals. In this way, optimized proposals for each customer can be delivered quickly, increasing the success rate of sales deals.

[0400] An example of a prompt message might be, "Please display additional information if the customer gives a positive response at the next business meeting."

[0401] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0402] Step 1:

[0403] The server extracts information from a database of past negotiations. This extraction process filters deal history and customer information related to the business domain to select the necessary datasets. The input is the entire deal history in the database, and the output is the selected deal data.

[0404] Step 2:

[0405] The server preprocesses the extracted sales opportunity data and transforms it appropriately as input for the generating AI model. This process includes data normalization, missing value imputation, and feature encoding. The input is selected sales opportunity data, and the output is in a data format that the machine learning model can understand.

[0406] Step 3:

[0407] The server trains a generative AI model using preprocessed data to create a trained model. This model learns from past successful patterns and generates predictions and suggestion materials for new data. The input is the training data, and the output is the trained generative AI model.

[0408] Step 4:

[0409] The server automatically generates proposal materials tailored to specific customers. It uses a generation AI model to construct customized materials based on customer attributes and past behavior. The input is customer information and a trained model, and the output is a specific proposal document.

[0410] Step 5:

[0411] The smart glasses, acting as a terminal, display proposal materials received from the server. Store staff, acting as users, interact with customers based on these materials. The input is the proposal materials, and the output is the staff's presentation to the customer.

[0412] Step 6:

[0413] The smart glasses use a built-in camera and microphone to record the customer's facial expressions and voice in real time and transmit them to a server. The input is the customer's facial expressions and voice information, and the output is the transmission of data to the server.

[0414] Step 7:

[0415] The server analyzes the received customer's facial expressions and voice data using computer vision and voice analysis technologies to extract emotional data. The input is the customer's visual and voice data, and the output is the analyzed emotional data.

[0416] Step 8:

[0417] The server optimizes proposal materials in real time based on sentiment data. It makes adjustments such as adding upsell suggestions based on positive emotions or emphasizing reassuring information based on negative emotions. The input is analyzed sentiment data, and the output is the optimized proposal material.

[0418] Step 9:

[0419] User feedback is sent from the smart glasses to the server and used to create proposals for future updates. The input consists of user evaluations and improvement requests, and the output is data that contributes to the continuous improvement of the system.

[0420] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0421] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0422] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0423] [Third Embodiment]

[0424] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0425] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0426] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0427] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0428] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0429] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0430] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0431] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0432] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0433] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0434] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0435] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0436] This invention is a system designed to improve the efficiency of proposal document creation in the sales department, and is implemented by the following means. First, the server extracts past sales negotiation data from the company's negotiation database. At this time, the server uses database queries to obtain necessary sales negotiation information (customer name, industry, negotiation details, etc.). This data forms the basis of a trained model for generating proposal documents.

[0437] Next, the server preprocesses the acquired sales opportunity data. In this process, the server performs data cleaning, such as imputing missing values, removing outliers, and normalizing the data. It also prepares the data into a format that the model can learn from, such as converting categorical data into numerical data.

[0438] The server uses pre-processed data to train a generative AI model. This trained model is designed to present optimal proposals for specific customers, and by referencing past success stories, it can generate more effective sales materials.

[0439] The generated proposal document is sent to the terminal, where the user can review it on the interface and edit it as needed. For example, the user can fine-tune the wording of the proposal document to suit a specific customer or to emphasize the product's features.

[0440] Finally, after the user completes the proposal document, the system collects feedback from the user. The user can point out which parts of the document were effective, and the server can incorporate this feedback data into the next learning cycle to improve the accuracy and quality of the generated AI model. This feedback loop continuously improves the efficiency and effectiveness of proposal document creation as a whole system.

[0441] The following describes the processing flow.

[0442] Step 1:

[0443] The server connects to the negotiation database and extracts past deal data. It issues SQL queries to collect data based on specific criteria (for example, deal history for the past 6 months).

[0444] Step 2:

[0445] The server preprocesses the extracted data. It sorts the data types and imputes any missing values ​​using appropriate methods. Furthermore, it encodes categorical variables into numerical values ​​and converts the data into a format that is easy for the model to handle.

[0446] Step 3:

[0447] The server trains the generative AI model using preprocessed data. Training is performed using an iterative learning algorithm, which learns features based on past successes.

[0448] Step 4:

[0449] The device automatically generates new proposal materials using a pre-trained generative AI model in response to user requests. These generated materials include personalized content for each customer.

[0450] Step 5:

[0451] Users review the proposal documents on their devices. If necessary, they edit the wording of the documents to tailor them to the specific needs of the client.

[0452] Step 6:

[0453] After the user gives final approval to the proposal, they enter feedback. The system collects this feedback and stores it on the server as data that can be used for future model improvements.

[0454] (Example 1)

[0455] Next, we will describe Example 1. 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."

[0456] In sales operations, efficiently and quickly creating proposal materials tailored to each customer is challenging, and improving the accuracy of these materials, especially by leveraging past success stories, requires significant time and effort. Furthermore, there is a need for a means to continuously improve the effectiveness of automatically generated proposal materials in actual business negotiations.

[0457] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0458] In this invention, the server includes means for extracting data from past negotiation information, means for preprocessing the extracted data and shaping it as input for an analysis model, and means for providing a trained model for proposal creation based on the preprocessed data. This enables a system that efficiently creates optimal proposals for each customer and includes a feedback process to improve their effectiveness.

[0459] "Past negotiation information" refers to documents and records related to business negotiations and transactions that have taken place in the past during a company's activities.

[0460] "Means of data extraction" refers to methods or devices for selecting and retrieving necessary information from databases or records.

[0461] "Preprocessing" refers to the process of preparing data to be analyzable, and specifically includes imputing missing values, removing outliers, and normalization.

[0462] "Model input for analysis" refers to data formatted in a way that machine learning models can use for learning and inference.

[0463] A "trained model" is a machine learning model that has been trained on existing data and is used to perform a specific task.

[0464] A "proposal" is a document outlining a plan or information submitted to a customer or business partner, and is used for business negotiations and contract signing.

[0465] "Feedback" refers to information provided to evaluate a system or process and to facilitate improvement.

[0466] "Methods for removing outliers" refer to methods or processes for identifying and eliminating inappropriate data points within a dataset.

[0467] "Means of converting categorical data into numerical format" refers to methods of replacing categorical data with numerical formats, enabling computational processing in machine learning.

[0468] This invention consists of a system that streamlines the creation of proposal materials in sales operations. Specifically, various processes are carried out primarily using a server, terminal, and user.

[0469] The server extracts necessary data from the company's negotiation information database. Using database queries, the server extracts important information related to past business negotiations, such as customer names, industries, and negotiation details. This data then forms the basis for proposal generation.

[0470] Next, the server preprocesses the extracted data. This preprocessing includes imputing missing values, removing outliers, and normalizing the data. It also converts categorical data into numerical format, making it suitable for machine learning models to learn from. This preprocessing improves the accuracy of the data.

[0471] The server uses the formatted data to train a generative AI model. The trained model is then used to provide customers with the most suitable proposals. This includes the ability to leverage past success stories to generate sales materials more effectively.

[0472] The generated proposal document is sent to the terminal and can be reviewed by the user through the interface. The user can review the document and edit it to suit a specific customer. Specifically, it is possible to adjust it to the customer's specialized terminology and emphasize the features of specific products.

[0473] For example, when a user creates a sales proposal for a new product, the server generates the initial proposal by retrieving data from past success stories and applying it to a learning model. An example of a prompt message as text might be: "Generate a proposal for a new product. The target market is small and medium-sized enterprises, and cost efficiency should be emphasized." This allows the system to provide the user with more specific and compelling materials, helping subsequent sales negotiations proceed smoothly.

[0474] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0475] Step 1:

[0476] The server queries a company's negotiation information database to extract data. The input database contains past business negotiation information. Specifically, it retrieves information such as customer name, industry, and negotiation details from the database. The output is negotiation data that forms the basis for creating proposal documents.

[0477] Step 2:

[0478] The server preprocesses the extracted sales opportunity data. The raw data obtained in the previous step is used as input. Specifically, it performs operations such as imputing missing values, removing outliers, and normalizing the data, converting categorical data into numerical format. The output is data converted into a format that can be processed by machine learning models.

[0479] Step 3:

[0480] The server trains a generative AI model using pre-processed data. The input is formatted data. Specifically, the model is trained based on past successful sales negotiations. The output is a trained model capable of generating optimal proposals for a particular customer.

[0481] Step 4:

[0482] The server generates proposal documents using a pre-trained model. The input consists of data and requirements related to a specific customer. Specifically, the generating AI model creates proposals tailored to the customer. The output is the automatically generated proposal document.

[0483] Step 5:

[0484] The terminal displays the generated proposal document to the user. The input is the proposal document sent from the server. Specifically, it visualizes the document on the user interface and makes it available for review and editing. The output is the document screen that the user views.

[0485] Step 6:

[0486] The user reviews the proposal document and edits it as needed. The input is the proposal document displayed on the terminal. Specifically, the user modifies the text to suit the customer's characteristics, emphasizing wording and product features. The output is the revised proposal document.

[0487] Step 7:

[0488] Users submit feedback on the materials to the system. The input consists of user ratings and opinions. Specifically, users input comments indicating which parts of the materials were effective. The output is feedback data used by the server for the next learning cycle.

[0489] (Application Example 1)

[0490] Next, we will explain Application Example 1. In the following explanation, 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."

[0491] In modern household proposal activities, there is a need to efficiently create optimal proposal materials based on individual conditions and past cases. However, with current methods, the analysis of past cases and the generation of proposal materials are often done manually, requiring a tremendous amount of time and effort, making efficient proposal activities difficult. To solve this problem, there is a need for a system that utilizes household robots and smart devices to support efficiently optimized proposal activities.

[0492] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0493] In this invention, the server includes means for extracting data from a set of past negotiation information, means for preprocessing the extracted information and formatting it as input data for a data analysis model, and means for presenting a trained model for creating proposal materials using the preprocessed data. This enables efficient support for proposal activities within the home and automatic generation of optimal proposal materials based on past cases.

[0494] "Past negotiation information" refers to historical data of past transactions and business negotiations related to proposal activities.

[0495] A "data analysis model" refers to the algorithms and analytical methods used to analyze historical data and generate proposal documents.

[0496] A "pre-trained model" refers to a model that has been developed using machine learning based on past information and is intended for use in generating proposal documents.

[0497] "Household suggestions" refer to suggestions made in relation to activities and decisions within the household, and specifically include things like savings plans and education plans.

[0498] "Information display means" refers to devices or functions that display generated proposal documents so that users can review and modify them.

[0499] "User feedback" refers to opinions and evaluations regarding the suggestion materials that users provide to the system.

[0500] An "optimized proposal document" refers to a document that contains proposals that are appropriately tailored to a specific individual or situation.

[0501] To implement this invention, the system is configured as follows: The server extracts past negotiation information from the database and uses it as the basic data for proposal materials. SQL queries are used to extract the data, and the specific data to be extracted includes transaction history and customer information. This forms a dataset for generating proposal materials.

[0502] Next, the server preprocesses the extracted data and formats it into a suitable format for training the data analysis model. This preprocessing includes data cleaning, normalization, and numerical conversion of categorical data. This process is carried out using Python, utilizing data processing libraries such as Numpy and Pandas.

[0503] Using the formatted data, the server trains machine learning models using TensorFlow or Keras. The trained models are then used to automatically generate personalized recommendations optimized for a specific user or household. These recommendations are intended for home use and may include, for example, savings plans or educational suggestions.

[0504] The terminal displays the generated proposal documents on an interface, allowing the user to review and edit them. The interface is created using HTML / CSS / JavaScript, providing an environment where users can intuitively manipulate the documents.

[0505] Finally, user feedback is collected and used to continuously improve the trained model. Feedback is collected in the form of questionnaires, and the analyzed data is reflected in the next model training.

[0506] One concrete example is the suggestion of energy-saving plans based on household electricity consumption data. The server uses AI to suggest ways to reduce the average daily consumption based on past electricity usage data. These suggestions are customized according to the user's preferences.

[0507] An example of a prompt for a generating AI model is, "Generate optimal energy-saving suggestions based on your household's electricity usage data for the past year."

[0508] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0509] Step 1:

[0510] The server extracts past negotiation information from the database. An SQL query is used as input, and the output is a dataset containing transaction history and customer information. This data forms the basis of the proposal document.

[0511] Step 2:

[0512] The server preprocesses the extracted data. This process uses Python, employing NumPy and Pandas for data cleaning, missing value imputation, and normalization. The input is the dataset, and the output is formatted data usable by the analysis model.

[0513] Step 3:

[0514] The server uses TensorFlow or Keras to train the formatted data. The input is pre-processed data, and the output is a trained generative AI model. This model is used to automatically generate proposal documents.

[0515] Step 4:

[0516] The server uses a pre-trained model to generate optimal recommendation materials tailored to a specific individual. The input consists of the individual's attributes and past usage data, while the output is customized recommendation materials.

[0517] Step 5:

[0518] The terminal displays the generated proposal document on the user interface. The user can review this document and edit its contents as needed. The displayed document is the generated proposal document, and its contents are updated based on user actions.

[0519] Step 6:

[0520] Users provide feedback on the proposed document. This feedback is entered in a questionnaire format and saved on the server as data for the next model training.

[0521] Step 7:

[0522] The server analyzes the collected feedback and uses it to improve the generated AI model. It uses user feedback as input data and obtains a learning model with improved accuracy as output.

[0523] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0524] This invention is a system that not only streamlines the creation of proposal materials in the sales department, but also has the function of recognizing the user's emotional state and optimizing the content of the proposal.

[0525] First, the server extracts past business negotiation data from the company's negotiation database and preprocesses it to prepare it for training. This data is then used to train a generative AI model, which generates proposal materials optimized for specific customers. This process allows for the creation of more persuasive proposal materials by learning patterns from past success stories.

[0526] Once the proposal document is generated, the terminal presents it to the user and provides an interface for reviewing and editing the proposal. At this stage, the user can adjust the document's content and add information tailored to specific customer needs.

[0527] Furthermore, this system incorporates an emotion engine that extracts emotional data from the user's voice and text input. For example, the emotion engine analyzes the emotions of the user based on the tone and content of comments they make while using the proposal document. The server then analyzes this emotional data and optimizes the proposal document, taking into account the user's current mental state.

[0528] For example, if a user is dissatisfied with the content of a proposal document, the emotion engine detects their stress level and adjusts the proposal to emphasize language and examples that are particularly well-received by customers. Conversely, when a positive user response is detected, additional information or suggestions are automatically inserted to maintain that positive tone.

[0529] Ultimately, the feedback users input into the system helps improve the trained model, allowing the entire system to evolve and provide higher-quality proposal materials for future sales meetings. This feedback loop continuously improves the quality of proposals and the efficiency of sales activities.

[0530] The following describes the processing flow.

[0531] Step 1:

[0532] The server extracts historical sales negotiation data from the negotiation database. To do this, it uses SQL queries to retrieve datasets that match specific time periods and criteria. The data includes customer names, industry, deal outcomes, and proposal details.

[0533] Step 2:

[0534] The server preprocesses the extracted data. It performs data cleaning, imputes missing values, and corrects outliers. It also encodes categorical data into numerical data and converts it into a format usable by the model.

[0535] Step 3:

[0536] The server uses pre-processed data to train a generative AI model. This model learns patterns based on past successes, forming the foundation for generating new proposal documents.

[0537] Step 4:

[0538] The device automatically generates new proposal documents using a generation AI model in response to user requests. These documents include customized content tailored to the specific customer's attributes and industry.

[0539] Step 5:

[0540] Users review the generated proposal documents on their devices and edit them as needed. They optimize the documents by adding vocabulary and details tailored to their business needs.

[0541] Step 6:

[0542] The device collects emotional data from the user's voice and input. An emotion engine analyzes this data to estimate the user's emotional state. This data is used to further refine the proposed materials.

[0543] Step 7:

[0544] The server optimizes the content of the proposal document based on sentiment data. For example, if a user expresses dissatisfaction, it automatically makes adjustments such as emphasizing more persuasive examples.

[0545] Step 8:

[0546] Once the user finally approves the proposal, their feedback is collected. The server analyzes this feedback and stores it to use for future model improvements.

[0547] (Example 2)

[0548] Next, we will describe Example 2. 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."

[0549] In modern sales activities, there is a need to streamline the creation of proposal materials in order to respond quickly and accurately to the diverse needs of customers. Furthermore, traditional methods make it difficult to optimize proposals while considering the emotional state of the user, making it a challenge to enhance their appeal to customers. Against this backdrop, there is a demand for a system that utilizes user emotional information to automatically generate optimal proposals for each customer.

[0550] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0551] In this invention, the server includes means for extracting information from a past commercial transaction database, means for preprocessing the extracted information and processing it as input for an analysis model, means for learning the preprocessed data and providing a trained model for generating proposal documents, means for automatically generating proposal documents suitable for specific customers, means for analyzing user sentiment information and optimizing the proposal documents, and means for collecting user feedback on the generated proposal documents and using it to improve the model. This makes it possible to effectively and efficiently generate optimal proposal documents tailored to individual customers.

[0552] A "commercial transaction database" is an information system that systematically stores information about past business negotiations and transactions, making it easy to search and extract that information.

[0553] "Preprocessing" refers to the process of preparing data into a format suitable for learning and analysis, and this includes denoising and normalizing the data.

[0554] An "analytical model" is a mathematical model that learns features from data and makes predictions and classifications based on new data.

[0555] A "trained model" is a model that has been trained based on past data and is in a state where it can make inferences on new data.

[0556] A "proposal document" is a document that contains proposals for products and services tailored to the individual needs of each customer, and is used in sales activities.

[0557] "Automatic generation" refers to a function where the system autonomously produces a certain output without user intervention.

[0558] "User emotional information" refers to data that indicates the emotional state of a user, extracted from their voice and text.

[0559] "Optimization" is the process of adjusting conditions and parameters to maximize their effectiveness for a specific purpose.

[0560] "Collecting feedback" is the activity of obtaining opinions and usage results from users to help improve the system.

[0561] The system of this invention operates collaboratively between a server and a terminal to streamline proposal document creation in the commercial sector. The server first extracts historical business negotiation data from the company's internal business transaction database. At this stage, information is retrieved using a database management system such as SQL. Next, the server preprocesses the extracted data to prepare it for use as input to an analysis model. Python libraries such as Pandas and NumPy are used for preprocessing, including data cleaning and normalization.

[0562] Next, the server trains a generative AI model based on this pre-processed data. This process uses deep learning libraries such as TensorFlow and PyTorch. The trained model forms the basis for generating proposal documents that are best suited to a specific customer. As an example of a prompt, the model is given the instruction, "Create a proposal for the latest product for customer A." This automatically generates a persuasive proposal tailored to the user.

[0563] The generated proposal document is presented to the user via a terminal. The terminal provides an interface for the user to review and edit the document. Through this interface, the user can adjust the document through drag-and-drop and direct editing, and add information that meets the customer's needs.

[0564] The system also features an emotion engine to analyze user sentiment. It detects emotions from the voice and text spoken by users while editing a proposed document, and the server optimizes the document based on that sentiment information. For example, if a user comments, "This content needs improvement," the server detects the negative emotion and restructures the document to make it more acceptable.

[0565] Ultimately, users provide feedback to the system, and this new data is used to retrain the model. The server uses this feedback to improve the model's accuracy for the next document generation. This feedback loop allows the system to improve over time, enabling the generation of higher-quality proposal documents.

[0566] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0567] Step 1:

[0568] The server extracts historical sales data from a company's business transaction database. SQL queries are executed as input, and the output includes datasets such as sales history and customer feedback. This data serves as the foundational information necessary for generating proposal documents.

[0569] Step 2:

[0570] The server performs preprocessing on the extracted data. The input is the raw data obtained in step 1, and the output is prepared into a format suitable for input to the analysis model through data cleaning and normalization. Python libraries such as Pandas and NumPy are used to impute missing values ​​and scale the data.

[0571] Step 3:

[0572] The server trains a generative AI model using preprocessed data. Preprocessed data is used as input, and a trained model usable for proposal document generation is obtained as output. TensorFlow or PyTorch is used to train the model by setting appropriate epoch counts and batch sizes.

[0573] Step 4:

[0574] The server uses a trained model to automatically generate proposal documents tailored to specific customers. The input is a prompt (e.g., "Create a proposal for the latest product for customer A"), and the output is a proposal document that meets that request. This document is optimized based on the specific customer's interests.

[0575] Step 5:

[0576] The terminal presents the generated proposal document to the user. The input is the proposal document sent from the server, and the output provides an interface that the user can view and edit. Through the interface, the user can visually review the document and perform drag-and-drop and text editing as needed.

[0577] Step 6:

[0578] The server acquires and analyzes user emotional information. Input includes user voice comments and text input, and output is an analysis result indicating the user's emotional state. An emotion engine is used to analyze tone and word choice to determine the user's mental state.

[0579] Step 7:

[0580] The server optimizes the proposal document based on sentiment information. The input is the sentiment information obtained in step 6 and the edited proposal document, and the output is the optimized proposal document. If a negative user reaction is detected, the wording is adjusted; if a positive reaction is received, the content is improved to maintain that tone.

[0581] Step 8:

[0582] Users provide feedback to the system. Inputs include usage results and opinions on the final proposed document, while output summarizes areas for improvement. This feedback is incorporated into subsequent model training sessions, contributing to overall system quality improvement.

[0583] (Application Example 2)

[0584] Next, we will explain application example 2. In the following explanation, 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."

[0585] In sales activities, it is crucial to quickly and effectively prepare optimal proposal materials for each customer. However, efficiently and automatically generating proposal materials based on vast amounts of past negotiation data, and further optimizing them in real time according to the customer's emotional state, is difficult. This requires advanced data analysis and user interaction. Conventional systems struggle to integrate and achieve such complex processing.

[0586] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0587] In this invention, the server includes means for extracting information from past negotiation data, means for preprocessing the extracted information and processing it as input to an analysis model, means for learning the preprocessed data and providing a trained model for generating proposal materials, means for automatically generating proposal materials suitable for a specific customer, means for collecting emotional data from the user's visual and voice input using computer vision technology, means for analyzing the emotional data and optimizing the proposal materials, and means for collecting user feedback and using it to improve the trained model. This enables the rapid provision of individually optimized proposal materials for each customer and real-time adjustment of materials based on emotions.

[0588] "Negotiation data" refers to data that includes information about past business negotiations and transactions within a company.

[0589] "Preprocessing" refers to the preparatory work required to appropriately transform raw data into input for an analytical model.

[0590] A "trained model" is an algorithmic model that uses knowledge gained from learning from past data to make predictions or generate data on new data.

[0591] A "proposal document" is a document or presentation created to explain the appeal of a product or service to a customer and encourage them to make a purchase.

[0592] "Computer vision technology" is a technology that allows computers to extract and analyze information from images and videos.

[0593] "Emotional data" refers to information about a person's emotional state, extracted from their voice, facial expressions, and other data.

[0594] "Optimization" is the act of adjusting a system or process to its best possible state according to its purpose.

[0595] "Feedback" refers to information provided as input, such as evaluations and opinions on the system's output, which is used to improve performance.

[0596] Based on this invention, the following embodiments can be considered in order to actually streamline sales activities and build a system that provides optimal proposals to customers.

[0597] The server first extracts data on past negotiations from the database. The extracted data is preprocessed and appropriately adapted as input for the analysis model. The adapted data is used to train the generative AI model, generating a trained model. This automatically generates proposal materials tailored to specific customers.

[0598] As a terminal, store staff use smart glasses during business negotiations. The smart glasses display generated proposal materials, which users use to make proposals to customers. Furthermore, the camera and microphone built into the smart glasses record the customer's facial expressions and voice, and transmit this data to a server in real time. Using computer vision and voice analysis technologies, the server extracts and analyzes the customer's emotional data.

[0599] The server further optimizes the proposal materials using the extracted emotional data. This process ensures that information and proposals are presented in a way that aligns with the customer's emotions and interests. For example, when a customer smiles, relevant product and special offer information is displayed on the smart glasses, facilitating a smoother business negotiation.

[0600] The feedback users provide to the system is collected by the server and used to improve subsequent sales materials. This feedback loop allows the system to continuously evolve and improve the quality of proposals. In this way, optimized proposals for each customer can be delivered quickly, increasing the success rate of sales deals.

[0601] An example of a prompt message might be, "Please display additional information if the customer gives a positive response at the next business meeting."

[0602] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0603] Step 1:

[0604] The server extracts information from a database of past negotiations. This extraction process filters deal history and customer information related to the business domain to select the necessary datasets. The input is the entire deal history in the database, and the output is the selected deal data.

[0605] Step 2:

[0606] The server preprocesses the extracted sales opportunity data and transforms it appropriately as input for the generating AI model. This process includes data normalization, missing value imputation, and feature encoding. The input is selected sales opportunity data, and the output is in a data format that the machine learning model can understand.

[0607] Step 3:

[0608] The server trains a generative AI model using preprocessed data to create a trained model. This model learns from past successful patterns and generates predictions and suggestion materials for new data. The input is the training data, and the output is the trained generative AI model.

[0609] Step 4:

[0610] The server automatically generates proposal materials tailored to specific customers. It uses a generation AI model to construct customized materials based on customer attributes and past behavior. The input is customer information and a trained model, and the output is a specific proposal document.

[0611] Step 5:

[0612] The smart glasses, acting as a terminal, display proposal materials received from the server. Store staff, acting as users, interact with customers based on these materials. The input is the proposal materials, and the output is the staff's presentation to the customer.

[0613] Step 6:

[0614] The smart glasses use a built-in camera and microphone to record the customer's facial expressions and voice in real time and transmit them to a server. The input is the customer's facial expressions and voice information, and the output is the transmission of data to the server.

[0615] Step 7:

[0616] The server analyzes the received customer's facial expressions and voice data using computer vision and voice analysis technologies to extract emotional data. The input is the customer's visual and voice data, and the output is the analyzed emotional data.

[0617] Step 8:

[0618] The server optimizes proposal materials in real time based on sentiment data. It makes adjustments such as adding upsell suggestions based on positive emotions or emphasizing reassuring information based on negative emotions. The input is analyzed sentiment data, and the output is the optimized proposal material.

[0619] Step 9:

[0620] User feedback is sent from the smart glasses to the server and used to create proposals for future updates. The input consists of user evaluations and improvement requests, and the output is data that contributes to the continuous improvement of the system.

[0621] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0622] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0623] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0624] [Fourth Embodiment]

[0625] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0626] As shown in Figure 7, the 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.

[0627] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0628] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0629] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0630] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0631] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0632] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive 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 robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0633] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0634] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0635] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0636] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0637] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0638] This invention is a system designed to improve the efficiency of proposal document creation in the sales department, and is implemented by the following means. First, the server extracts past sales negotiation data from the company's negotiation database. At this time, the server uses database queries to obtain necessary sales negotiation information (customer name, industry, negotiation details, etc.). This data forms the basis of a trained model for generating proposal documents.

[0639] Next, the server preprocesses the acquired sales opportunity data. In this process, the server performs data cleaning, such as imputing missing values, removing outliers, and normalizing the data. It also prepares the data into a format that the model can learn from, such as converting categorical data into numerical data.

[0640] The server uses pre-processed data to train a generative AI model. This trained model is designed to present optimal proposals for specific customers, and by referencing past success stories, it can generate more effective sales materials.

[0641] The generated proposal document is sent to the terminal, where the user can review it on the interface and edit it as needed. For example, the user can fine-tune the wording of the proposal document to suit a specific customer or to emphasize the product's features.

[0642] Finally, after the user completes the proposal document, the system collects feedback from the user. The user can point out which parts of the document were effective, and the server can incorporate this feedback data into the next learning cycle to improve the accuracy and quality of the generated AI model. This feedback loop continuously improves the efficiency and effectiveness of proposal document creation as a whole system.

[0643] The following describes the processing flow.

[0644] Step 1:

[0645] The server connects to the negotiation database and extracts past deal data. It issues SQL queries to collect data based on specific criteria (for example, deal history for the past 6 months).

[0646] Step 2:

[0647] The server preprocesses the extracted data. It sorts the data types and imputes any missing values ​​using appropriate methods. Furthermore, it encodes categorical variables into numerical values ​​and converts the data into a format that is easy for the model to handle.

[0648] Step 3:

[0649] The server trains the generative AI model using preprocessed data. Training is performed using an iterative learning algorithm, which learns features based on past successes.

[0650] Step 4:

[0651] The device automatically generates new proposal materials using a pre-trained generative AI model in response to user requests. These generated materials include personalized content for each customer.

[0652] Step 5:

[0653] Users review the proposal documents on their devices. If necessary, they edit the wording of the documents to tailor them to the specific needs of the client.

[0654] Step 6:

[0655] After the user gives final approval to the proposal, they enter feedback. The system collects this feedback and stores it on the server as data that can be used for future model improvements.

[0656] (Example 1)

[0657] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0658] In sales operations, efficiently and quickly creating proposal materials tailored to each customer is challenging, and improving the accuracy of these materials, especially by leveraging past success stories, requires significant time and effort. Furthermore, there is a need for a means to continuously improve the effectiveness of automatically generated proposal materials in actual business negotiations.

[0659] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0660] In this invention, the server includes means for extracting data from past negotiation information, means for preprocessing the extracted data and shaping it as input for an analysis model, and means for providing a trained model for proposal creation based on the preprocessed data. This enables a system that efficiently creates optimal proposals for each customer and includes a feedback process to improve their effectiveness.

[0661] "Past negotiation information" refers to documents and records related to business negotiations and transactions that have taken place in the past during a company's activities.

[0662] "Means of data extraction" refers to methods or devices for selecting and retrieving necessary information from databases or records.

[0663] "Preprocessing" refers to the process of preparing data to be analyzable, and specifically includes imputing missing values, removing outliers, and normalization.

[0664] "Model input for analysis" refers to data formatted in a way that machine learning models can use for learning and inference.

[0665] A "trained model" is a machine learning model that has been trained on existing data and is used to perform a specific task.

[0666] A "proposal" is a document outlining a plan or information submitted to a customer or business partner, and is used for business negotiations and contract signing.

[0667] "Feedback" refers to information provided to evaluate a system or process and to facilitate improvement.

[0668] "Methods for removing outliers" refer to methods or processes for identifying and eliminating inappropriate data points within a dataset.

[0669] "Means of converting categorical data into numerical format" refers to methods of replacing categorical data with numerical formats, enabling computational processing in machine learning.

[0670] This invention consists of a system that streamlines the creation of proposal materials in sales operations. Specifically, various processes are carried out primarily using a server, terminal, and user.

[0671] The server extracts necessary data from the company's negotiation information database. Using database queries, the server extracts important information related to past business negotiations, such as customer names, industries, and negotiation details. This data then forms the basis for proposal generation.

[0672] Next, the server preprocesses the extracted data. This preprocessing includes imputing missing values, removing outliers, and normalizing the data. It also converts categorical data into numerical format, making it suitable for machine learning models to learn from. This preprocessing improves the accuracy of the data.

[0673] The server uses the formatted data to train a generative AI model. The trained model is then used to provide customers with the most suitable proposals. This includes the ability to leverage past success stories to generate sales materials more effectively.

[0674] The generated proposal document is sent to the terminal and can be reviewed by the user through the interface. The user can review the document and edit it to suit a specific customer. Specifically, it is possible to adjust it to the customer's specialized terminology and emphasize the features of specific products.

[0675] For example, when a user creates a sales proposal for a new product, the server generates the initial proposal by retrieving data from past success stories and applying it to a learning model. An example of a prompt message as text might be: "Generate a proposal for a new product. The target market is small and medium-sized enterprises, and cost efficiency should be emphasized." This allows the system to provide the user with more specific and compelling materials, helping subsequent sales negotiations proceed smoothly.

[0676] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0677] Step 1:

[0678] The server queries a company's negotiation information database to extract data. The input database contains past business negotiation information. Specifically, it retrieves information such as customer name, industry, and negotiation details from the database. The output is negotiation data that forms the basis for creating proposal documents.

[0679] Step 2:

[0680] The server preprocesses the extracted sales opportunity data. The raw data obtained in the previous step is used as input. Specifically, it performs operations such as imputing missing values, removing outliers, and normalizing the data, converting categorical data into numerical format. The output is data converted into a format that can be processed by machine learning models.

[0681] Step 3:

[0682] The server trains a generative AI model using pre-processed data. The input is formatted data. Specifically, the model is trained based on past successful sales negotiations. The output is a trained model capable of generating optimal proposals for a particular customer.

[0683] Step 4:

[0684] The server generates proposal documents using a pre-trained model. The input consists of data and requirements related to a specific customer. Specifically, the generating AI model creates proposals tailored to the customer. The output is the automatically generated proposal document.

[0685] Step 5:

[0686] The terminal displays the generated proposal document to the user. The input is the proposal document sent from the server. Specifically, it visualizes the document on the user interface and makes it available for review and editing. The output is the document screen that the user views.

[0687] Step 6:

[0688] The user reviews the proposal document and edits it as needed. The input is the proposal document displayed on the terminal. Specifically, the user modifies the text to suit the customer's characteristics, emphasizing wording and product features. The output is the revised proposal document.

[0689] Step 7:

[0690] Users submit feedback on the materials to the system. The input consists of user ratings and opinions. Specifically, users input comments indicating which parts of the materials were effective. The output is feedback data used by the server for the next learning cycle.

[0691] (Application Example 1)

[0692] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0693] In modern household proposal activities, there is a need to efficiently create optimal proposal materials based on individual conditions and past cases. However, with current methods, the analysis of past cases and the generation of proposal materials are often done manually, requiring a tremendous amount of time and effort, making efficient proposal activities difficult. To solve this problem, there is a need for a system that utilizes household robots and smart devices to support efficiently optimized proposal activities.

[0694] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0695] In this invention, the server includes means for extracting data from a set of past negotiation information, means for preprocessing the extracted information and formatting it as input data for a data analysis model, and means for presenting a trained model for creating proposal materials using the preprocessed data. This enables efficient support for proposal activities within the home and automatic generation of optimal proposal materials based on past cases.

[0696] "Past negotiation information" refers to historical data of past transactions and business negotiations related to proposal activities.

[0697] A "data analysis model" refers to the algorithms and analytical methods used to analyze historical data and generate proposal documents.

[0698] A "pre-trained model" refers to a model that has been developed using machine learning based on past information and is intended for use in generating proposal documents.

[0699] "Household suggestions" refer to suggestions made in relation to activities and decisions within the household, and specifically include things like savings plans and education plans.

[0700] "Information display means" refers to devices or functions that display generated proposal documents so that users can review and modify them.

[0701] "User feedback" refers to opinions and evaluations regarding the suggestion materials that users provide to the system.

[0702] An "optimized proposal document" refers to a document that contains proposals that are appropriately tailored to a specific individual or situation.

[0703] To implement this invention, the system is configured as follows: The server extracts past negotiation information from the database and uses it as the basic data for proposal materials. SQL queries are used to extract the data, and the specific data to be extracted includes transaction history and customer information. This forms a dataset for generating proposal materials.

[0704] Next, the server preprocesses the extracted data and formats it into a suitable format for training the data analysis model. This preprocessing includes data cleaning, normalization, and numerical conversion of categorical data. This process is carried out using Python, utilizing data processing libraries such as Numpy and Pandas.

[0705] Using the formatted data, the server trains machine learning models using TensorFlow or Keras. The trained models are then used to automatically generate personalized recommendations optimized for a specific user or household. These recommendations are intended for home use and may include, for example, savings plans or educational suggestions.

[0706] The terminal displays the generated proposal documents on an interface, allowing the user to review and edit them. The interface is created using HTML / CSS / JavaScript, providing an environment where users can intuitively manipulate the documents.

[0707] Finally, user feedback is collected and used to continuously improve the trained model. Feedback is collected in the form of questionnaires, and the analyzed data is reflected in the next model training.

[0708] One concrete example is the suggestion of energy-saving plans based on household electricity consumption data. The server uses AI to suggest ways to reduce the average daily consumption based on past electricity usage data. These suggestions are customized according to the user's preferences.

[0709] An example of a prompt for a generating AI model is, "Generate optimal energy-saving suggestions based on your household's electricity usage data for the past year."

[0710] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0711] Step 1:

[0712] The server extracts past negotiation information from the database. An SQL query is used as input, and the output is a dataset containing transaction history and customer information. This data forms the basis of the proposal document.

[0713] Step 2:

[0714] The server preprocesses the extracted data. This process uses Python, employing NumPy and Pandas for data cleaning, missing value imputation, and normalization. The input is the dataset, and the output is formatted data usable by the analysis model.

[0715] Step 3:

[0716] The server uses TensorFlow or Keras to train the formatted data. The input is pre-processed data, and the output is a trained generative AI model. This model is used to automatically generate proposal documents.

[0717] Step 4:

[0718] The server uses a pre-trained model to generate optimal recommendation materials tailored to a specific individual. The input consists of the individual's attributes and past usage data, while the output is customized recommendation materials.

[0719] Step 5:

[0720] The terminal displays the generated proposal document on the user interface. The user can review this document and edit its contents as needed. The displayed document is the generated proposal document, and its contents are updated based on user actions.

[0721] Step 6:

[0722] Users provide feedback on the proposed document. This feedback is entered in a questionnaire format and saved on the server as data for the next model training.

[0723] Step 7:

[0724] The server analyzes the collected feedback and uses it to improve the generated AI model. It uses user feedback as input data and obtains a learning model with improved accuracy as output.

[0725] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0726] This invention is a system that not only streamlines the creation of proposal materials in the sales department, but also has the function of recognizing the user's emotional state and optimizing the content of the proposal.

[0727] First, the server extracts past business negotiation data from the company's negotiation database and preprocesses it to prepare it for training. This data is then used to train a generative AI model, which generates proposal materials optimized for specific customers. This process allows for the creation of more persuasive proposal materials by learning patterns from past success stories.

[0728] Once the proposal document is generated, the terminal presents it to the user and provides an interface for reviewing and editing the proposal. At this stage, the user can adjust the document's content and add information tailored to specific customer needs.

[0729] Furthermore, this system incorporates an emotion engine that extracts emotional data from the user's voice and text input. For example, the emotion engine analyzes the emotions of the user based on the tone and content of comments they make while using the proposal document. The server then analyzes this emotional data and optimizes the proposal document, taking into account the user's current mental state.

[0730] For example, if a user is dissatisfied with the content of a proposal document, the emotion engine detects their stress level and adjusts the proposal to emphasize language and examples that are particularly well-received by customers. Conversely, when a positive user response is detected, additional information or suggestions are automatically inserted to maintain that positive tone.

[0731] Ultimately, the feedback users input into the system helps improve the trained model, allowing the entire system to evolve and provide higher-quality proposal materials for future sales meetings. This feedback loop continuously improves the quality of proposals and the efficiency of sales activities.

[0732] The following describes the processing flow.

[0733] Step 1:

[0734] The server extracts historical sales negotiation data from the negotiation database. To do this, it uses SQL queries to retrieve datasets that match specific time periods and criteria. The data includes customer names, industry, deal outcomes, and proposal details.

[0735] Step 2:

[0736] The server preprocesses the extracted data. It performs data cleaning, imputes missing values, and corrects outliers. It also encodes categorical data into numerical data and converts it into a format usable by the model.

[0737] Step 3:

[0738] The server uses pre-processed data to train a generative AI model. This model learns patterns based on past successes, forming the foundation for generating new proposal documents.

[0739] Step 4:

[0740] The device automatically generates new proposal documents using a generation AI model in response to user requests. These documents include customized content tailored to the specific customer's attributes and industry.

[0741] Step 5:

[0742] Users review the generated proposal documents on their devices and edit them as needed. They optimize the documents by adding vocabulary and details tailored to their business needs.

[0743] Step 6:

[0744] The device collects emotional data from the user's voice and input. An emotion engine analyzes this data to estimate the user's emotional state. This data is used to further refine the proposed materials.

[0745] Step 7:

[0746] The server optimizes the content of the proposal document based on sentiment data. For example, if a user expresses dissatisfaction, it automatically makes adjustments such as emphasizing more persuasive examples.

[0747] Step 8:

[0748] Once the user finally approves the proposal, their feedback is collected. The server analyzes this feedback and stores it to use for future model improvements.

[0749] (Example 2)

[0750] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0751] In modern sales activities, there is a need to streamline the creation of proposal materials in order to respond quickly and accurately to the diverse needs of customers. Furthermore, traditional methods make it difficult to optimize proposals while considering the emotional state of the user, making it a challenge to enhance their appeal to customers. Against this backdrop, there is a demand for a system that utilizes user emotional information to automatically generate optimal proposals for each customer.

[0752] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0753] In this invention, the server includes means for extracting information from a past commercial transaction database, means for preprocessing the extracted information and processing it as input for an analysis model, means for learning the preprocessed data and providing a trained model for generating proposal documents, means for automatically generating proposal documents suitable for specific customers, means for analyzing user sentiment information and optimizing the proposal documents, and means for collecting user feedback on the generated proposal documents and using it to improve the model. This makes it possible to effectively and efficiently generate optimal proposal documents tailored to individual customers.

[0754] A "commercial transaction database" is an information system that systematically stores information about past business negotiations and transactions, making it easy to search and extract that information.

[0755] "Preprocessing" refers to the process of preparing data into a format suitable for learning and analysis, and this includes denoising and normalizing the data.

[0756] An "analytical model" is a mathematical model that learns features from data and makes predictions and classifications based on new data.

[0757] A "trained model" is a model that has been trained based on past data and is in a state where it can make inferences on new data.

[0758] A "proposal document" is a document that contains proposals for products and services tailored to the individual needs of each customer, and is used in sales activities.

[0759] "Automatic generation" refers to a function where the system autonomously produces a certain output without user intervention.

[0760] "User emotional information" refers to data that indicates the emotional state of a user, extracted from their voice and text.

[0761] "Optimization" is the process of adjusting conditions and parameters to maximize their effectiveness for a specific purpose.

[0762] "Collecting feedback" is the activity of obtaining opinions and usage results from users to help improve the system.

[0763] The system of this invention operates collaboratively between a server and a terminal to streamline proposal document creation in the commercial sector. The server first extracts historical business negotiation data from the company's internal business transaction database. At this stage, information is retrieved using a database management system such as SQL. Next, the server preprocesses the extracted data to prepare it for use as input to an analysis model. Python libraries such as Pandas and NumPy are used for preprocessing, including data cleaning and normalization.

[0764] Next, the server trains a generative AI model based on this pre-processed data. This process uses deep learning libraries such as TensorFlow and PyTorch. The trained model forms the basis for generating proposal documents that are best suited to a specific customer. As an example of a prompt, the model is given the instruction, "Create a proposal for the latest product for customer A." This automatically generates a persuasive proposal tailored to the user.

[0765] The generated proposal document is presented to the user via a terminal. The terminal provides an interface for the user to review and edit the document. Through this interface, the user can adjust the document through drag-and-drop and direct editing, and add information that meets the customer's needs.

[0766] The system also features an emotion engine to analyze user sentiment. It detects emotions from the voice and text spoken by users while editing a proposed document, and the server optimizes the document based on that sentiment information. For example, if a user comments, "This content needs improvement," the server detects the negative emotion and restructures the document to make it more acceptable.

[0767] Ultimately, users provide feedback to the system, and this new data is used to retrain the model. The server uses this feedback to improve the model's accuracy for the next document generation. This feedback loop allows the system to improve over time, enabling the generation of higher-quality proposal documents.

[0768] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0769] Step 1:

[0770] The server extracts historical sales data from a company's business transaction database. SQL queries are executed as input, and the output includes datasets such as sales history and customer feedback. This data serves as the foundational information necessary for generating proposal documents.

[0771] Step 2:

[0772] The server performs preprocessing on the extracted data. The input is the raw data obtained in step 1, and the output is prepared into a format suitable for input to the analysis model through data cleaning and normalization. Python libraries such as Pandas and NumPy are used to impute missing values ​​and scale the data.

[0773] Step 3:

[0774] The server trains a generative AI model using preprocessed data. Preprocessed data is used as input, and a trained model usable for proposal document generation is obtained as output. TensorFlow or PyTorch is used to train the model by setting appropriate epoch counts and batch sizes.

[0775] Step 4:

[0776] The server uses a trained model to automatically generate proposal documents tailored to specific customers. The input is a prompt (e.g., "Create a proposal for the latest product for customer A"), and the output is a proposal document that meets that request. This document is optimized based on the specific customer's interests.

[0777] Step 5:

[0778] The terminal presents the generated proposal document to the user. The input is the proposal document sent from the server, and the output provides an interface that the user can view and edit. Through the interface, the user can visually review the document and perform drag-and-drop and text editing as needed.

[0779] Step 6:

[0780] The server acquires and analyzes user emotional information. Input includes user voice comments and text input, and output is an analysis result indicating the user's emotional state. An emotion engine is used to analyze tone and word choice to determine the user's mental state.

[0781] Step 7:

[0782] The server optimizes the proposal document based on sentiment information. The input is the sentiment information obtained in step 6 and the edited proposal document, and the output is the optimized proposal document. If a negative user reaction is detected, the wording is adjusted; if a positive reaction is received, the content is improved to maintain that tone.

[0783] Step 8:

[0784] Users provide feedback to the system. Inputs include usage results and opinions on the final proposed document, while output summarizes areas for improvement. This feedback is incorporated into subsequent model training sessions, contributing to overall system quality improvement.

[0785] (Application Example 2)

[0786] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0787] In sales activities, it is crucial to quickly and effectively prepare optimal proposal materials for each customer. However, efficiently and automatically generating proposal materials based on vast amounts of past negotiation data, and further optimizing them in real time according to the customer's emotional state, is difficult. This requires advanced data analysis and user interaction. Conventional systems struggle to integrate and achieve such complex processing.

[0788] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0789] In this invention, the server includes means for extracting information from past negotiation data, means for preprocessing the extracted information and processing it as input to an analysis model, means for learning the preprocessed data and providing a trained model for generating proposal materials, means for automatically generating proposal materials suitable for a specific customer, means for collecting emotional data from the user's visual and voice input using computer vision technology, means for analyzing the emotional data and optimizing the proposal materials, and means for collecting user feedback and using it to improve the trained model. This enables the rapid provision of individually optimized proposal materials for each customer and real-time adjustment of materials based on emotions.

[0790] "Negotiation data" refers to data that includes information about past business negotiations and transactions within a company.

[0791] "Preprocessing" refers to the preparatory work required to appropriately transform raw data into input for an analytical model.

[0792] A "trained model" is an algorithmic model that uses knowledge gained from learning from past data to make predictions or generate data on new data.

[0793] A "proposal document" is a document or presentation created to explain the appeal of a product or service to a customer and encourage them to make a purchase.

[0794] "Computer vision technology" is a technology that allows computers to extract and analyze information from images and videos.

[0795] "Emotional data" refers to information about a person's emotional state, extracted from their voice, facial expressions, and other data.

[0796] "Optimization" is the act of adjusting a system or process to its best possible state according to its purpose.

[0797] "Feedback" refers to information provided as input, such as evaluations and opinions on the system's output, which is used to improve performance.

[0798] Based on this invention, the following embodiments can be considered in order to actually streamline sales activities and build a system that provides optimal proposals to customers.

[0799] The server first extracts data on past negotiations from the database. The extracted data is preprocessed and appropriately adapted as input for the analysis model. The adapted data is used to train the generative AI model, generating a trained model. This automatically generates proposal materials tailored to specific customers.

[0800] As a terminal, store staff use smart glasses during business negotiations. The smart glasses display generated proposal materials, which users use to make proposals to customers. Furthermore, the camera and microphone built into the smart glasses record the customer's facial expressions and voice, and transmit this data to a server in real time. Using computer vision and voice analysis technologies, the server extracts and analyzes the customer's emotional data.

[0801] The server further optimizes the proposal materials using the extracted emotional data. This process ensures that information and proposals are presented in a way that aligns with the customer's emotions and interests. For example, when a customer smiles, relevant product and special offer information is displayed on the smart glasses, facilitating a smoother business negotiation.

[0802] The feedback users provide to the system is collected by the server and used to improve subsequent sales materials. This feedback loop allows the system to continuously evolve and improve the quality of proposals. In this way, optimized proposals for each customer can be delivered quickly, increasing the success rate of sales deals.

[0803] An example of a prompt message might be, "Please display additional information if the customer gives a positive response at the next business meeting."

[0804] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0805] Step 1:

[0806] The server extracts information from a database of past negotiations. This extraction process filters deal history and customer information related to the business domain to select the necessary datasets. The input is the entire deal history in the database, and the output is the selected deal data.

[0807] Step 2:

[0808] The server preprocesses the extracted sales opportunity data and transforms it appropriately as input for the generating AI model. This process includes data normalization, missing value imputation, and feature encoding. The input is selected sales opportunity data, and the output is in a data format that the machine learning model can understand.

[0809] Step 3:

[0810] The server trains a generative AI model using preprocessed data to create a trained model. This model learns from past successful patterns and generates predictions and suggestion materials for new data. The input is the training data, and the output is the trained generative AI model.

[0811] Step 4:

[0812] The server automatically generates proposal materials tailored to specific customers. It uses a generation AI model to construct customized materials based on customer attributes and past behavior. The input is customer information and a trained model, and the output is a specific proposal document.

[0813] Step 5:

[0814] The smart glasses, acting as a terminal, display proposal materials received from the server. Store staff, acting as users, interact with customers based on these materials. The input is the proposal materials, and the output is the staff's presentation to the customer.

[0815] Step 6:

[0816] The smart glasses use a built-in camera and microphone to record the customer's facial expressions and voice in real time and transmit them to a server. The input is the customer's facial expressions and voice information, and the output is the transmission of data to the server.

[0817] Step 7:

[0818] The server analyzes the received customer's facial expressions and voice data using computer vision and voice analysis technologies to extract emotional data. The input is the customer's visual and voice data, and the output is the analyzed emotional data.

[0819] Step 8:

[0820] The server optimizes proposal materials in real time based on sentiment data. It makes adjustments such as adding upsell suggestions based on positive emotions or emphasizing reassuring information based on negative emotions. The input is analyzed sentiment data, and the output is the optimized proposal material.

[0821] Step 9:

[0822] User feedback is sent from the smart glasses to the server and used to create proposals for future updates. The input consists of user evaluations and improvement requests, and the output is data that contributes to the continuous improvement of the system.

[0823] 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 controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0824] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0825] 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 this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0826] Furthermore, the emotion identification model 59, acting 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 a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0827] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0828] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0829] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0830] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0831] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is 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 the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0832] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0833] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0834] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0835] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0836] 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.

[0837] Furthermore, it is not necessary to store the entirety 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 the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0838] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0839] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0840] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0841] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0842] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0843] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[0844] The following is further disclosed regarding the embodiments described above.

[0845] (Claim 1)

[0846] A method for extracting information from past negotiation databases,

[0847] A means of preprocessing the extracted information and using it as input for an analysis model,

[0848] A means for learning from preprocessed data and providing a trained model for generating proposal materials,

[0849] A means of automatically generating proposal materials tailored to specific customers,

[0850] A means of providing an interface that allows the generated proposal document to be reviewed and edited,

[0851] A means of collecting user feedback and using it to improve trained models,

[0852] A system that includes this.

[0853] (Claim 2)

[0854] The system according to claim 1, wherein the proposed materials include a configuration that includes success stories.

[0855] (Claim 3)

[0856] The system according to claim 1, comprising an algorithm that sequentially improves the model based on user feedback.

[0857] "Example 1"

[0858] (Claim 1)

[0859] Methods for extracting data from past negotiation information,

[0860] A means of preprocessing the extracted data and formatting it as input for an analysis model,

[0861] A means of providing a trained model for proposal creation based on pre-processed data,

[0862] A method for automatically generating proposals tailored to specific clients,

[0863] A means of providing an interface that allows the generated proposal to be reviewed and edited,

[0864] A means of collecting feedback from users and using it to improve trained models,

[0865] The above process includes means for removing outliers and converting categorical data into numerical format,

[0866] A system that includes this.

[0867] (Claim 2)

[0868] The system according to claim 1, which has a configuration for including success stories in proposals to business partners.

[0869] (Claim 3)

[0870] The system according to claim 1, comprising an algorithm for progressively improving the model in accordance with user feedback.

[0871] "Application Example 1"

[0872] (Claim 1)

[0873] A method for extracting data from past negotiation information sets,

[0874] A means for preprocessing the extracted information and formatting it as input data for a data analysis model,

[0875] A means of presenting a pre-trained model for creating proposal materials using pre-processed data,

[0876] A means of automatically generating proposal materials optimized for a specific individual,

[0877] A means for displaying information that allows for review and modification of the generated proposal document,

[0878] A means of collecting user feedback and using it to improve trained models,

[0879] A means of generating proposal materials based on past cases to support proposals within the home,

[0880] A system that includes this.

[0881] (Claim 2)

[0882] The system according to claim 1, having a structure in which the proposal materials include success stories and a function to optimize in-house proposals.

[0883] (Claim 3)

[0884] The system according to claim 1, comprising a program for improving a sequential learning model based on user feedback.

[0885] "Example 2 of combining an emotion engine"

[0886] (Claim 1)

[0887] A method for extracting information from past commercial transaction databases,

[0888] A means of preprocessing the extracted information and using it as input for an analysis model,

[0889] A means for learning from preprocessed data and providing a trained model for generating proposed documents,

[0890] A means of automatically generating proposal documents tailored to specific customers,

[0891] Means for providing an interface that allows the generated proposal document to be reviewed and edited,

[0892] A means of analyzing user sentiment information and optimizing the proposed document,

[0893] A means of collecting user feedback on the generated proposal document and using it to improve the model,

[0894] A system that includes this.

[0895] (Claim 2)

[0896] The system according to claim 1, wherein the proposal document has a structure that includes past success stories and optimizes based on user sentiment information.

[0897] (Claim 3)

[0898] The system according to claim 1, comprising an algorithm that sequentially improves the model based on user feedback and sentiment information.

[0899] "Application example 2 when combining with an emotional engine"

[0900] (Claim 1)

[0901] Methods for extracting information from past negotiation data,

[0902] A means of preprocessing the extracted information and using it as input for an analysis model,

[0903] A means for learning from preprocessed data and providing a trained model for generating proposal materials,

[0904] A means of automatically generating proposal materials tailored to specific customers,

[0905] A means of providing information that can be reviewed and edited in the generated proposal document,

[0906] A means for collecting emotional data from a user's visual and audio input using computer vision technology,

[0907] A method for analyzing emotional data and optimizing proposal materials,

[0908] A means of collecting user feedback and using it to improve trained models,

[0909] A system that includes this.

[0910] (Claim 2)

[0911] The system according to claim 1, wherein the proposal document includes success stories and adjusts the content in real time based on the user's emotional state.

[0912] (Claim 3)

[0913] The system according to claim 1, comprising an algorithm that sequentially improves the model based on the user's visual and auditory feedback. [Explanation of Symbols]

[0914] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A method for extracting information from past negotiation databases, A means of preprocessing the extracted information and using it as input for an analysis model, A means for learning from preprocessed data and providing a trained model for generating proposal materials, A means of automatically generating proposal materials tailored to specific customers, A means of providing an interface that allows the generated proposal document to be reviewed and edited, A means of collecting user feedback and using it to improve trained models, A system that includes this.

2. The system according to claim 1, wherein the proposed materials include a configuration that includes success stories.

3. The system according to claim 1, comprising an algorithm that sequentially improves the model based on user feedback.

Images