system

Abstract

Provide a system. 【Solution means】 Means to collect general information of corporate customers, Means to collect past order performance data, Means to preprocess the above general information and order performance data, Means to apply a machine learning model based on the preprocessed data to predict future needs, Means to propose solutions corresponding to the predicted needs, Means to automatically generate a proposal based on the proposed solution, Means to transmit the generated proposal to a terminal, A system including the above.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, the method including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a character of the chatbot, 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 modern corporate sales, it is very important to anticipate customer needs and propose appropriate solutions. However, it is a difficult task to efficiently analyze a vast amount of customer information and past order performance data to make accurate proposals. Conventional methods require a great deal of labor and time, and there are limitations in the accuracy and effectiveness of the proposals. As a result, sales activities become inefficient, potentially leading to a decrease in customer satisfaction and a loss of sales opportunities.

Means for Solving the Problems

[0005] To solve the above problems, this invention provides a means for collecting general information and past order performance data of corporate customers and for preprocessing this data. By applying a machine learning model based on the preprocessed data and using a means to predict future needs, it is possible to propose appropriate solutions. Furthermore, the invention also provides a means for automatically proposing solutions that address the predicted needs and automatically generating a proposal document. This proposal document is customized by selecting the appropriate one from multiple templates and sent to the terminal. As a result, sales representatives can make proposals to customers quickly and effectively, thereby improving the efficiency of sales activities and enhancing customer satisfaction.

[0006] A "corporate customer" refers to a company or organization that conducts specific business activities and purchases goods or services as part of those business activities.

[0007] "General information" refers to basic information about a specific corporate client, including industry, company size, main business activities, and location.

[0008] "Order history data" refers to historical data on past orders and transactions, including the products and services purchased by the customer, the transaction amount, the date and time of the transaction, and the terms and conditions of the transaction.

[0009] "Preprocessing" refers to the process of preparing collected data into a format suitable for data analysis by performing operations such as imputing missing values, data cleansing, filtering, and normalization.

[0010] A "machine learning model" is an algorithm that recognizes patterns in data and makes predictions and classifications based on future data, and is trained using historical data.

[0011] "Needs forecasting" involves combining collected data with machine learning models to predict the products and services that corporate customers will need in the future.

[0012] "Solution proposal" refers to providing appropriate products or services to anticipated needs, and includes the process of shaping this into a concrete proposal document.

[0013] A "proposal" is a document that outlines the proposed content to a corporate client, including details of the proposed product or service, the benefits of implementation, and past performance.

[0014] A "template" is a framework with a basic structure and layout used when automatically generating a proposal, and multiple options are available. [Brief explanation of the drawing]

[0015] [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]It is a sequence diagram showing the processing flow of the data processing system in Embodiment 1. [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Embodiment 2 when combined with an emotion engine. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when combined with an emotion engine.

Mode for Carrying Out the Invention

[0016] Hereinafter, an example of an embodiment of the system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0017] First, the terms used in the following description will be explained.

[0018] In the following embodiments, a labeled processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include CPU (Central Processing Unit), GPU (Graphics Processing Unit), GPGPU (General-Purpose computing on Graphics Processing Units), APU (Accelerated Processing Unit), etc.

[0019] In the following embodiments, a labeled RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0020] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0021] 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).

[0022] 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."

[0023] [First Embodiment]

[0024] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0025] 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.

[0026] 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).

[0027] 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.

[0028] 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.

[0029] 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.

[0030] 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.

[0031] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0032] 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.

[0033] 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.

[0034] 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.

[0035] 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".

[0036] This invention relates to a system that accurately predicts the needs of corporate clients and proposes solutions to address them. It primarily functions through the interaction of a server, terminals, and users to perform processes such as data collection, data preprocessing, data analysis, solution proposal, and automatic generation and transmission of proposal documents.

[0037] Server Processing

[0038] Data collection

[0039] The server first collects general information and past order data for corporate customers from internal databases and external APIs. For example, for a major manufacturing company, it retrieves information such as the number of employees, main product lines, location, and past transaction history.

[0040] Data preprocessing

[0041] Next, the server preprocesses the collected data. It fills in any incomplete parts of the data, corrects erroneous data, and filters out unnecessary information. It also unifies and normalizes data in different formats. For example, it fills in missing purchase amounts in past transaction data with the average value.

[0042] Data Analysis

[0043] The server uses pre-processed data to apply a trained machine learning model to predict future needs. For example, it is predicted that large manufacturing companies are likely to adopt IoT technology in the future.

[0044] Solution proposal

[0045] The server searches the database for appropriate products and services to propose solutions that address anticipated needs. For example, it might suggest our IoT sensor system for a predicted IoT technology deployment.

[0046] Proposal generation

[0047] Once a proposal is approved, the server selects a proposal template and automatically generates the proposal based on it. This proposal includes basic company information, anticipated needs, details of the proposed products or services, benefits of implementation, and implementation track record with similar companies. The final proposal is saved in PDF format and prepared for transmission to the terminal.

[0048] Terminal processing

[0049] User input

[0050] The terminal displays a screen for the user to input company information. The sales representative (user) enters basic information about the company they are scheduled to visit. For example, they might enter information such as "a company in the manufacturing industry with 5,000 employees."

[0051] Data transmission

[0052] The entered information is sent from the terminal to the server.

[0053] Receiving and displaying proposals

[0054] A proposal generated from the server is sent to the terminal and displayed to the user. The proposal details solutions that address anticipated needs.

[0055] Proposal customization

[0056] The device also provides editing capabilities, allowing users to customize proposals. For example, users can fine-tune the layout of the proposal or add comments.

[0057] Sending a proposal

[0058] The customized proposal is sent to the client. It is sent using email or cloud sharing features.

[0059] Specific example

[0060] Example 1: Proposal of IT solutions for the manufacturing industry

[0061] For a manufacturing company called "ABC Manufacturing Co., Ltd.," a user (sales representative) inputs company information into a terminal. The terminal sends this information to a server, which analyzes past order data and market trends to predict that improvements to the manufacturing process are needed. The server selects our IoT sensor system as a proposal and generates a proposal document based on it. The generated proposal document is sent to the terminal, and the user customizes it before sending it to ABC Manufacturing Co., Ltd.

[0062] Example 2: Proposal of environmental protection products for the food industry

[0063] For the food industry company "XYZ Food Co., Ltd.", the user enters company information into a terminal. Based on this information, the server predicts that there is a growing need for food safety and environmental measures. The server selects our eco-friendly packaging materials as a suggestion and generates a proposal document based on it. The proposal document is sent to the terminal, the user customizes it, and finally sends it to XYZ Food Co., Ltd.

[0064] This system enables sales representatives to make appropriate proposals to clients quickly and effectively, resulting in increased efficiency in sales activities and improved customer satisfaction.

[0065] The following describes the processing flow.

[0066] Step 1:

[0067] The server collects general information about corporate customers from internal databases and external APIs. This general information includes industry information, company size, main business activities, and location.

[0068] Step 2:

[0069] The server collects past order data from its internal database. This data includes information such as the products and services purchased by the customer, the transaction amount, the date and time of the transaction, and the terms and conditions of the transaction.

[0070] Step 3:

[0071] The server preprocesses the collected general information and order performance data. This includes imputing missing values, correcting incorrect data, filtering out unnecessary information, and normalizing data in different formats.

[0072] Step 4:

[0073] The server applies machine learning models based on pre-processed data to predict the future needs of corporate customers. For example, it uses historical transaction data and market trends to predict increased demand for IT infrastructure.

[0074] Step 5:

[0075] The server searches the database for solutions that address anticipated needs. For example, it extracts our products and services, such as cloud services and security solutions, as potential suggestions.

[0076] Step 6:

[0077] The server selects the optimal solution from the proposed candidates. This selection takes into account factors such as cost-effectiveness, implementation benefits, and past performance.

[0078] Step 7:

[0079] The server automatically generates a proposal based on the selected solution. The proposal includes general company information, projected needs, details of the proposed products and services, benefits of implementation, and past success stories.

[0080] Step 8:

[0081] The server sends the generated proposal to the terminal. The proposal is saved and sent in PDF or Word format.

[0082] Step 9:

[0083] The user enters information about the company they plan to visit into an input form displayed on the terminal screen. This information includes the company name, industry, and past transaction history.

[0084] Step 10:

[0085] The terminal sends the entered company information to the server. The data is securely transferred using an API endpoint.

[0086] Step 11:

[0087] The terminal receives the proposal sent from the server. It saves the proposal and displays it in a user-friendly format.

[0088] Step 12:

[0089] Users review the received proposals and customize them as needed. For example, they can add comments to the proposal or adjust the layout.

[0090] Step 13:

[0091] The user sends the customized proposal to the client. The proposal is sent via email or cloud sharing.

[0092] (Example 1)

[0093] 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."

[0094] Traditional systems required significant time and effort to predict the needs of corporate clients and propose appropriate solutions. Furthermore, proposal creation and customization were often done manually, leading to inefficiencies and a high likelihood of errors. This hindered the efficiency of sales activities and improved customer satisfaction.

[0095] 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.

[0096] In this invention, the server includes means for collecting general information of corporate customers, means for collecting past order performance data, means for pre-processing the above general information and order performance data, means for applying a machine learning model based on the pre-processed data to predict future needs, means for proposing solutions that address the predicted needs, means for automatically generating a proposal document based on the proposed solutions, means for sending the generated proposal document to a terminal, means for the user to input company information using the terminal, means for sending the input information to the server, means for displaying the proposal document sent from the server on the terminal, means for customizing the proposal document on the terminal, and means for sending the customized proposal document to the client. This enables the generation and transmission of proposal documents that respond quickly and accurately to the requests of corporate customers, thereby improving the efficiency of sales activities and enhancing customer satisfaction.

[0097] A "corporate client" is a client that has a specific legal entity, such as a company or organization.

[0098] "General information" refers to basic information about corporate clients, including, for example, name, address, industry, and number of employees.

[0099] "Order history data" refers to historical information on transactions previously received from corporate clients, including, for example, the date and time of the transaction, the details of the transaction, the transaction amount, and product details.

[0100] "Preprocessing" refers to the process of supplementing incomplete parts of collected data, correcting errors, and filtering out unnecessary information.

[0101] A "machine learning model" is a statistical model that includes algorithms to learn specific patterns and relationships based on data and make future predictions.

[0102] "Needs forecasting" refers to estimating the future needs and requirements of corporate customers using machine learning models.

[0103] "Solution proposal" means presenting appropriate products or services to meet anticipated needs.

[0104] A "proposal" is a document that details a solution and presents it to a corporate client.

[0105] "Automatic generation" refers to the process of automatically creating a document based on a series of data points using a program.

[0106] A "device" refers to a device operated by a user, such as a computer, tablet, or smartphone.

[0107] "Customization" refers to modifying and adjusting the generated proposal to meet the user's needs.

[0108] "Sending" refers to sending data from a server to a terminal, or from a terminal to a client.

[0109] "Client" refers to the recipient of the proposal, and usually refers to the person in charge at a corporate client company.

[0110] This invention relates to a system that accurately predicts the needs of corporate clients and proposes optimal solutions based on those predictions. It primarily functions through the interaction of a server, terminals, and users, via processes such as data collection, data preprocessing, data analysis, solution proposal, and automatic generation and transmission of proposal documents.

[0111] Hardware and software configuration

[0112] Server Configuration

[0113] Hardware: A server is a computer system equipped with a high-performance processor, sufficient memory, and large-capacity storage. It includes hardware for database management systems (e.g., MySQL®) and integration with external APIs.

[0114] Software: Python is used for data processing, MySQL for databases, the requests library for communication with external APIs, and various libraries such as pandas, numpy, scikit-learn, and TENSORFLOW® are used for data analysis. Python-docx is used for document generation, and WeasyPrint is used for PDF conversion.

[0115] Device configuration

[0116] Hardware: A terminal is a device that a user accesses, and includes PCs, tablets, smartphones, etc.

[0117] Software: The user interface uses HTML and JavaScript (registered trademark), data transmission uses the JavaScript fetch API or Axios, and libraries such as PDF.js are used for displaying proposals.

[0118] Process details

[0119] Data collection

[0120] The server collects general information and past order data for corporate customers from an internal database and external APIs. Specifically, it retrieves information such as customer ID, order date, purchase amount, and purchased items from the internal database, and uses external APIs to retrieve additional data such as the company's location and industry classification.

[0121] Data preprocessing

[0122] The collected data is supplemented to fill in incomplete parts, erroneous data is corrected, and unnecessary information is filtered out. Python's pandas and numpy libraries are used to unify and normalize data in different formats. Additionally, sklearn.impute.SimpleImputer is used to impute missing values.

[0123] Data Analysis

[0124] Based on preprocessed data, machine learning models are applied to predict the future needs of corporate customers. The prediction models include scikit-learn's Random Forest and TensorFlow's deep learning models. Future demand is predicted by loading the models and calling the predict method.

[0125] Solution proposal

[0126] The system searches the database for the most suitable products and services to meet anticipated needs. For example, it might select an IoT sensor system as a proposal for "implementing IoT technology." It then executes an SQL query to retrieve product information.

[0127] Proposal generation

[0128] Load the proposal template and insert the anticipated needs and corresponding solutions. Create the proposal using Python-docx, convert it to PDF format using WeasyPrint, and save it. This proposal will then be sent to the terminal.

[0129] User input on the terminal

[0130] Users input basic corporate customer information through the terminal screen, and the terminal sends this information to the server. A JavaScript and HTML interface is used to enable efficient data entry.

[0131] Receiving and displaying proposals

[0132] The generated proposal is sent to the device and displayed to the user. PDF.js is used to display the contents of the proposal.

[0133] Proposal customization

[0134] Users can customize their proposals on their devices using a WYSIWYG editor (such as TinyMCE or CKEditor) to adjust the content and layout of the proposal.

[0135] Sending a proposal

[0136] Customized proposals are sent as email attachments or provided to clients via a generated cloud sharing link. They are sent using SMTP clients or cloud storage APIs (e.g., Google Drive API).

[0137] Specific example

[0138] A concrete example of a prompt is, "Based on transaction data and company information from the past five years, please tell me how to predict future needs and propose appropriate solutions." By inputting this prompt into the AI ​​generation model, effective suggestions can be obtained.

[0139] This system enables us to provide optimal solutions to corporate clients quickly and accurately, resulting in increased efficiency in sales activities and improved customer satisfaction.

[0140] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0141] The program's processing flow will be explained in steps.

[0142] Step 1:

[0143] Data collection

[0144] The server retrieves general information and past order data for corporate customers from its internal database. Furthermore, it calls external APIs (e.g., Google Places API) to obtain additional data such as company location information and industry classification.

[0145] Input fields: Company ID, number of employees, order date, purchase amount, purchased items, etc.

[0146] Output: A set of collected general information and past order performance data.

[0147] Specific operation: The server uses the Python requests library to collect data from an external API and executes SQL queries to retrieve the necessary information from an internal database.

[0148] Step 2:

[0149] Data preprocessing

[0150] The server preprocesses the collected data. Specifically, it uses the pandas library to impute missing data using appropriate methods (e.g., imputing with the mean), corrects incorrect data (e.g., negative purchase amounts), and filters out unnecessary information. It also normalizes data by unifying different formats.

[0151] Input: A set of collected general information and past order performance data.

[0152] Output: Preprocessed, clean dataset.

[0153] Specific operation: The server uses pandas and numpy to perform data cleaning and preprocessing.

[0154] Step 3:

[0155] Data Analysis

[0156] The server uses pre-processed data to train models such as scikit-learn's Random Forest and TensorFlow's deep learning models, predicting the future needs of corporate clients.

[0157] Input: A pre-processed, clean dataset.

[0158] Output: Predicted future needs (e.g., potential for IoT technology implementation).

[0159] Specific operation: The server loads the trained model, makes predictions using the predict method, and saves the results.

[0160] Step 4:

[0161] Solution proposal

[0162] The server searches the database for solutions that address anticipated needs and selects the most suitable products and services.

[0163] Input: Predicted future needs.

[0164] Output: A set of proposed solutions (e.g., IoT sensor system).

[0165] Specific operation: The server uses SQL queries to retrieve information about the best solution from the database.

[0166] Step 5:

[0167] Proposal generation

[0168] The server loads a proposal template, fills in the anticipated needs and corresponding solutions, and automatically generates the proposal. The generated proposal is converted to PDF format.

[0169] Input: A set of proposed solutions.

[0170] Output: Automatically generated proposal (PDF format).

[0171] Specific operation: The server uses Python-docx to create a proposal in DOCX format and converts it to PDF using WeasyPrint.

[0172] Step 6:

[0173] Sending a proposal

[0174] The server sends the generated proposal to the terminal.

[0175] Input: Automated proposal (PDF format).

[0176] Output: Send proposal data to the terminal.

[0177] Specific operation: The server sends the PDF data to the terminal as an HTTP response.

[0178] Step 7:

[0179] User input on the terminal

[0180] The user enters the company information they plan to visit on the terminal's input screen.

[0181] Input: Company information (e.g., company name, number of employees, industry, location).

[0182] Output: Sending corporate information from the terminal to the server.

[0183] Specific operation: The terminal uses HTML forms and JavaScript to build the user interface, validate the input data, and submit it.

[0184] Step 8:

[0185] Display of the proposal

[0186] The terminal displays the proposal sent from the server using a PDF viewer.

[0187] Input: Proposal submitted from the server (PDF format).

[0188] Output: The proposal displayed to the user.

[0189] Specific operation: The terminal uses the PDF.js library to display the contents of the PDF.

[0190] Step 9:

[0191] Proposal customization

[0192] Users customize proposals on their devices. For example, they can edit the layout and content of the proposal.

[0193] Input: The displayed proposal.

[0194] Output: Customized proposal.

[0195] Specific operation: The terminal allows users to customize proposals using WYSIWYG editors such as TinyMCE or CKEditor.

[0196] Step 10:

[0197] Sending a proposal

[0198] Customized proposals are sent to clients via email or cloud sharing.

[0199] Input: Customized proposal.

[0200] Output: Sending the proposal to the client.

[0201] Specific actions: The device will send emails using an SMTP client or share files using the Google Drive API, etc.

[0202] Examples of prompt statements that have been explicitly stated so far

[0203] A concrete example of a prompt is, "Based on transaction data and company information from the past five years, please tell me how to predict future needs and propose appropriate solutions." By inputting this prompt into the AI ​​generation model, effective suggestions can be obtained.

[0204] (Application Example 1)

[0205] 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."

[0206] Accurately predicting the needs of corporate clients and promptly proposing solutions that address those needs is crucial for improving the efficiency of sales activities and enhancing customer satisfaction. However, the systems required to achieve this are complex and necessitate the accurate processing and analysis of large amounts of data. Furthermore, customizing proposals and engaging with clients are also important elements. This invention aims to solve these problems and realize efficient and accurate solution proposals.

[0207] 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.

[0208] In this invention, the server includes means for collecting summary information of corporate customers, means for collecting past transaction history data, means for pre-processing the summary information and transaction history data, means for applying a machine learning model based on the pre-processed data to predict future demand, means for proposing solutions that correspond to the predicted demand, means for automatically generating a proposal document based on the proposed solutions, means for transmitting the generated proposal document to a terminal, means for customizing the proposal document, and means for interacting with corporate customers using robots in the factory. This makes it possible to accurately predict the needs of corporate customers and propose appropriate solutions efficiently and quickly.

[0209] "Corporate clients" refer to customers such as companies and organizations that have commercial or business needs.

[0210] "Summary information" refers to basic attribute information of corporate clients, including information such as industry, number of employees, location, and year of establishment.

[0211] "Transaction history data" refers to data related to past transactions with corporate customers, including information such as purchased items, transaction amount, number of transactions, and transaction date.

[0212] "Preprocessing" refers to a series of processes to prepare collected data for analysis, including imputation of missing values ​​and data normalization.

[0213] A "machine learning model" refers to an algorithm used to make specific predictions or classifications based on data, and is often trained on a training dataset.

[0214] "Demand forecasting" refers to predicting the products and services that corporate customers will need in the future, based on collected data.

[0215] A "solution" refers to a specific product or service proposed to address anticipated needs.

[0216] A "proposal" is a document that details a proposed solution and is submitted to corporate clients.

[0217] A "device" is a device used by a user to operate something, and includes smartphones, tablets, and personal computers.

[0218] "Customization" refers to modifying and adjusting a generated proposal to suit a specific corporate client.

[0219] "Robots in a factory" refers to mechanical devices that operate within a factory and perform designated tasks.

[0220] "Dialogue" refers to the act of robots in a factory communicating with corporate clients, and includes activities such as gathering information and proposing solutions.

[0221] This invention relates to the realization of a system that accurately predicts the needs of corporate clients and proposes solutions that address those needs. The specific operation and components of the system are described below.

[0222] Server Processing

[0223] Data collection

[0224] The server first collects summary information and historical transaction data for corporate customers. This is done using an internal database and an external API. The server retrieves data through API endpoints and uses the Python pandas library to organize it into a data frame.

[0225] Data preprocessing

[0226] The collected data is preprocessed by imputing missing values, normalizing the data, and filtering out unnecessary information. This is done using pandas and the StandardScaler class from scikit-learn.

[0227] Data Analysis

[0228] A machine learning model is applied based on preprocessed data. The predictive model is assumed to be a linear regression model using scikit-learn, which is trained on historical transaction data. This analysis predicts future demand.

[0229] Solution proposal

[0230] Based on the prediction results, appropriate solutions are proposed. For example, if high demand is predicted, the introduction of IoT sensors is suggested.

[0231] Proposal generation

[0232] Based on the proposed solutions, a proposal document is automatically generated. This proposal document includes an overview of the corporate client, projected demand, details of the proposed products and services, and the benefits of implementation. This process is carried out by converting the text data generated by Python into PDF format.

[0233] Sending a proposal

[0234] The generated proposal is sent to the terminal. An email sending library is used for this purpose. The proposal is sent via a cloud platform and is provided in real time.

[0235] Terminal processing

[0236] User input

[0237] The terminal displays a screen for the user to enter basic information about the corporate customer. The user (sales representative) enters the necessary information through this screen.

[0238] Data transmission

[0239] The information entered is sent from the terminal to the server. This allows the data to be collected on the server in real time.

[0240] Receiving and displaying proposals

[0241] Proposals sent from the server are displayed on the terminal. Users can review them and present them to corporate clients.

[0242] Proposal customization

[0243] The proposal document is customizable by the user. For example, the layout of the proposal document can be fine-tuned, or comments can be added.

[0244] Processing by robots in the factory

[0245] Dialogue

[0246] Factory robots interact with corporate clients. Through these interactions, they can collect additional data and offer immediate suggestions. Equipped with advanced sensors and voice recognition systems, the robots can delve deeper into the needs of corporate clients.

[0247] Specific example

[0248] Example 1: Proposal of IT solutions for the manufacturing industry

[0249] For corporate clients in the industrial manufacturing sector, the user inputs basic information into a terminal, and the server collects, preprocesses, and analyzes this information. Based on the analysis results, the introduction of IoT sensors is proposed, and a proposal document is generated. The proposal document is sent to the terminal, and the user customizes it and presents it to the corporate client.

[0250] Examples of prompts for generative AI models

[0251] "Based on the following company information, predict their needs and propose appropriate solutions. Company Information: Industry = Manufacturing, Number of Employees = 3000, Past Transaction History = High, Technology Implementation History = Medium"

[0252] Hardware and software

[0253] Server: Data collection, data preprocessing, data analysis, solution proposal, proposal generation.

[0254] Libraries used: Python, pandas, scikit-learn, requests

[0255] Terminals: User input, data transmission, proposal reception, proposal customization

[0256] Devices: Smartphones, tablets, PCs

[0257] Factory robots: Interacting with corporate clients, collecting data, and proposing solutions.

[0258] Technology: Advanced sensors, voice recognition systems

[0259] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0260] Step 1:

[0261] Data collection

[0262] The server collects summary information and historical transaction history data for corporate customers from API endpoints and internal databases. Specifically, it uses the requests module to retrieve data from external APIs in JSON format and converts it into a DataFrame using the pandas library.

[0263] Input: API endpoint URL, database connection information

[0264] Output: Dataframe of customer information and transaction history

[0265] Step 2:

[0266] Data preprocessing

[0267] The server imputes missing values ​​in the collected data and normalizes the data. Specifically, it uses pandas to impute missing values ​​with the mean and scikit-learn's StandardScaler to standardize the data.

[0268] Input: Dataframe of customer information and transaction history

[0269] Output: Preprocessed data frame

[0270] Step 3:

[0271] Data Analysis

[0272] The server applies a machine learning model based on the preprocessed data to predict future demand. Specifically, a linear regression model from scikit-learn is used, and predictions are made using the trained model.

[0273] Input: Preprocessed data frame

[0274] Output: Result of demand prediction (predicted value)

[0275] Step 4:

[0276] Solution proposal

[0277] The server proposes an appropriate solution based on the prediction results. For example, for a high demand prediction value, it proposes the introduction of IoT sensors. The proposed content is retrieved by searching for appropriate product information from the database.

[0278] Input: Result of demand prediction

[0279] Output: Proposed solution (list of products and services)

[0280] Step 5:

[0281] Generation of proposal

[0282] The server automatically generates a proposal. The proposal includes summary information of corporate customers, predicted demand, and details of the proposed products and services. Since it is generated based on a template, Python-docx or ReportLab libraries are often used.

[0283] Input: Summary information of corporate customers, predicted demand, details of the proposed products and services

[0284] Output: Proposal (in PDF or DOCX format)

[0285] Step 6:

[0286] Submission of Proposal

[0287] The server sends the generated proposal to the terminal. Here, an email sending library (e.g., smtplib) is used to attach the proposal in PDF format and send it to the end user's terminal.

[0288] Input: Proposal (in PDF or DOCX format), end user's email address

[0289] Output: Notification of proposal sending to the end user

[0290] Step 7:

[0291] User Input

[0292] The user inputs the basic information of the corporate customer from the terminal. Specifically, the information is input in the form format and sent to the system.

[0293] Input: Basic information of the corporate customer (e.g., industry, number of employees, location)

[0294] Output: Sending of the input information to the system

[0295] Step 8:

[0296] Receiving and Displaying of the Proposal

[0297] The terminal receives the proposal sent from the server and displays it to the user. Thereby, the user can check the content of the generated proposal.

[0298] Input: Proposal (in PDF or DOCX format)

[0299] Output: Display of the proposal

[0300] Step 9:

[0301] Customization of the Proposal

[0302] The user can customize the proposal generated by the terminal. For example, using a text editor or specific software, it is possible to modify the layout and add comments.

[0303] Input: Proposal (in PDF or DOCX format)

[0304] Output: Customized proposal

[0305] Step 10:

[0306] Interaction with in-plant robots

[0307] The robots in the factory interact with corporate customers to collect additional data. Specifically, using a speech recognition system and sensors, customer feedback is obtained and sent to the server.

[0308] Input: Customer voice instructions, robot sensor data

[0309] Output: Transmission of customer feedback to the server

[0310] Through this series of steps, the needs of corporate customers can be accurately predicted, and appropriate solutions can be efficiently proposed.

[0311] Furthermore, an emotion engine for estimating the user's emotions may be combined. That is, the specific processing unit 290 may estimate the user's emotions using the emotion recognition model 59 and perform specific processing using the user's emotions.

[0312] This invention relates to a system that combines an emotion engine for recognizing the user's emotions in addition to a system for accurately predicting the needs of corporate customers and proposing corresponding solutions. It mainly functions through the interaction of the server, terminal, and user to execute processes such as data collection, data preprocessing, data analysis, emotion recognition, solution proposal, and automatic generation and transmission of proposals.

[0313] Server Processing

[0314] Data collection

[0315] The server first collects general information and past order data for corporate customers from internal databases and external APIs. For example, for a major manufacturing company, it retrieves information such as the number of employees, main product lines, location, and past transaction history.

[0316] Data preprocessing

[0317] Next, the server preprocesses the collected data. It fills in any incomplete parts of the data, corrects erroneous data, and filters out unnecessary information. It also unifies and normalizes data in different formats. For example, it fills in missing purchase amounts in past transaction data with the average value.

[0318] Data Analysis

[0319] The server uses pre-processed data to apply a trained machine learning model to predict future needs. For example, it might predict that a large manufacturing company is likely to implement IT infrastructure in the future.

[0320] emotion recognition

[0321] The server uses an emotion engine to analyze user emotions based on input data. The emotion engine detects emotions from user-entered text, for example, by using natural language processing techniques. For instance, it analyzes user-entered comments and feedback to identify emotions such as "satisfied" or "dissatisfied."

[0322] Solution proposal

[0323] The server searches the database for the optimal solution based on predicted needs and sentiment analysis. For example, it extracts our products and services, such as cloud services and security solutions, as potential suggestions. Taking the results of the sentiment analysis into account, the system is customized, for example, by providing more detailed suggestions if the user is "dissatisfied."

[0324] Proposal generation

[0325] Once a proposal is approved, the server selects a proposal template based on the sentiment analysis results and automatically generates the proposal. This proposal includes basic company information, anticipated needs, details of the proposed products or services, benefits of implementation, and implementation track record with similar companies. The final proposal is saved in PDF format and prepared for transmission to the terminal.

[0326] Terminal processing

[0327] User input

[0328] The terminal displays a screen for the user to input company information. The sales representative (user) enters basic information about the company they are scheduled to visit. For example, they might enter information such as "a company in the manufacturing industry with 5,000 employees."

[0329] Data transmission

[0330] The entered information is sent from the terminal to the server.

[0331] Receiving and displaying proposals

[0332] A proposal generated from the server is sent to the terminal and displayed to the user. The proposal details solutions that address anticipated needs.

[0333] Proposal customization

[0334] The device also provides editing capabilities, allowing users to customize proposals. For example, users can fine-tune the layout of the proposal or add comments.

[0335] Sending a proposal

[0336] The customized proposal is sent to the client. It is sent using email or cloud sharing features.

[0337] Specific example

[0338] Example 1: Proposal of IT solutions for the manufacturing industry

[0339] For a manufacturing company, a user (sales representative) inputs company information into a terminal. The terminal sends this information to a server, which analyzes past order data and market trends to predict that improvements to the manufacturing process are needed. Furthermore, an emotion engine detects a "somewhat disappointed" emotion from the user's input comments. The server selects our IoT sensor system as a proposal and generates a proposal document based on it. The generated proposal document is sent to the terminal, where the user customizes it and then sends it to the company.

[0340] Example 2: Proposal of environmental protection products for the food industry

[0341] For companies in the food industry, users input company information into a terminal. Based on this information, the server predicts that there is a growing need for food safety and environmental measures. Furthermore, an emotion engine detects a "highly interested" sentiment from the user's input comments. The server selects our eco-friendly packaging materials as a suggestion and generates a proposal based on this. The proposal is sent to the terminal, where the user customizes it to increase satisfaction and finally sends it to the company.

[0342] This system enables sales representatives to quickly and effectively make appropriate proposals to clients, resulting in increased efficiency in sales activities, improved customer satisfaction, and more personalized proposals that reflect user emotions.

[0343] The following describes the processing flow.

[0344] Step 1:

[0345] The server collects general information about corporate customers from internal databases and external APIs. This general information includes industry, company size, main business activities, and location.

[0346] Step 2:

[0347] The server collects past order data from its internal database. This data includes information such as the products and services purchased by the customer, the transaction amount, the date and time of the transaction, and the terms and conditions of the transaction.

[0348] Step 3:

[0349] The server preprocesses the collected general information and order performance data. Specifically, it performs tasks such as imputing missing values, correcting incorrect data, filtering out unnecessary information, and normalizing data in different formats.

[0350] Step 4:

[0351] The server applies a pre-trained machine learning model based on pre-processed data to predict future needs. For example, it might predict that a large manufacturing company is likely to implement IT infrastructure in the future.

[0352] Step 5:

[0353] The server searches the database for solutions that address anticipated needs. For example, it extracts our products and services, such as cloud services and security solutions, as potential suggestions.

[0354] Step 6:

[0355] The server analyzes the text data entered by the user using an emotion engine to recognize emotions. For example, it identifies emotions such as "satisfied" or "dissatisfied" from the user's input comments.

[0356] Step 7:

[0357] The server combines emotion recognition results with needs predictions to select the optimal solution. Based on the emotion recognition results, it fine-tunes the content of the proposal and the selection of templates.

[0358] Step 8:

[0359] The server automatically generates a proposal based on the selected solution. The proposal includes company information, anticipated needs, details of the proposed products and services, benefits of implementation, and past success stories.

[0360] Step 9:

[0361] The server sends the generated proposal to the terminal. The proposal is saved and sent in PDF or Word format.

[0362] Step 10:

[0363] The user enters information about the company they plan to visit into an input form displayed on the terminal screen. This information includes the company name, industry, and past transaction history.

[0364] Step 11:

[0365] The terminal sends the entered company information to the server. The data is securely transferred using an API endpoint.

[0366] Step 12:

[0367] The terminal receives the proposal sent from the server and displays it in a format that is easy for the user to understand.

[0368] Step 13:

[0369] Users review the received proposals and customize them as needed. For example, they can add comments to the proposal or adjust the layout.

[0370] Step 14:

[0371] The user sends the customized proposal to the client. The proposal is sent via email or cloud sharing.

[0372] Specific example

[0373] Example 1: Proposal of IT solutions for the manufacturing industry

[0374] Step 1: The server collects general information about major manufacturing customers from a database and external APIs.

[0375] Step 2: The server collects the customer's past order history data.

[0376] Step 3: The server preprocesses the collected data, performing missing value imputation and data normalization.

[0377] Step 4: The server applies a machine learning model based on the pre-processed data to predict that the customer is likely to need IT infrastructure in the future.

[0378] Step 5: The server searches the database for IT infrastructure-related solutions.

[0379] Step 6: The server analyzes the comments entered by the user on the device using an emotion engine and detects the emotion "slightly anxious".

[0380] Step 7: Based on emotion recognition and needs prediction, the server selects a detailed IT infrastructure proposal.

[0381] Step 8: The server automatically generates a proposal and customizes the content to reflect emotions.

[0382] Step 9: Send the completed proposal to your device.

[0383] Step 10: The user enters the company information of the company they are visiting into the terminal.

[0384] Step 11: The terminal sends information to the server.

[0385] Step 12: Receive proposals sent from the server and display them to the user.

[0386] Step 13: The user reviews the proposal and makes customizations as needed.

[0387] Step 14: The user sends the final proposal to the client.

[0388] Example 2: Proposal of environmental protection products for the food industry

[0389] Step 1: The server collects customer information from the food industry.

[0390] Step 2: The server collects past transaction history.

[0391] Step 3: The server performs data imputation and normalization.

[0392] Step 4: The server predicts future needs using a trained machine learning model. It predicts that the need for safety and environmental measures will increase.

[0393] Step 5: Search the database for related products such as eco-friendly packaging materials.

[0394] Step 6: The server detects a "very interested" sentiment from the user's input comments.

[0395] Step 7: Based on the emotion recognition results, select detailed proposals for eco-friendly packaging materials.

[0396] Step 8: Automatically generate a proposal that reflects emotions.

[0397] Step 9: Send the proposal to your device.

[0398] Step 10: The user enters company information for a food industry company.

[0399] Step 11: The terminal sends information to the server.

[0400] Step 12: Receive proposals from the server and display them to the user.

[0401] Step 13: The user customizes the proposal.

[0402] Step 14: Send the final proposal to the client.

[0403] This system allows sales representatives to not only make optimal proposals to clients quickly and effectively, but also to make personalized proposals based on the user's emotions.

[0404] (Example 2)

[0405] 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".

[0406] Traditional proposal systems for corporate clients predicted customer needs based on general information and past order data, but they were insufficient in considering user emotions. As a result, customer satisfaction decreased, and proposals lacked personalization. Furthermore, even in automated proposal generation, the selection of templates was uniform, making it difficult to respond flexibly to reflect user emotions. There is a need to solve these problems and provide more accurate and personalized proposals.

[0407] 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.

[0408] In this invention, the server includes means for collecting general information of corporate customers, means for collecting past order performance data, means for pre-processing the above general information and order performance data, means for applying a machine learning model based on the pre-processed data to predict future needs, means for recognizing emotions based on user input data, means for proposing solutions corresponding to the predicted needs and the results of emotion recognition, means for automatically generating a proposal document based on the proposed solutions, and means for transmitting the generated proposal document to a terminal. This enables personalized proposals that take into account the user's emotions, thereby improving customer satisfaction and streamlining sales activities.

[0409] "Corporate clients" refer to customers other than individuals, such as companies and organizations.

[0410] "General information" refers to basic attribute information about corporate clients, including the number of employees, location, and main business activities.

[0411] "Order performance data" refers to past transaction history and order-related data with corporate clients.

[0412] "Preprocessing" refers to the process of formatting collected data to make it analyzable, supplementing incomplete data, and unifying data in different formats.

[0413] A "machine learning model" refers to an algorithm that learns certain patterns and rules based on data and uses that knowledge to make predictions and classifications on new data.

[0414] "Emotion recognition" refers to the process of identifying and specifying emotions based on user input data.

[0415] A "solution" refers to a specific product or service, or a combination thereof, proposed to address anticipated needs or problems.

[0416] A "proposal" is a document that details the proposed solution, including basic company information, anticipated needs, proposed content, and its benefits.

[0417] A "terminal" refers to a hardware device used by a user to interact with a system, and includes personal computers, tablets, smartphones, and other similar devices.

[0418] Modes for carrying out the invention

[0419] This invention relates to a system that combines a system for accurately predicting the needs of corporate clients and proposing corresponding solutions with an emotion engine that recognizes user emotions. It primarily functions through the interaction of a server, terminal, and user to perform processes such as data collection, data preprocessing, data analysis, emotion recognition, solution proposal, and automatic generation and transmission of proposal documents.

[0420] Server Processing

[0421] The server first collects general information and past order history data for corporate customers from its internal database and external APIs. For example, for a manufacturing company, it retrieves information such as the number of employees, main product lines, location, and past transaction history from the "customer_db" database and the "https: / / api.example.com / company-data" API.

[0422] Next, the server preprocesses the collected data. It uses algorithms (such as Python's Pandas library) to impart incomplete data and fills missing values ​​with the mean. It unifies data in different formats (e.g., date formatting) and filters out unnecessary information.

[0423] Based on the pre-processed data, the server applies a pre-trained machine learning model (such as "future_needs_model.pkl") to predict future needs. For example, it might predict that a certain manufacturing company is likely to implement IT infrastructure in the future.

[0424] Furthermore, the server uses an emotion engine to recognize emotions based on user input data. Using natural language processing (NLP) techniques, it assigns emotion labels such as "satisfied" or "dissatisfied" to the text data provided by the user.

[0425] Next, the server searches the database (e.g., "solutions_db") for the optimal solution based on the predicted needs and sentiment recognition results. For example, products and services such as cloud services and security solutions may be extracted as suggested options. Taking the sentiment analysis results into consideration, more detailed suggestions are made if the user indicates dissatisfaction.

[0426] Once a proposal is approved, the server selects a proposal template (such as "template1.html") based on the sentiment analysis results and automatically generates the proposal. This proposal includes basic company information, predicted needs, details of the proposed products or services, implementation benefits, and implementation track record of similar companies. The final proposal is saved in PDF format (through the "pdfkit" library) and ready to be sent to the terminal.

[0427] Terminal processing

[0428] The terminal displays a screen for the user to input company information. The sales representative (user) enters basic information about the company they are scheduled to visit (e.g., manufacturing industry, 5,000 employees) into an HTML form. The entered information is sent from the terminal to the server in JSON format.

[0429] A proposal generated from the server is sent to the terminal and displayed to the user. The proposal details solutions that address anticipated needs. Users can further customize the proposal using the provided editing functions. For example, they can fine-tune the layout using a WYSIWYG editor or add comments.

[0430] The customized proposal is sent to the client via email or cloud sharing.

[0431] Specific example

[0432] Example 1: Proposal of IT solutions for the manufacturing industry

[0433] For a manufacturing company, a user (sales representative) inputs company information into a terminal. The terminal sends this information to a server, which analyzes past order data and market trends to predict that improvements to the manufacturing process are needed. Furthermore, an emotion engine detects a "somewhat disappointed" emotion from the user's input comments. The server selects our IoT sensor system as a proposal and generates a proposal document based on it. The generated proposal document is sent to the terminal, where the user customizes it and then sends it to the company.

[0434] Example 2: Proposal of environmental protection products for the food industry

[0435] For companies in the food industry, users input company information into a terminal. Based on this information, the server predicts that there is a growing need for food safety and environmental measures. Furthermore, an emotion engine detects a "highly interested" sentiment from the user's input comments. The server selects our eco-friendly packaging materials as a suggestion and generates a proposal based on this. The proposal is sent to the terminal, where the user customizes it to increase satisfaction and finally sends it to the company.

[0436] Examples of prompts for generative AI models

[0437] "Please describe the system's overview, which predicts the needs of corporate clients and proposes appropriate solutions. Please also provide details on the emotion recognition component based on user input data."

[0438] This prompt statement allows you to request specific explanations from the generated AI model.

[0439] This system enables sales representatives to quickly and effectively make appropriate proposals to clients, resulting in increased sales efficiency and improved customer satisfaction. It also allows for more personalized proposals that reflect user emotions.

[0440] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0441] Step 1:

[0442] The server collects general information about corporate customers from an internal database (customer_db) and an external API (https: / / api.example.com / company-data). Specifically, the server uses an API key to access the external API and retrieves data in JSON format. The input is the API endpoint and authentication information, and the output is general information data about corporate customers.

[0443] Step 2:

[0444] The server collects past order data from its internal database. It establishes a database connection and executes SQL queries such as "SELECT FROM orders WHERE customer_id = ?". The input is the customer ID, and the output is past order data.

[0445] Step 3:

[0446] The server preprocesses the collected general information and order history data. Specifically, it uses the Pandas library to impute missing values ​​with the mean, correct inaccuracies in the data, and filter out unnecessary information. The input is raw data, and the output is preprocessed, clean data.

[0447] Step 4:

[0448] The server uses preprocessed data to apply machine learning models and predict future needs. For example, it loads "future_needs_model.pkl" and uses the Scikit-learn library to predict future demand. The input is preprocessed data, and the output is the predicted needs.

[0449] Step 5:

[0450] The server recognizes emotions based on user input data. Using an NLP library (e.g., SpaCy), it identifies emotions from the input text and assigns emotion labels such as "satisfied" or "dissatisfied." The input is user text data, and the output is emotion labels.

[0451] Step 6:

[0452] The server searches for appropriate solutions based on predicted needs and sentiment recognition results. It uses SQL queries to retrieve the best solutions from the database (solutions_db) and extracts them in list format. The input is the predicted needs and sentiment labels, and the output is a list of suggested solutions.

[0453] Step 7:

[0454] The server automatically generates a proposal based on the proposed solution. Using a template engine (such as Jinja), it embeds data into the selected template to generate the final proposal. PDFkit is used to convert it to PDF format. The input is the proposed solution and template, and the output is a PDF file of the proposal.

[0455] Step 8:

[0456] The server sends the generated proposal to the user's terminal. It sends a PDF file as an HTTP response, which the user receives. The input is the generated PDF file, and the output is the transmission to the user's terminal.

[0457] Step 9:

[0458] The terminal displays the proposal to the user. It opens a PDF viewer to display the received PDF file. The input is the proposal PDF file, and the output is the proposal that the user can view.

[0459] Step 10:

[0460] The user customizes the proposal. They use the editing functions provided by the terminal (such as a WYSIWYG editor) to modify text and change the layout. The input is the received proposal and the user's edits, and the output is the customized proposal.

[0461] Step 11:

[0462] The user sends the customized proposal to the client. The proposal is sent via email or cloud sharing. The input is the customized proposal, and the output is the document sent to the client.

[0463] (Application Example 2)

[0464] 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".

[0465] Traditional corporate client solution proposal systems provided proposals based on predicted customer needs, but lacked personalization of proposals and customization of proposals to consider customer emotions. Therefore, it was difficult to improve customer satisfaction and maximize the effectiveness of proposals. In the security services sector in particular, there is a need to provide more accurate proposals that take customer emotions into consideration.

[0466] 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.

[0467] In this invention, the server includes means for collecting general information of corporate customers, means for collecting past order performance data, means for pre-processing the above general information and order performance data, means for applying a machine learning model based on the pre-processed data to predict future needs, means for recognizing emotions from user input data, means for proposing solutions based on the predicted needs and recognized emotions, means for automatically generating a proposal document based on the proposed solutions, and means for transmitting the generated proposal document to a terminal. This enables more personalized proposals that take into account the predicted needs and emotions of corporate customers.

[0468] A "corporate customer" refers to a specific corporation or company that is targeted as a recipient of products or services.

[0469] "General information" refers to basic attribute information of a company, including data such as industry, number of employees, and location.

[0470] "Order history data" refers to information about past transactions and contracts, specifically including historical data such as transaction amount, item, and transaction date.

[0471] "Preprocessing" refers to the process of preparing collected data into an analyzable format, and includes operations such as data cleaning, imputation of missing values, and normalization.

[0472] A "machine learning model" refers to an algorithm or model used to predict future outcomes based on data analysis, and is one that has learned past patterns using training data.

[0473] "Needs forecasting" is the process of predicting the products and services that corporate customers may need in the future, based on collected data and machine learning models.

[0474] "Emotion recognition" refers to the process of analyzing emotions from user input data and expressing the results numerically or in classifications, using technologies such as natural language processing.

[0475] "Solution proposal" refers to the act of selecting the most suitable products and services based on anticipated needs and perceived emotions, and proposing them to corporate clients.

[0476] "Automatic proposal generation" refers to the process of automatically creating a proposal based on a solution proposal, using templates to combine content.

[0477] "Sending to a device" refers to the process of electronically sending an automatically generated proposal to a device, specifically using methods such as email or cloud sharing.

[0478] This invention combines a system that accurately predicts the needs of corporate clients and proposes corresponding solutions with an emotion engine that recognizes user emotions. This maximizes the effectiveness of solution proposals to corporate clients.

[0479] Server Processing

[0480] The server performs various processes using the following methods.

[0481] 1. Data Collection

[0482] The server collects general information and past order history data of corporate customers from internal databases and external APIs. The collected data includes basic company data and transaction history.

[0483] 2. Data preprocessing

[0484] The collected data is cleaned and normalized using Pandas. Missing values ​​are imputed and the data is standardized, preparing it for analysis.

[0485] 3. Data Analysis

[0486] Based on pre-processed data, a pre-trained machine learning model (TensorFlow or PyTorch) is applied to predict future needs.

[0487] 4. Emotion recognition

[0488] The server uses the Google Cloud Natural Language API to analyze emotions from text entered by the user. This allows it to quantify and extract emotions such as "satisfied" or "dissatisfied."

[0489] 5. Solution Proposal

[0490] Based on predicted needs and sentiment analysis results, the system proposes the most suitable products and services from its solution database. This includes customization, such as providing detailed suggestions if the sentiment is "dissatisfied."

[0491] 6. Proposal generation

[0492] Based on the proposed content, a proposal document will be automatically generated. The proposal document will be generated using a selection of multiple templates and will include the proposed content, company information, and implementation benefits. It will be generated and saved in PDF format.

[0493] Terminal processing

[0494] The device interacts with the user using the following means:

[0495] 1. User input

[0496] The terminal provides an interface for entering corporate customer information. The user enters basic information about the company they plan to visit (for example, "a manufacturing company with 2,000 employees and 5 security incidents in the past 3 years").

[0497] 2. Data transmission

[0498] The entered information is sent from the terminal to the server.

[0499] 3. Receiving and displaying proposals

[0500] The proposal generated from the server is sent to the terminal and displayed to the user. The proposal details solutions that address anticipated needs.

[0501] 4. Customizing the proposal

[0502] The device provides editing capabilities that allow users to customize proposals, enabling fine-tuning of the layout and the addition of comments.

[0503] 5. Submission of proposal

[0504] The customized proposal is sent to the corporate client via email or cloud sharing.

[0505] Specific example

[0506] For a manufacturing company, a user inputs information into a terminal, such as "a manufacturing company with 2,000 employees and five security incidents in the past three years." The terminal sends the information to a server, which analyzes the data, recognizes emotions, and proposes the most suitable security measures.

[0507] Example of a prompt

[0508] "I would like to propose IT security measures for a manufacturing company with 2,000 employees that has experienced five security incidents in the past three years."

[0509] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0510] Step 1:

[0511] Users enter general and basic information about their corporate clients into the terminal. This includes industry, number of employees, location, and past security incidents. The entered data is temporarily stored in JSON format.

[0512] Step 2:

[0513] The terminal sends the entered information to the server. Data transmission is performed securely and encrypted using the HTTPS protocol. The server stores the received data in a database.

[0514] Step 3:

[0515] The server collects general information and past order performance data for corporate customers from internal databases and external APIs. Data collection involves obtaining data in JSON format from each API and integrating it into the database.

[0516] Step 4:

[0517] The server preprocesses the collected data using Pandas. This includes imputing missing values, filtering out unnecessary data, and normalizing the data. The preprocessed data is then converted into a format that is easy for machine learning models to handle.

[0518] Step 5:

[0519] The server applies a pre-trained machine learning model (TensorFlow or PyTorch) based on pre-processed data to predict the future needs of corporate customers. Specifically, if it determines that security measures are necessary, it outputs the details of those measures.

[0520] Step 6:

[0521] The server uses the Google Cloud Natural Language API to recognize emotions from user input data. It parses text data and extracts a numerical emotion score. The emotion score is used to customize suggestions.

[0522] Step 7:

[0523] Based on predicted needs and sentiment analysis results, the server suggests the most suitable products and services from its solution database. If the sentiment score indicates "dissatisfaction," the suggestions are adjusted to include more detailed explanations and additional options.

[0524] Step 8:

[0525] The server automatically generates proposals. These proposals, containing details such as the proposed solution, company information, and implementation benefits, are generated in PDF format. The PDFkit library is used for generation, and multiple templates are selected.

[0526] Step 9:

[0527] The server sends the generated proposal to the terminal. The terminal displays the received proposal to the user and provides an editing interface for customization. The user can fine-tune the content of the proposal and add comments.

[0528] Step 10:

[0529] The customized proposals, completed on the terminal, are sent to corporate clients via email or cloud sharing. This ensures that the proposals reach the clients and that effective security measures are proposed.

[0530] 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.

[0531] 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.

[0532] 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.

[0533] [Second Embodiment]

[0534] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0535] 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.

[0536] 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).

[0537] 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.

[0538] 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.

[0539] 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).

[0540] 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.

[0541] 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.

[0542] 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.

[0543] 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.

[0544] 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.

[0545] 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".

[0546] This invention relates to a system that accurately predicts the needs of corporate clients and proposes solutions to address them. It primarily functions through the interaction of a server, terminals, and users to perform processes such as data collection, data preprocessing, data analysis, solution proposal, and automatic generation and transmission of proposal documents.

[0547] Server Processing

[0548] Data collection

[0549] The server first collects general information and past order data for corporate customers from internal databases and external APIs. For example, for a major manufacturing company, it retrieves information such as the number of employees, main product lines, location, and past transaction history.

[0550] Data preprocessing

[0551] Next, the server preprocesses the collected data. It fills in any incomplete parts of the data, corrects erroneous data, and filters out unnecessary information. It also unifies and normalizes data in different formats. For example, it fills in missing purchase amounts in past transaction data with the average value.

[0552] Data Analysis

[0553] The server uses pre-processed data to apply a trained machine learning model to predict future needs. For example, it is predicted that large manufacturing companies are likely to adopt IoT technology in the future.

[0554] Solution proposal

[0555] The server searches the database for appropriate products and services to propose solutions that address anticipated needs. For example, it might suggest our IoT sensor system for a predicted IoT technology deployment.

[0556] Proposal generation

[0557] Once a proposal is approved, the server selects a proposal template and automatically generates the proposal based on it. This proposal includes basic company information, anticipated needs, details of the proposed products or services, benefits of implementation, and implementation track record with similar companies. The final proposal is saved in PDF format and prepared for transmission to the terminal.

[0558] Terminal processing

[0559] User input

[0560] The terminal displays a screen for the user to input company information. The sales representative (user) enters basic information about the company they are scheduled to visit. For example, they might enter information such as "a company in the manufacturing industry with 5,000 employees."

[0561] Data transmission

[0562] The entered information is sent from the terminal to the server.

[0563] Receiving and displaying proposals

[0564] A proposal generated from the server is sent to the terminal and displayed to the user. The proposal details solutions that address anticipated needs.

[0565] Proposal customization

[0566] The device also provides editing capabilities, allowing users to customize proposals. For example, users can fine-tune the layout of the proposal or add comments.

[0567] Sending a proposal

[0568] The customized proposal is sent to the client. It is sent using email or cloud sharing features.

[0569] Specific example

[0570] Example 1: Proposal of IT solutions for the manufacturing industry

[0571] For a manufacturing company called "ABC Manufacturing Co., Ltd.," a user (sales representative) inputs company information into a terminal. The terminal sends this information to a server, which analyzes past order data and market trends to predict that improvements to the manufacturing process are needed. The server selects our IoT sensor system as a proposal and generates a proposal document based on it. The generated proposal document is sent to the terminal, and the user customizes it before sending it to ABC Manufacturing Co., Ltd.

[0572] Example 2: Proposal of environmental protection products for the food industry

[0573] For the food industry company "XYZ Food Co., Ltd.", the user enters company information into a terminal. Based on this information, the server predicts that there is a growing need for food safety and environmental measures. The server selects our eco-friendly packaging materials as a suggestion and generates a proposal document based on it. The proposal document is sent to the terminal, the user customizes it, and finally sends it to XYZ Food Co., Ltd.

[0574] This system enables sales representatives to make appropriate proposals to clients quickly and effectively, resulting in increased efficiency in sales activities and improved customer satisfaction.

[0575] The following describes the processing flow.

[0576] Step 1:

[0577] The server collects general information about corporate customers from internal databases and external APIs. This general information includes industry information, company size, main business activities, and location.

[0578] Step 2:

[0579] The server collects past order data from its internal database. This data includes information such as the products and services purchased by the customer, the transaction amount, the date and time of the transaction, and the terms and conditions of the transaction.

[0580] Step 3:

[0581] The server preprocesses the collected general information and order performance data. This includes imputing missing values, correcting incorrect data, filtering out unnecessary information, and normalizing data in different formats.

[0582] Step 4:

[0583] The server applies machine learning models based on pre-processed data to predict the future needs of corporate customers. For example, it uses historical transaction data and market trends to predict increased demand for IT infrastructure.

[0584] Step 5:

[0585] The server searches the database for solutions that address anticipated needs. For example, it extracts our products and services, such as cloud services and security solutions, as potential suggestions.

[0586] Step 6:

[0587] The server selects the optimal solution from the proposed candidates. This selection takes into account factors such as cost-effectiveness, implementation benefits, and past performance.

[0588] Step 7:

[0589] The server automatically generates a proposal based on the selected solution. The proposal includes general company information, projected needs, details of the proposed products and services, benefits of implementation, and past success stories.

[0590] Step 8:

[0591] The server sends the generated proposal to the terminal. The proposal is saved and sent in PDF or Word format.

[0592] Step 9:

[0593] The user enters information about the company they plan to visit into an input form displayed on the terminal screen. This information includes the company name, industry, and past transaction history.

[0594] Step 10:

[0595] The terminal sends the entered company information to the server. The data is securely transferred using an API endpoint.

[0596] Step 11:

[0597] The terminal receives the proposal sent from the server. It saves the proposal and displays it in a user-friendly format.

[0598] Step 12:

[0599] Users review the received proposals and customize them as needed. For example, they can add comments to the proposal or adjust the layout.

[0600] Step 13:

[0601] The user sends the customized proposal to the client. The proposal is sent via email or cloud sharing.

[0602] (Example 1)

[0603] 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."

[0604] Traditional systems required significant time and effort to predict the needs of corporate clients and propose appropriate solutions. Furthermore, proposal creation and customization were often done manually, leading to inefficiencies and a high likelihood of errors. This hindered the efficiency of sales activities and improved customer satisfaction.

[0605] 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.

[0606] In this invention, the server includes means for collecting general information of corporate customers, means for collecting past order performance data, means for pre-processing the above general information and order performance data, means for applying a machine learning model based on the pre-processed data to predict future needs, means for proposing solutions that address the predicted needs, means for automatically generating a proposal document based on the proposed solutions, means for sending the generated proposal document to a terminal, means for the user to input company information using the terminal, means for sending the input information to the server, means for displaying the proposal document sent from the server on the terminal, means for customizing the proposal document on the terminal, and means for sending the customized proposal document to the client. This enables the generation and transmission of proposal documents that respond quickly and accurately to the requests of corporate customers, thereby improving the efficiency of sales activities and enhancing customer satisfaction.

[0607] A "corporate client" is a client that has a specific legal entity, such as a company or organization.

[0608] "General information" refers to basic information about corporate clients, including, for example, name, address, industry, and number of employees.

[0609] "Order history data" refers to historical information on transactions previously received from corporate clients, including, for example, the date and time of the transaction, the details of the transaction, the transaction amount, and product details.

[0610] "Preprocessing" refers to the process of supplementing incomplete parts of collected data, correcting errors, and filtering out unnecessary information.

[0611] A "machine learning model" is a statistical model that includes algorithms to learn specific patterns and relationships based on data and make future predictions.

[0612] "Needs forecasting" refers to estimating the future needs and requirements of corporate customers using machine learning models.

[0613] "Solution proposal" means presenting appropriate products or services to meet anticipated needs.

[0614] A "proposal" is a document that details a solution and presents it to a corporate client.

[0615] "Automatic generation" refers to the process of automatically creating a document based on a series of data points using a program.

[0616] A "device" refers to a device operated by a user, such as a computer, tablet, or smartphone.

[0617] "Customization" refers to modifying and adjusting the generated proposal to meet the user's needs.

[0618] "Sending" refers to sending data from a server to a terminal, or from a terminal to a client.

[0619] "Client" refers to the recipient of the proposal, and usually refers to the person in charge at a corporate client company.

[0620] This invention relates to a system that accurately predicts the needs of corporate clients and proposes optimal solutions based on those predictions. It primarily functions through the interaction of a server, terminals, and users, via processes such as data collection, data preprocessing, data analysis, solution proposal, and automatic generation and transmission of proposal documents.

[0621] Hardware and software configuration

[0622] Server Configuration

[0623] Hardware: A server is a computer system equipped with a high-performance processor, sufficient memory, and large-capacity storage. It also includes hardware for database management systems (e.g., MySQL) and integration with external APIs.

[0624] Software: Python is used for data processing, MySQL for databases, the requests library for communication with external APIs, and various libraries such as pandas, numpy, scikit-learn, and TensorFlow for data analysis. Python-docx is used for document generation, and WeasyPrint is used for PDF conversion.

[0625] Device configuration

[0626] Hardware: A terminal is a device that a user accesses, and includes PCs, tablets, smartphones, etc.

[0627] Software: The user interface uses HTML and JavaScript, data transmission uses the JavaScript fetch API or Axios, and libraries such as PDF.js are used for displaying proposals.

[0628] Process details

[0629] Data collection

[0630] The server collects general information and past order data for corporate customers from an internal database and external APIs. Specifically, it retrieves information such as customer ID, order date, purchase amount, and purchased items from the internal database, and uses external APIs to retrieve additional data such as the company's location and industry classification.

[0631] Data preprocessing

[0632] The collected data is supplemented to fill in incomplete parts, erroneous data is corrected, and unnecessary information is filtered out. Python's pandas and numpy libraries are used to unify and normalize data in different formats. Additionally, sklearn.impute.SimpleImputer is used to impute missing values.

[0633] Data Analysis

[0634] Based on preprocessed data, machine learning models are applied to predict the future needs of corporate customers. The prediction models include scikit-learn's Random Forest and TensorFlow's deep learning models. Future demand is predicted by loading the models and calling the predict method.

[0635] Solution proposal

[0636] The system searches the database for the most suitable products and services to meet anticipated needs. For example, it might select an IoT sensor system as a proposal for "implementing IoT technology." It then executes an SQL query to retrieve product information.

[0637] Proposal generation

[0638] Load the proposal template and insert the anticipated needs and corresponding solutions. Create the proposal using Python-docx, convert it to PDF format using WeasyPrint, and save it. This proposal will then be sent to the terminal.

[0639] User input on the terminal

[0640] Users input basic corporate customer information through the terminal screen, and the terminal sends this information to the server. A JavaScript and HTML interface is used to enable efficient data entry.

[0641] Receiving and displaying proposals

[0642] The generated proposal is sent to the device and displayed to the user. PDF.js is used to display the contents of the proposal.

[0643] Proposal customization

[0644] Users can customize their proposals on their devices using a WYSIWYG editor (such as TinyMCE or CKEditor) to adjust the content and layout of the proposal.

[0645] Sending a proposal

[0646] Customized proposals are sent as email attachments or provided to clients via a generated cloud sharing link. They are sent using SMTP clients or cloud storage APIs (e.g., Google Drive API).

[0647] Specific example

[0648] A concrete example of a prompt is, "Based on transaction data and company information from the past five years, please tell me how to predict future needs and propose appropriate solutions." By inputting this prompt into the AI ​​generation model, effective suggestions can be obtained.

[0649] This system enables us to provide optimal solutions to corporate clients quickly and accurately, resulting in increased efficiency in sales activities and improved customer satisfaction.

[0650] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0651] The program's processing flow will be explained in steps.

[0652] Step 1:

[0653] Data collection

[0654] The server retrieves general information and past order data for corporate customers from its internal database. Furthermore, it calls external APIs (e.g., Google Places API) to obtain additional data such as company location information and industry classification.

[0655] Input fields: Company ID, number of employees, order date, purchase amount, purchased items, etc.

[0656] Output: A set of collected general information and past order performance data.

[0657] Specific operation: The server uses the Python requests library to collect data from an external API and executes SQL queries to retrieve the necessary information from an internal database.

[0658] Step 2:

[0659] Data preprocessing

[0660] The server preprocesses the collected data. Specifically, it uses the pandas library to impute missing data using appropriate methods (e.g., imputing with the mean), corrects incorrect data (e.g., negative purchase amounts), and filters out unnecessary information. It also normalizes data by unifying different formats.

[0661] Input: A set of collected general information and past order performance data.

[0662] Output: Preprocessed, clean dataset.

[0663] Specific operation: The server uses pandas and numpy to perform data cleaning and preprocessing.

[0664] Step 3:

[0665] Data Analysis

[0666] The server uses pre-processed data to train models such as scikit-learn's Random Forest and TensorFlow's deep learning models, predicting the future needs of corporate clients.

[0667] Input: A pre-processed, clean dataset.

[0668] Output: Predicted future needs (e.g., potential for IoT technology implementation).

[0669] Specific operation: The server loads the trained model, makes predictions using the predict method, and saves the results.

[0670] Step 4:

[0671] Solution proposal

[0672] The server searches the database for solutions that address anticipated needs and selects the most suitable products and services.

[0673] Input: Predicted future needs.

[0674] Output: A set of proposed solutions (e.g., IoT sensor system).

[0675] Specific operation: The server uses SQL queries to retrieve information about the best solution from the database.

[0676] Step 5:

[0677] Proposal generation

[0678] The server loads a proposal template, fills in the anticipated needs and corresponding solutions, and automatically generates the proposal. The generated proposal is converted to PDF format.

[0679] Input: A set of proposed solutions.

[0680] Output: Automatically generated proposal (PDF format).

[0681] Specific operation: The server uses Python-docx to create a proposal in DOCX format and converts it to PDF using WeasyPrint.

[0682] Step 6:

[0683] Sending a proposal

[0684] The server sends the generated proposal to the terminal.

[0685] Input: Automated proposal (PDF format).

[0686] Output: Send proposal data to the terminal.

[0687] Specific operation: The server sends the PDF data to the terminal as an HTTP response.

[0688] Step 7:

[0689] User input on the terminal

[0690] The user enters the company information they plan to visit on the terminal's input screen.

[0691] Input: Company information (e.g., company name, number of employees, industry, location).

[0692] Output: Sending corporate information from the terminal to the server.

[0693] Specific operation: The terminal uses HTML forms and JavaScript to build the user interface, validate the input data, and submit it.

[0694] Step 8:

[0695] Display of the proposal

[0696] The terminal displays the proposal sent from the server using a PDF viewer.

[0697] Input: Proposal submitted from the server (PDF format).

[0698] Output: The proposal displayed to the user.

[0699] Specific operation: The terminal uses the PDF.js library to display the contents of the PDF.

[0700] Step 9:

[0701] Proposal customization

[0702] Users customize proposals on their devices. For example, they can edit the layout and content of the proposal.

[0703] Input: The displayed proposal.

[0704] Output: Customized proposal.

[0705] Specific operation: The terminal allows users to customize proposals using WYSIWYG editors such as TinyMCE or CKEditor.

[0706] Step 10:

[0707] Sending a proposal

[0708] Customized proposals are sent to clients via email or cloud sharing.

[0709] Input: Customized proposal.

[0710] Output: Sending the proposal to the client.

[0711] Specific actions: The device will send emails using an SMTP client or share files using the Google Drive API, etc.

[0712] Examples of prompt statements that have been explicitly stated so far

[0713] A concrete example of a prompt is, "Based on transaction data and company information from the past five years, please tell me how to predict future needs and propose appropriate solutions." By inputting this prompt into the AI ​​generation model, effective suggestions can be obtained.

[0714] (Application Example 1)

[0715] 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."

[0716] Accurately predicting the needs of corporate clients and promptly proposing solutions that address those needs is crucial for improving the efficiency of sales activities and enhancing customer satisfaction. However, the systems required to achieve this are complex and necessitate the accurate processing and analysis of large amounts of data. Furthermore, customizing proposals and engaging with clients are also important elements. This invention aims to solve these problems and realize efficient and accurate solution proposals.

[0717] 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.

[0718] In this invention, the server includes means for collecting summary information of corporate customers, means for collecting past transaction history data, means for pre-processing the summary information and transaction history data, means for applying a machine learning model based on the pre-processed data to predict future demand, means for proposing solutions that correspond to the predicted demand, means for automatically generating a proposal document based on the proposed solutions, means for transmitting the generated proposal document to a terminal, means for customizing the proposal document, and means for interacting with corporate customers using robots in the factory. This makes it possible to accurately predict the needs of corporate customers and propose appropriate solutions efficiently and quickly.

[0719] "Corporate clients" refer to customers such as companies and organizations that have commercial or business needs.

[0720] "Summary information" refers to basic attribute information of corporate clients, including information such as industry, number of employees, location, and year of establishment.

[0721] "Transaction history data" refers to data related to past transactions with corporate customers, including information such as purchased items, transaction amount, number of transactions, and transaction date.

[0722] "Preprocessing" refers to a series of processes to prepare collected data for analysis, including imputation of missing values ​​and data normalization.

[0723] A "machine learning model" refers to an algorithm used to make specific predictions or classifications based on data, and is often trained on a training dataset.

[0724] "Demand forecasting" refers to predicting the products and services that corporate customers will need in the future, based on collected data.

[0725] A "solution" refers to a specific product or service proposed to address anticipated needs.

[0726] A "proposal" is a document that details a proposed solution and is submitted to corporate clients.

[0727] A "device" is a device used by a user to operate something, and includes smartphones, tablets, and personal computers.

[0728] "Customization" refers to modifying and adjusting a generated proposal to suit a specific corporate client.

[0729] "Robots in a factory" refers to mechanical devices that operate within a factory and perform designated tasks.

[0730] "Dialogue" refers to the act of robots in a factory communicating with corporate clients, and includes activities such as gathering information and proposing solutions.

[0731] This invention relates to the realization of a system that accurately predicts the needs of corporate clients and proposes solutions that address those needs. The specific operation and components of the system are described below.

[0732] Server Processing

[0733] Data collection

[0734] The server first collects summary information and historical transaction data for corporate customers. This is done using an internal database and an external API. The server retrieves data through API endpoints and uses the Python pandas library to organize it into a data frame.

[0735] Data preprocessing

[0736] The collected data is preprocessed by imputing missing values, normalizing the data, and filtering out unnecessary information. This is done using pandas and the StandardScaler class from scikit-learn.

[0737] Data Analysis

[0738] A machine learning model is applied based on preprocessed data. The predictive model is assumed to be a linear regression model using scikit-learn, which is trained on historical transaction data. This analysis predicts future demand.

[0739] Solution proposal

[0740] Based on the prediction results, appropriate solutions are proposed. For example, if high demand is predicted, the introduction of IoT sensors is suggested.

[0741] Proposal generation

[0742] Based on the proposed solutions, a proposal document is automatically generated. This proposal document includes an overview of the corporate client, projected demand, details of the proposed products and services, and the benefits of implementation. This process is carried out by converting the text data generated by Python into PDF format.

[0743] Sending a proposal

[0744] The generated proposal is sent to the terminal. An email sending library is used for this purpose. The proposal is sent via a cloud platform and is provided in real time.

[0745] Terminal processing

[0746] User input

[0747] The terminal displays a screen for the user to enter basic information about the corporate customer. The user (sales representative) enters the necessary information through this screen.

[0748] Data transmission

[0749] The information entered is sent from the terminal to the server. This allows the data to be collected on the server in real time.

[0750] Receiving and displaying proposals

[0751] Proposals sent from the server are displayed on the terminal. Users can review them and present them to corporate clients.

[0752] Proposal customization

[0753] The proposal document is customizable by the user. For example, the layout of the proposal document can be fine-tuned, or comments can be added.

[0754] Processing by robots in the factory

[0755] Dialogue

[0756] Factory robots interact with corporate clients. Through these interactions, they can collect additional data and offer immediate suggestions. Equipped with advanced sensors and voice recognition systems, the robots can delve deeper into the needs of corporate clients.

[0757] Specific example

[0758] Example 1: Proposal of IT solutions for the manufacturing industry

[0759] For corporate clients in the industrial manufacturing sector, the user inputs basic information into a terminal, and the server collects, preprocesses, and analyzes this information. Based on the analysis results, the introduction of IoT sensors is proposed, and a proposal document is generated. The proposal document is sent to the terminal, and the user customizes it and presents it to the corporate client.

[0760] Examples of prompts for generative AI models

[0761] "Based on the following company information, predict their needs and propose appropriate solutions. Company Information: Industry = Manufacturing, Number of Employees = 3000, Past Transaction History = High, Technology Implementation History = Medium"

[0762] Hardware and software

[0763] Server: Data collection, data preprocessing, data analysis, solution proposal, proposal generation.

[0764] Libraries used: Python, pandas, scikit-learn, requests

[0765] Terminals: User input, data transmission, proposal reception, proposal customization

[0766] Devices: Smartphones, tablets, PCs

[0767] Factory robots: Interacting with corporate clients, collecting data, and proposing solutions.

[0768] Technology: Advanced sensors, voice recognition systems

[0769] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0770] Step 1:

[0771] Data collection

[0772] The server collects summary information and historical transaction history data for corporate customers from API endpoints and internal databases. Specifically, it uses the requests module to retrieve data from external APIs in JSON format and converts it into a DataFrame using the pandas library.

[0773] Input: API endpoint URL, database connection information

[0774] Output: Dataframe of customer information and transaction history

[0775] Step 2:

[0776] Data preprocessing

[0777] The server imputes missing values ​​in the collected data and normalizes the data. Specifically, it uses pandas to impute missing values ​​with the mean and scikit-learn's StandardScaler to standardize the data.

[0778] Input: Dataframe of customer information and transaction history

[0779] Output: Preprocessed data frame

[0780] Step 3:

[0781] Data Analysis

[0782] The server applies a machine learning model to preprocessed data to predict future demand. Specifically, it uses a linear regression model from scikit-learn and makes predictions using the pre-trained model.

[0783] Input: Preprocessed data frame

[0784] Output: Demand forecast results (predicted values)

[0785] Step 4:

[0786] Solution proposal

[0787] The server proposes appropriate solutions based on the prediction results. For example, for high demand forecasts, it suggests the introduction of IoT sensors. The proposed solutions are obtained by searching for appropriate product information in the database.

[0788] Input: Demand forecast results

[0789] Output: Proposed solutions (list of products and services)

[0790] Step 5:

[0791] Proposal generation

[0792] The server automatically generates proposals. These proposals include an overview of the corporate client, projected demand, and details of the proposed products and services. Because they are generated based on templates, Python-docx and the ReportLab library are often used.

[0793] Input: Corporate customer overview information, projected demand, and details of proposed products and services.

[0794] Output: Proposal (PDF or DOCX format)

[0795] Step 6:

[0796] Sending a proposal

[0797] The server sends the generated proposal to the terminal. Here, it uses an email sending library (e.g., smtplib) to attach the proposal to the end user's terminal and send it.

[0798] Input: Proposal (PDF or DOCX format), end user's email address

[0799] Output: Notification of proposal submission to end users

[0800] Step 7:

[0801] User input

[0802] Users input basic information about corporate customers from their terminals. Specifically, they enter information using a form and submit it to the system.

[0803] Input: Basic information of the corporate customer (e.g., industry, number of employees, location)

[0804] Output: Sending input information to the system

[0805] Step 8:

[0806] Receiving and displaying proposals

[0807] The terminal receives the proposal sent from the server and displays it to the user. This allows the user to review the contents of the generated proposal.

[0808] Input: Proposal (PDF or DOCX format)

[0809] Output: Display of proposal

[0810] Step 9:

[0811] Proposal customization

[0812] Users can customize proposals generated on their devices. For example, they can modify the layout or add comments using a text editor or specific software.

[0813] Input: Proposal (PDF or DOCX format)

[0814] Output: Customized proposal

[0815] Step 10:

[0816] Dialogue between factory robots

[0817] The robots in the factory interact with corporate clients and collect additional data. Specifically, they use voice recognition systems and sensors to obtain customer feedback and send it to a server.

[0818] Input: Customer voice commands, robot sensor data

[0819] Output: Sending customer feedback to the server

[0820] This series of steps allows us to accurately predict the needs of corporate clients and efficiently propose appropriate solutions.

[0821] 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.

[0822] This invention relates to a system that combines a system for accurately predicting the needs of corporate clients and proposing corresponding solutions with an emotion engine that recognizes user emotions. It primarily functions through the interaction of a server, terminal, and user to perform processes such as data collection, data preprocessing, data analysis, emotion recognition, solution proposal, and automatic generation and transmission of proposal documents.

[0823] Server Processing

[0824] Data collection

[0825] The server first collects general information and past order data for corporate customers from internal databases and external APIs. For example, for a major manufacturing company, it retrieves information such as the number of employees, main product lines, location, and past transaction history.

[0826] Data preprocessing

[0827] Next, the server preprocesses the collected data. It fills in any incomplete parts of the data, corrects erroneous data, and filters out unnecessary information. It also unifies and normalizes data in different formats. For example, it fills in missing purchase amounts in past transaction data with the average value.

[0828] Data Analysis

[0829] The server uses pre-processed data to apply a trained machine learning model to predict future needs. For example, it might predict that a large manufacturing company is likely to implement IT infrastructure in the future.

[0830] emotion recognition

[0831] The server uses an emotion engine to analyze user emotions based on input data. The emotion engine detects emotions from user-entered text, for example, by using natural language processing techniques. For instance, it analyzes user-entered comments and feedback to identify emotions such as "satisfied" or "dissatisfied."

[0832] Solution proposal

[0833] The server searches the database for the optimal solution based on predicted needs and sentiment analysis. For example, it extracts our products and services, such as cloud services and security solutions, as potential suggestions. Taking the results of the sentiment analysis into account, the system is customized, for example, by providing more detailed suggestions if the user is "dissatisfied."

[0834] Proposal generation

[0835] Once a proposal is approved, the server selects a proposal template based on the sentiment analysis results and automatically generates the proposal. This proposal includes basic company information, anticipated needs, details of the proposed products or services, benefits of implementation, and implementation track record with similar companies. The final proposal is saved in PDF format and prepared for transmission to the terminal.

[0836] Terminal processing

[0837] User input

[0838] The terminal displays a screen for the user to input company information. The sales representative (user) enters basic information about the company they are scheduled to visit. For example, they might enter information such as "a company in the manufacturing industry with 5,000 employees."

[0839] Data transmission

[0840] The entered information is sent from the terminal to the server.

[0841] Receiving and displaying proposals

[0842] A proposal generated from the server is sent to the terminal and displayed to the user. The proposal details solutions that address anticipated needs.

[0843] Proposal customization

[0844] The device also provides editing capabilities, allowing users to customize proposals. For example, users can fine-tune the layout of the proposal or add comments.

[0845] Sending a proposal

[0846] The customized proposal is sent to the client. It is sent using email or cloud sharing features.

[0847] Specific example

[0848] Example 1: Proposal of IT solutions for the manufacturing industry

[0849] For a manufacturing company, a user (sales representative) inputs company information into a terminal. The terminal sends this information to a server, which analyzes past order data and market trends to predict that improvements to the manufacturing process are needed. Furthermore, an emotion engine detects a "somewhat disappointed" emotion from the user's input comments. The server selects our IoT sensor system as a proposal and generates a proposal document based on it. The generated proposal document is sent to the terminal, where the user customizes it and then sends it to the company.

[0850] Example 2: Proposal of environmental protection products for the food industry

[0851] For companies in the food industry, users input company information into a terminal. Based on this information, the server predicts that there is a growing need for food safety and environmental measures. Furthermore, an emotion engine detects a "highly interested" sentiment from the user's input comments. The server selects our eco-friendly packaging materials as a suggestion and generates a proposal based on this. The proposal is sent to the terminal, where the user customizes it to increase satisfaction and finally sends it to the company.

[0852] This system enables sales representatives to quickly and effectively make appropriate proposals to clients, resulting in increased efficiency in sales activities, improved customer satisfaction, and more personalized proposals that reflect user emotions.

[0853] The following describes the processing flow.

[0854] Step 1:

[0855] The server collects general information about corporate customers from internal databases and external APIs. This general information includes industry, company size, main business activities, and location.

[0856] Step 2:

[0857] The server collects past order data from its internal database. This data includes information such as the products and services purchased by the customer, the transaction amount, the date and time of the transaction, and the terms and conditions of the transaction.

[0858] Step 3:

[0859] The server preprocesses the collected general information and order performance data. Specifically, it performs tasks such as imputing missing values, correcting incorrect data, filtering out unnecessary information, and normalizing data in different formats.

[0860] Step 4:

[0861] The server applies a pre-trained machine learning model based on pre-processed data to predict future needs. For example, it might predict that a large manufacturing company is likely to implement IT infrastructure in the future.

[0862] Step 5:

[0863] The server searches the database for solutions that address anticipated needs. For example, it extracts our products and services, such as cloud services and security solutions, as potential suggestions.

[0864] Step 6:

[0865] The server analyzes the text data entered by the user using an emotion engine to recognize emotions. For example, it identifies emotions such as "satisfied" or "dissatisfied" from the user's input comments.

[0866] Step 7:

[0867] The server combines emotion recognition results with needs predictions to select the optimal solution. Based on the emotion recognition results, it fine-tunes the content of the proposal and the selection of templates.

[0868] Step 8:

[0869] The server automatically generates a proposal based on the selected solution. The proposal includes company information, anticipated needs, details of the proposed products and services, benefits of implementation, and past success stories.

[0870] Step 9:

[0871] The server sends the generated proposal to the terminal. The proposal is saved and sent in PDF or Word format.

[0872] Step 10:

[0873] The user enters information about the company they plan to visit into an input form displayed on the terminal screen. This information includes the company name, industry, and past transaction history.

[0874] Step 11:

[0875] The terminal sends the entered company information to the server. The data is securely transferred using an API endpoint.

[0876] Step 12:

[0877] The terminal receives the proposal sent from the server and displays it in a format that is easy for the user to understand.

[0878] Step 13:

[0879] Users review the received proposals and customize them as needed. For example, they can add comments to the proposal or adjust the layout.

[0880] Step 14:

[0881] The user sends the customized proposal to the client. The proposal is sent via email or cloud sharing.

[0882] Specific example

[0883] Example 1: Proposal of IT solutions for the manufacturing industry

[0884] Step 1: The server collects general information about major manufacturing customers from a database and external APIs.

[0885] Step 2: The server collects the customer's past order history data.

[0886] Step 3: The server preprocesses the collected data, performing missing value imputation and data normalization.

[0887] Step 4: The server applies a machine learning model based on the pre-processed data to predict that the customer is likely to need IT infrastructure in the future.

[0888] Step 5: The server searches the database for IT infrastructure-related solutions.

[0889] Step 6: The server analyzes the comments entered by the user on the device using an emotion engine and detects the emotion "slightly anxious".

[0890] Step 7: Based on emotion recognition and needs prediction, the server selects a detailed IT infrastructure proposal.

[0891] Step 8: The server automatically generates a proposal and customizes the content to reflect emotions.

[0892] Step 9: Send the completed proposal to your device.

[0893] Step 10: The user enters the company information of the company they are visiting into the terminal.

[0894] Step 11: The terminal sends information to the server.

[0895] Step 12: Receive proposals sent from the server and display them to the user.

[0896] Step 13: The user reviews the proposal and makes customizations as needed.

[0897] Step 14: The user sends the final proposal to the client.

[0898] Example 2: Proposal of environmental protection products for the food industry

[0899] Step 1: The server collects customer information from the food industry.

[0900] Step 2: The server collects past transaction history.

[0901] Step 3: The server performs data imputation and normalization.

[0902] Step 4: The server predicts future needs using a trained machine learning model. It predicts that the need for safety and environmental measures will increase.

[0903] Step 5: Search the database for related products such as eco-friendly packaging materials.

[0904] Step 6: The server detects a "very interested" sentiment from the user's input comments.

[0905] Step 7: Based on the emotion recognition results, select detailed proposals for eco-friendly packaging materials.

[0906] Step 8: Automatically generate a proposal that reflects emotions.

[0907] Step 9: Send the proposal to your device.

[0908] Step 10: The user enters company information for a food industry company.

[0909] Step 11: The terminal sends information to the server.

[0910] Step 12: Receive proposals from the server and display them to the user.

[0911] Step 13: The user customizes the proposal.

[0912] Step 14: Send the final proposal to the client.

[0913] This system allows sales representatives to not only make optimal proposals to clients quickly and effectively, but also to make personalized proposals based on the user's emotions.

[0914] (Example 2)

[0915] 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".

[0916] Traditional proposal systems for corporate clients predicted customer needs based on general information and past order data, but they were insufficient in considering user emotions. As a result, customer satisfaction decreased, and proposals lacked personalization. Furthermore, even in automated proposal generation, the selection of templates was uniform, making it difficult to respond flexibly to reflect user emotions. There is a need to solve these problems and provide more accurate and personalized proposals.

[0917] 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.

[0918] In this invention, the server includes means for collecting general information of corporate customers, means for collecting past order performance data, means for pre-processing the above general information and order performance data, means for applying a machine learning model based on the pre-processed data to predict future needs, means for recognizing emotions based on user input data, means for proposing solutions corresponding to the predicted needs and the results of emotion recognition, means for automatically generating a proposal document based on the proposed solutions, and means for transmitting the generated proposal document to a terminal. This enables personalized proposals that take into account the user's emotions, thereby improving customer satisfaction and streamlining sales activities.

[0919] "Corporate clients" refer to customers other than individuals, such as companies and organizations.

[0920] "General information" refers to basic attribute information about corporate clients, including the number of employees, location, and main business activities.

[0921] "Order performance data" refers to past transaction history and order-related data with corporate clients.

[0922] "Preprocessing" refers to the process of formatting collected data to make it analyzable, supplementing incomplete data, and unifying data in different formats.

[0923] A "machine learning model" refers to an algorithm that learns certain patterns and rules based on data and uses that knowledge to make predictions and classifications on new data.

[0924] "Emotion recognition" refers to the process of identifying and specifying emotions based on user input data.

[0925] A "solution" refers to a specific product or service, or a combination thereof, proposed to address anticipated needs or problems.

[0926] A "proposal" is a document that details the proposed solution, including basic company information, anticipated needs, proposed content, and its benefits.

[0927] A "terminal" refers to a hardware device used by a user to interact with a system, and includes personal computers, tablets, smartphones, and other similar devices.

[0928] Modes for carrying out the invention

[0929] This invention relates to a system that combines a system for accurately predicting the needs of corporate clients and proposing corresponding solutions with an emotion engine that recognizes user emotions. It primarily functions through the interaction of a server, terminal, and user to perform processes such as data collection, data preprocessing, data analysis, emotion recognition, solution proposal, and automatic generation and transmission of proposal documents.

[0930] Server Processing

[0931] The server first collects general information and past order history data for corporate customers from its internal database and external APIs. For example, for a manufacturing company, it retrieves information such as the number of employees, main product lines, location, and past transaction history from the "customer_db" database and the "https: / / api.example.com / company-data" API.

[0932] Next, the server preprocesses the collected data. It uses algorithms (such as Python's Pandas library) to impart incomplete data and fills missing values ​​with the mean. It unifies data in different formats (e.g., date formatting) and filters out unnecessary information.

[0933] Based on the pre-processed data, the server applies a pre-trained machine learning model (such as "future_needs_model.pkl") to predict future needs. For example, it might predict that a certain manufacturing company is likely to implement IT infrastructure in the future.

[0934] Furthermore, the server uses an emotion engine to recognize emotions based on user input data. Using natural language processing (NLP) techniques, it assigns emotion labels such as "satisfied" or "dissatisfied" to the text data provided by the user.

[0935] Next, the server searches the database (e.g., "solutions_db") for the optimal solution based on the predicted needs and sentiment recognition results. For example, products and services such as cloud services and security solutions may be extracted as suggested options. Taking the sentiment analysis results into consideration, more detailed suggestions are made if the user indicates dissatisfaction.

[0936] Once a proposal is approved, the server selects a proposal template (such as "template1.html") based on the sentiment analysis results and automatically generates the proposal. This proposal includes basic company information, predicted needs, details of the proposed products or services, implementation benefits, and implementation track record of similar companies. The final proposal is saved in PDF format (through the "pdfkit" library) and ready to be sent to the terminal.

[0937] Terminal processing

[0938] The terminal displays a screen for the user to input company information. The sales representative (user) enters basic information about the company they are scheduled to visit (e.g., manufacturing industry, 5,000 employees) into an HTML form. The entered information is sent from the terminal to the server in JSON format.

[0939] A proposal generated from the server is sent to the terminal and displayed to the user. The proposal details solutions that address anticipated needs. Users can further customize the proposal using the provided editing functions. For example, they can fine-tune the layout using a WYSIWYG editor or add comments.

[0940] The customized proposal is sent to the client via email or cloud sharing.

[0941] Specific example

[0942] Example 1: Proposal of IT solutions for the manufacturing industry

[0943] For a manufacturing company, a user (sales representative) inputs company information into a terminal. The terminal sends this information to a server, which analyzes past order data and market trends to predict that improvements to the manufacturing process are needed. Furthermore, an emotion engine detects a "somewhat disappointed" emotion from the user's input comments. The server selects our IoT sensor system as a proposal and generates a proposal document based on it. The generated proposal document is sent to the terminal, where the user customizes it and then sends it to the company.

[0944] Example 2: Proposal of environmental protection products for the food industry

[0945] For companies in the food industry, users input company information into a terminal. Based on this information, the server predicts that there is a growing need for food safety and environmental measures. Furthermore, an emotion engine detects a "highly interested" sentiment from the user's input comments. The server selects our eco-friendly packaging materials as a suggestion and generates a proposal based on this. The proposal is sent to the terminal, where the user customizes it to increase satisfaction and finally sends it to the company.

[0946] Examples of prompts for generative AI models

[0947] "Please describe the system's overview, which predicts the needs of corporate clients and proposes appropriate solutions. Please also provide details on the emotion recognition component based on user input data."

[0948] This prompt statement allows you to request specific explanations from the generated AI model.

[0949] This system enables sales representatives to quickly and effectively make appropriate proposals to clients, resulting in increased sales efficiency and improved customer satisfaction. It also allows for more personalized proposals that reflect user emotions.

[0950] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0951] Step 1:

[0952] The server collects general information about corporate customers from an internal database (customer_db) and an external API (https: / / api.example.com / company-data). Specifically, the server uses an API key to access the external API and retrieves data in JSON format. The input is the API endpoint and authentication information, and the output is general information data about corporate customers.

[0953] Step 2:

[0954] The server collects past order data from its internal database. It establishes a database connection and executes SQL queries such as "SELECT FROM orders WHERE customer_id = ?". The input is the customer ID, and the output is past order data.

[0955] Step 3:

[0956] The server preprocesses the collected general information and order history data. Specifically, it uses the Pandas library to impute missing values ​​with the mean, correct inaccuracies in the data, and filter out unnecessary information. The input is raw data, and the output is preprocessed, clean data.

[0957] Step 4:

[0958] The server uses preprocessed data to apply machine learning models and predict future needs. For example, it loads "future_needs_model.pkl" and uses the Scikit-learn library to predict future demand. The input is preprocessed data, and the output is the predicted needs.

[0959] Step 5:

[0960] The server recognizes emotions based on user input data. Using an NLP library (e.g., SpaCy), it identifies emotions from the input text and assigns emotion labels such as "satisfied" or "dissatisfied." The input is user text data, and the output is emotion labels.

[0961] Step 6:

[0962] The server searches for appropriate solutions based on predicted needs and sentiment recognition results. It uses SQL queries to retrieve the best solutions from the database (solutions_db) and extracts them in list format. The input is the predicted needs and sentiment labels, and the output is a list of suggested solutions.

[0963] Step 7:

[0964] The server automatically generates a proposal based on the proposed solution. Using a template engine (such as Jinja), it embeds data into the selected template to generate the final proposal. PDFkit is used to convert it to PDF format. The input is the proposed solution and template, and the output is a PDF file of the proposal.

[0965] Step 8:

[0966] The server sends the generated proposal to the user's terminal. It sends a PDF file as an HTTP response, which the user receives. The input is the generated PDF file, and the output is the transmission to the user's terminal.

[0967] Step 9:

[0968] The terminal displays the proposal to the user. It opens a PDF viewer to display the received PDF file. The input is the proposal PDF file, and the output is the proposal that the user can view.

[0969] Step 10:

[0970] The user customizes the proposal. They use the editing functions provided by the terminal (such as a WYSIWYG editor) to modify text and change the layout. The input is the received proposal and the user's edits, and the output is the customized proposal.

[0971] Step 11:

[0972] The user sends the customized proposal to the client. The proposal is sent via email or cloud sharing. The input is the customized proposal, and the output is the document sent to the client.

[0973] (Application Example 2)

[0974] 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."

[0975] Traditional corporate client solution proposal systems provided proposals based on predicted customer needs, but lacked personalization of proposals and customization of proposals to consider customer emotions. Therefore, it was difficult to improve customer satisfaction and maximize the effectiveness of proposals. In the security services sector in particular, there is a need to provide more accurate proposals that take customer emotions into consideration.

[0976] 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.

[0977] In this invention, the server includes means for collecting general information of corporate customers, means for collecting past order performance data, means for pre-processing the above general information and order performance data, means for applying a machine learning model based on the pre-processed data to predict future needs, means for recognizing emotions from user input data, means for proposing solutions based on the predicted needs and recognized emotions, means for automatically generating a proposal document based on the proposed solutions, and means for transmitting the generated proposal document to a terminal. This enables more personalized proposals that take into account the predicted needs and emotions of corporate customers.

[0978] A "corporate customer" refers to a specific corporation or company that is targeted as a recipient of products or services.

[0979] "General information" refers to basic attribute information of a company, including data such as industry, number of employees, and location.

[0980] "Order history data" refers to information about past transactions and contracts, specifically including historical data such as transaction amount, item, and transaction date.

[0981] "Preprocessing" refers to the process of preparing collected data into an analyzable format, and includes operations such as data cleaning, imputation of missing values, and normalization.

[0982] A "machine learning model" refers to an algorithm or model used to predict future outcomes based on data analysis, and is one that has learned past patterns using training data.

[0983] "Needs forecasting" is the process of predicting the products and services that corporate customers may need in the future, based on collected data and machine learning models.

[0984] "Emotion recognition" refers to the process of analyzing emotions from user input data and expressing the results numerically or in classifications, using technologies such as natural language processing.

[0985] "Solution proposal" refers to the act of selecting the most suitable products and services based on anticipated needs and perceived emotions, and proposing them to corporate clients.

[0986] "Automatic proposal generation" refers to the process of automatically creating a proposal based on a solution proposal, using templates to combine content.

[0987] "Sending to a device" refers to the process of electronically sending an automatically generated proposal to a device, specifically using methods such as email or cloud sharing.

[0988] This invention combines a system that accurately predicts the needs of corporate clients and proposes corresponding solutions with an emotion engine that recognizes user emotions. This maximizes the effectiveness of solution proposals to corporate clients.

[0989] Server Processing

[0990] The server performs various processes using the following methods.

[0991] 1. Data Collection

[0992] The server collects general information and past order history data of corporate customers from internal databases and external APIs. The collected data includes basic company data and transaction history.

[0993] 2. Data preprocessing

[0994] The collected data is cleaned and normalized using Pandas. Missing values ​​are imputed and the data is standardized, preparing it for analysis.

[0995] 3. Data Analysis

[0996] Based on pre-processed data, a pre-trained machine learning model (TensorFlow or PyTorch) is applied to predict future needs.

[0997] 4. Emotion recognition

[0998] The server uses the Google Cloud Natural Language API to analyze emotions from text entered by the user. This allows it to quantify and extract emotions such as "satisfied" or "dissatisfied."

[0999] 5. Solution Proposal

[1000] Based on predicted needs and sentiment analysis results, the system proposes the most suitable products and services from its solution database. This includes customization, such as providing detailed suggestions if the sentiment is "dissatisfied."

[1001] 6. Proposal generation

[1002] Based on the proposed content, a proposal document will be automatically generated. The proposal document will be generated using a selection of multiple templates and will include the proposed content, company information, and implementation benefits. It will be generated and saved in PDF format.

[1003] Terminal processing

[1004] The device interacts with the user using the following means:

[1005] 1. User input

[1006] The terminal provides an interface for entering corporate customer information. The user enters basic information about the company they plan to visit (for example, "a manufacturing company with 2,000 employees and 5 security incidents in the past 3 years").

[1007] 2. Data transmission

[1008] The entered information is sent from the terminal to the server.

[1009] 3. Receiving and displaying proposals

[1010] The proposal generated from the server is sent to the terminal and displayed to the user. The proposal details solutions that address anticipated needs.

[1011] 4. Customizing the proposal

[1012] The device provides editing capabilities that allow users to customize proposals, enabling fine-tuning of the layout and the addition of comments.

[1013] 5. Submission of proposal

[1014] The customized proposal is sent to the corporate client via email or cloud sharing.

[1015] Specific example

[1016] For a manufacturing company, a user inputs information into a terminal, such as "a manufacturing company with 2,000 employees and five security incidents in the past three years." The terminal sends the information to a server, which analyzes the data, recognizes emotions, and proposes the most suitable security measures.

[1017] Example of a prompt

[1018] "I would like to propose IT security measures for a manufacturing company with 2,000 employees that has experienced five security incidents in the past three years."

[1019] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1020] Step 1:

[1021] Users enter general and basic information about their corporate clients into the terminal. This includes industry, number of employees, location, and past security incidents. The entered data is temporarily stored in JSON format.

[1022] Step 2:

[1023] The terminal sends the entered information to the server. Data transmission is performed securely and encrypted using the HTTPS protocol. The server stores the received data in a database.

[1024] Step 3:

[1025] The server collects general information and past order performance data for corporate customers from internal databases and external APIs. Data collection involves obtaining data in JSON format from each API and integrating it into the database.

[1026] Step 4:

[1027] The server preprocesses the collected data using Pandas. This includes imputing missing values, filtering out unnecessary data, and normalizing the data. The preprocessed data is then converted into a format that is easy for machine learning models to handle.

[1028] Step 5:

[1029] The server applies a pre-trained machine learning model (TensorFlow or PyTorch) based on pre-processed data to predict the future needs of corporate customers. Specifically, if it determines that security measures are necessary, it outputs the details of those measures.

[1030] Step 6:

[1031] The server uses the Google Cloud Natural Language API to recognize emotions from user input data. It parses text data and extracts a numerical emotion score. The emotion score is used to customize suggestions.

[1032] Step 7:

[1033] Based on predicted needs and sentiment analysis results, the server suggests the most suitable products and services from its solution database. If the sentiment score indicates "dissatisfaction," the suggestions are adjusted to include more detailed explanations and additional options.

[1034] Step 8:

[1035] The server automatically generates proposals. These proposals, containing details such as the proposed solution, company information, and implementation benefits, are generated in PDF format. The PDFkit library is used for generation, and multiple templates are selected.

[1036] Step 9:

[1037] The server sends the generated proposal to the terminal. The terminal displays the received proposal to the user and provides an editing interface for customization. The user can fine-tune the content of the proposal and add comments.

[1038] Step 10:

[1039] The customized proposals, completed on the terminal, are sent to corporate clients via email or cloud sharing. This ensures that the proposals reach the clients and that effective security measures are proposed.

[1040] 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.

[1041] 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.

[1042] 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.

[1043] [Third Embodiment]

[1044] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[1045] 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.

[1046] 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).

[1047] 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.

[1048] 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.

[1049] 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).

[1050] 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.

[1051] 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.

[1052] 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.

[1053] 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.

[1054] 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.

[1055] 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".

[1056] This invention relates to a system that accurately predicts the needs of corporate clients and proposes solutions to address them. It primarily functions through the interaction of a server, terminals, and users to perform processes such as data collection, data preprocessing, data analysis, solution proposal, and automatic generation and transmission of proposal documents.

[1057] Server Processing

[1058] Data collection

[1059] The server first collects general information and past order data for corporate customers from internal databases and external APIs. For example, for a major manufacturing company, it retrieves information such as the number of employees, main product lines, location, and past transaction history.

[1060] Data preprocessing

[1061] Next, the server preprocesses the collected data. It fills in any incomplete parts of the data, corrects erroneous data, and filters out unnecessary information. It also unifies and normalizes data in different formats. For example, it fills in missing purchase amounts in past transaction data with the average value.

[1062] Data Analysis

[1063] The server uses pre-processed data to apply a trained machine learning model to predict future needs. For example, it is predicted that large manufacturing companies are likely to adopt IoT technology in the future.

[1064] Solution proposal

[1065] The server searches the database for appropriate products and services to propose solutions that address anticipated needs. For example, it might suggest our IoT sensor system for a predicted IoT technology deployment.

[1066] Proposal generation

[1067] Once a proposal is approved, the server selects a proposal template and automatically generates the proposal based on it. This proposal includes basic company information, anticipated needs, details of the proposed products or services, benefits of implementation, and implementation track record with similar companies. The final proposal is saved in PDF format and prepared for transmission to the terminal.

[1068] Terminal processing

[1069] User input

[1070] The terminal displays a screen for the user to input company information. The sales representative (user) enters basic information about the company they are scheduled to visit. For example, they might enter information such as "a company in the manufacturing industry with 5,000 employees."

[1071] Data transmission

[1072] The entered information is sent from the terminal to the server.

[1073] Receiving and displaying proposals

[1074] A proposal generated from the server is sent to the terminal and displayed to the user. The proposal details solutions that address anticipated needs.

[1075] Proposal customization

[1076] The device also provides editing capabilities, allowing users to customize proposals. For example, users can fine-tune the layout of the proposal or add comments.

[1077] Sending a proposal

[1078] The customized proposal is sent to the client. It is sent using email or cloud sharing features.

[1079] Specific example

[1080] Example 1: Proposal of IT solutions for the manufacturing industry

[1081] For a manufacturing company called "ABC Manufacturing Co., Ltd.," a user (sales representative) inputs company information into a terminal. The terminal sends this information to a server, which analyzes past order data and market trends to predict that improvements to the manufacturing process are needed. The server selects our IoT sensor system as a proposal and generates a proposal document based on it. The generated proposal document is sent to the terminal, and the user customizes it before sending it to ABC Manufacturing Co., Ltd.

[1082] Example 2: Proposal of environmental protection products for the food industry

[1083] For the food industry company "XYZ Food Co., Ltd.", the user enters company information into a terminal. Based on this information, the server predicts that there is a growing need for food safety and environmental measures. The server selects our eco-friendly packaging materials as a suggestion and generates a proposal document based on it. The proposal document is sent to the terminal, the user customizes it, and finally sends it to XYZ Food Co., Ltd.

[1084] This system enables sales representatives to make appropriate proposals to clients quickly and effectively, resulting in increased efficiency in sales activities and improved customer satisfaction.

[1085] The following describes the processing flow.

[1086] Step 1:

[1087] The server collects general information about corporate customers from internal databases and external APIs. This general information includes industry information, company size, main business activities, and location.

[1088] Step 2:

[1089] The server collects past order data from its internal database. This data includes information such as the products and services purchased by the customer, the transaction amount, the date and time of the transaction, and the terms and conditions of the transaction.

[1090] Step 3:

[1091] The server preprocesses the collected general information and order performance data. This includes imputing missing values, correcting incorrect data, filtering out unnecessary information, and normalizing data in different formats.

[1092] Step 4:

[1093] The server applies machine learning models based on pre-processed data to predict the future needs of corporate customers. For example, it uses historical transaction data and market trends to predict increased demand for IT infrastructure.

[1094] Step 5:

[1095] The server searches the database for solutions that address anticipated needs. For example, it extracts our products and services, such as cloud services and security solutions, as potential suggestions.

[1096] Step 6:

[1097] The server selects the optimal solution from the proposed candidates. This selection takes into account factors such as cost-effectiveness, implementation benefits, and past performance.

[1098] Step 7:

[1099] The server automatically generates a proposal based on the selected solution. The proposal includes general company information, projected needs, details of the proposed products and services, benefits of implementation, and past success stories.

[1100] Step 8:

[1101] The server sends the generated proposal to the terminal. The proposal is saved and sent in PDF or Word format.

[1102] Step 9:

[1103] The user enters information about the company they plan to visit into an input form displayed on the terminal screen. This information includes the company name, industry, and past transaction history.

[1104] Step 10:

[1105] The terminal sends the entered company information to the server. The data is securely transferred using an API endpoint.

[1106] Step 11:

[1107] The terminal receives the proposal sent from the server. It saves the proposal and displays it in a user-friendly format.

[1108] Step 12:

[1109] Users review the received proposals and customize them as needed. For example, they can add comments to the proposal or adjust the layout.

[1110] Step 13:

[1111] The user sends the customized proposal to the client. The proposal is sent via email or cloud sharing.

[1112] (Example 1)

[1113] 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."

[1114] Traditional systems required significant time and effort to predict the needs of corporate clients and propose appropriate solutions. Furthermore, proposal creation and customization were often done manually, leading to inefficiencies and a high likelihood of errors. This hindered the efficiency of sales activities and improved customer satisfaction.

[1115] 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.

[1116] In this invention, the server includes means for collecting general information of corporate customers, means for collecting past order performance data, means for pre-processing the above general information and order performance data, means for applying a machine learning model based on the pre-processed data to predict future needs, means for proposing solutions that address the predicted needs, means for automatically generating a proposal document based on the proposed solutions, means for sending the generated proposal document to a terminal, means for the user to input company information using the terminal, means for sending the input information to the server, means for displaying the proposal document sent from the server on the terminal, means for customizing the proposal document on the terminal, and means for sending the customized proposal document to the client. This enables the generation and transmission of proposal documents that respond quickly and accurately to the requests of corporate customers, thereby improving the efficiency of sales activities and enhancing customer satisfaction.

[1117] A "corporate client" is a client that has a specific legal entity, such as a company or organization.

[1118] "General information" refers to basic information about corporate clients, including, for example, name, address, industry, and number of employees.

[1119] "Order history data" refers to historical information on transactions previously received from corporate clients, including, for example, the date and time of the transaction, the details of the transaction, the transaction amount, and product details.

[1120] "Preprocessing" refers to the process of supplementing incomplete parts of collected data, correcting errors, and filtering out unnecessary information.

[1121] A "machine learning model" is a statistical model that includes algorithms to learn specific patterns and relationships based on data and make future predictions.

[1122] "Needs forecasting" refers to estimating the future needs and requirements of corporate customers using machine learning models.

[1123] "Solution proposal" means presenting appropriate products or services to meet anticipated needs.

[1124] A "proposal" is a document that details a solution and presents it to a corporate client.

[1125] "Automatic generation" refers to the process of automatically creating a document based on a series of data points using a program.

[1126] A "device" refers to a device operated by a user, such as a computer, tablet, or smartphone.

[1127] "Customization" refers to modifying and adjusting the generated proposal to meet the user's needs.

[1128] "Sending" refers to sending data from a server to a terminal, or from a terminal to a client.

[1129] "Client" refers to the recipient of the proposal, and usually refers to the person in charge at a corporate client company.

[1130] This invention relates to a system that accurately predicts the needs of corporate clients and proposes optimal solutions based on those predictions. It primarily functions through the interaction of a server, terminals, and users, via processes such as data collection, data preprocessing, data analysis, solution proposal, and automatic generation and transmission of proposal documents.

[1131] Hardware and software configuration

[1132] Server Configuration

[1133] Hardware: A server is a computer system equipped with a high-performance processor, sufficient memory, and large-capacity storage. It also includes hardware for database management systems (e.g., MySQL) and integration with external APIs.

[1134] Software: Python is used for data processing, MySQL for databases, the requests library for communication with external APIs, and various libraries such as pandas, numpy, scikit-learn, and TensorFlow for data analysis. Python-docx is used for document generation, and WeasyPrint is used for PDF conversion.

[1135] Device configuration

[1136] Hardware: A terminal is a device that a user accesses, and includes PCs, tablets, smartphones, etc.

[1137] Software: The user interface uses HTML and JavaScript, data transmission uses the JavaScript fetch API or Axios, and libraries such as PDF.js are used for displaying proposals.

[1138] Process details

[1139] Data collection

[1140] The server collects general information and past order data for corporate customers from an internal database and external APIs. Specifically, it retrieves information such as customer ID, order date, purchase amount, and purchased items from the internal database, and uses external APIs to retrieve additional data such as the company's location and industry classification.

[1141] Data preprocessing

[1142] The collected data is supplemented to fill in incomplete parts, erroneous data is corrected, and unnecessary information is filtered out. Python's pandas and numpy libraries are used to unify and normalize data in different formats. Additionally, sklearn.impute.SimpleImputer is used to impute missing values.

[1143] Data Analysis

[1144] Based on preprocessed data, machine learning models are applied to predict the future needs of corporate customers. The prediction models include scikit-learn's Random Forest and TensorFlow's deep learning models. Future demand is predicted by loading the models and calling the predict method.

[1145] Solution proposal

[1146] The system searches the database for the most suitable products and services to meet anticipated needs. For example, it might select an IoT sensor system as a proposal for "implementing IoT technology." It then executes an SQL query to retrieve product information.

[1147] Proposal generation

[1148] Load the proposal template and insert the anticipated needs and corresponding solutions. Create the proposal using Python-docx, convert it to PDF format using WeasyPrint, and save it. This proposal will then be sent to the terminal.

[1149] User input on the terminal

[1150] Users input basic corporate customer information through the terminal screen, and the terminal sends this information to the server. A JavaScript and HTML interface is used to enable efficient data entry.

[1151] Receiving and displaying proposals

[1152] The generated proposal is sent to the device and displayed to the user. PDF.js is used to display the contents of the proposal.

[1153] Proposal customization

[1154] Users can customize their proposals on their devices using a WYSIWYG editor (such as TinyMCE or CKEditor) to adjust the content and layout of the proposal.

[1155] Sending a proposal

[1156] Customized proposals are sent as email attachments or provided to clients via a generated cloud sharing link. They are sent using SMTP clients or cloud storage APIs (e.g., Google Drive API).

[1157] Specific example

[1158] A concrete example of a prompt is, "Based on transaction data and company information from the past five years, please tell me how to predict future needs and propose appropriate solutions." By inputting this prompt into the AI ​​generation model, effective suggestions can be obtained.

[1159] This system enables us to provide optimal solutions to corporate clients quickly and accurately, resulting in increased efficiency in sales activities and improved customer satisfaction.

[1160] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1161] The program's processing flow will be explained in steps.

[1162] Step 1:

[1163] Data collection

[1164] The server retrieves general information and past order data for corporate customers from its internal database. Furthermore, it calls external APIs (e.g., Google Places API) to obtain additional data such as company location information and industry classification.

[1165] Input fields: Company ID, number of employees, order date, purchase amount, purchased items, etc.

[1166] Output: A set of collected general information and past order performance data.

[1167] Specific operation: The server uses the Python requests library to collect data from an external API and executes SQL queries to retrieve the necessary information from an internal database.

[1168] Step 2:

[1169] Data preprocessing

[1170] The server preprocesses the collected data. Specifically, it uses the pandas library to impute missing data using appropriate methods (e.g., imputing with the mean), corrects incorrect data (e.g., negative purchase amounts), and filters out unnecessary information. It also normalizes data by unifying different formats.

[1171] Input: A set of collected general information and past order performance data.

[1172] Output: Preprocessed, clean dataset.

[1173] Specific operation: The server uses pandas and numpy to perform data cleaning and preprocessing.

[1174] Step 3:

[1175] Data Analysis

[1176] The server uses pre-processed data to train models such as scikit-learn's Random Forest and TensorFlow's deep learning models, predicting the future needs of corporate clients.

[1177] Input: A pre-processed, clean dataset.

[1178] Output: Predicted future needs (e.g., potential for IoT technology implementation).

[1179] Specific operation: The server loads the trained model, makes predictions using the predict method, and saves the results.

[1180] Step 4:

[1181] Solution proposal

[1182] The server searches the database for solutions that address anticipated needs and selects the most suitable products and services.

[1183] Input: Predicted future needs.

[1184] Output: A set of proposed solutions (e.g., IoT sensor system).

[1185] Specific operation: The server uses SQL queries to retrieve information about the best solution from the database.

[1186] Step 5:

[1187] Proposal generation

[1188] The server loads a proposal template, fills in the anticipated needs and corresponding solutions, and automatically generates the proposal. The generated proposal is converted to PDF format.

[1189] Input: A set of proposed solutions.

[1190] Output: Automatically generated proposal (PDF format).

[1191] Specific operation: The server uses Python-docx to create a proposal in DOCX format and converts it to PDF using WeasyPrint.

[1192] Step 6:

[1193] Sending a proposal

[1194] The server sends the generated proposal to the terminal.

[1195] Input: Automated proposal (PDF format).

[1196] Output: Send proposal data to the terminal.

[1197] Specific operation: The server sends the PDF data to the terminal as an HTTP response.

[1198] Step 7:

[1199] User input on the terminal

[1200] The user enters the company information they plan to visit on the terminal's input screen.

[1201] Input: Company information (e.g., company name, number of employees, industry, location).

[1202] Output: Sending corporate information from the terminal to the server.

[1203] Specific operation: The terminal uses HTML forms and JavaScript to build the user interface, validate the input data, and submit it.

[1204] Step 8:

[1205] Display of the proposal

[1206] The terminal displays the proposal sent from the server using a PDF viewer.

[1207] Input: Proposal submitted from the server (PDF format).

[1208] Output: The proposal displayed to the user.

[1209] Specific operation: The terminal uses the PDF.js library to display the contents of the PDF.

[1210] Step 9:

[1211] Proposal customization

[1212] Users customize proposals on their devices. For example, they can edit the layout and content of the proposal.

[1213] Input: The displayed proposal.

[1214] Output: Customized proposal.

[1215] Specific operation: The terminal allows users to customize proposals using WYSIWYG editors such as TinyMCE or CKEditor.

[1216] Step 10:

[1217] Sending a proposal

[1218] Customized proposals are sent to clients via email or cloud sharing.

[1219] Input: Customized proposal.

[1220] Output: Sending the proposal to the client.

[1221] Specific actions: The device will send emails using an SMTP client or share files using the Google Drive API, etc.

[1222] Examples of prompt statements that have been explicitly stated so far

[1223] A concrete example of a prompt is, "Based on transaction data and company information from the past five years, please tell me how to predict future needs and propose appropriate solutions." By inputting this prompt into the AI ​​generation model, effective suggestions can be obtained.

[1224] (Application Example 1)

[1225] 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."

[1226] Accurately predicting the needs of corporate clients and promptly proposing solutions that address those needs is crucial for improving the efficiency of sales activities and enhancing customer satisfaction. However, the systems required to achieve this are complex and necessitate the accurate processing and analysis of large amounts of data. Furthermore, customizing proposals and engaging with clients are also important elements. This invention aims to solve these problems and realize efficient and accurate solution proposals.

[1227] 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.

[1228] In this invention, the server includes means for collecting summary information of corporate customers, means for collecting past transaction history data, means for pre-processing the summary information and transaction history data, means for applying a machine learning model based on the pre-processed data to predict future demand, means for proposing solutions that correspond to the predicted demand, means for automatically generating a proposal document based on the proposed solutions, means for transmitting the generated proposal document to a terminal, means for customizing the proposal document, and means for interacting with corporate customers using robots in the factory. This makes it possible to accurately predict the needs of corporate customers and propose appropriate solutions efficiently and quickly.

[1229] "Corporate clients" refer to customers such as companies and organizations that have commercial or business needs.

[1230] "Summary information" refers to basic attribute information of corporate clients, including information such as industry, number of employees, location, and year of establishment.

[1231] "Transaction history data" refers to data related to past transactions with corporate customers, including information such as purchased items, transaction amount, number of transactions, and transaction date.

[1232] "Preprocessing" refers to a series of processes to prepare collected data for analysis, including imputation of missing values ​​and data normalization.

[1233] A "machine learning model" refers to an algorithm used to make specific predictions or classifications based on data, and is often trained on a training dataset.

[1234] "Demand forecasting" refers to predicting the products and services that corporate customers will need in the future, based on collected data.

[1235] A "solution" refers to a specific product or service proposed to address anticipated needs.

[1236] A "proposal" is a document that details a proposed solution and is submitted to corporate clients.

[1237] A "device" is a device used by a user to operate something, and includes smartphones, tablets, and personal computers.

[1238] "Customization" refers to modifying and adjusting a generated proposal to suit a specific corporate client.

[1239] "Robots in a factory" refers to mechanical devices that operate within a factory and perform designated tasks.

[1240] "Dialogue" refers to the act of robots in a factory communicating with corporate clients, and includes activities such as gathering information and proposing solutions.

[1241] This invention relates to the realization of a system that accurately predicts the needs of corporate clients and proposes solutions that address those needs. The specific operation and components of the system are described below.

[1242] Server Processing

[1243] Data collection

[1244] The server first collects summary information and historical transaction data for corporate customers. This is done using an internal database and an external API. The server retrieves data through API endpoints and uses the Python pandas library to organize it into a data frame.

[1245] Data preprocessing

[1246] The collected data is preprocessed by imputing missing values, normalizing the data, and filtering out unnecessary information. This is done using pandas and the StandardScaler class from scikit-learn.

[1247] Data Analysis

[1248] A machine learning model is applied based on preprocessed data. The predictive model is assumed to be a linear regression model using scikit-learn, which is trained on historical transaction data. This analysis predicts future demand.

[1249] Solution proposal

[1250] Based on the prediction results, appropriate solutions are proposed. For example, if high demand is predicted, the introduction of IoT sensors is suggested.

[1251] Proposal generation

[1252] Based on the proposed solutions, a proposal document is automatically generated. This proposal document includes an overview of the corporate client, projected demand, details of the proposed products and services, and the benefits of implementation. This process is carried out by converting the text data generated by Python into PDF format.

[1253] Sending a proposal

[1254] The generated proposal is sent to the terminal. An email sending library is used for this purpose. The proposal is sent via a cloud platform and is provided in real time.

[1255] Terminal processing

[1256] User input

[1257] The terminal displays a screen for the user to enter basic information about the corporate customer. The user (sales representative) enters the necessary information through this screen.

[1258] Data transmission

[1259] The information entered is sent from the terminal to the server. This allows the data to be collected on the server in real time.

[1260] Receiving and displaying proposals

[1261] Proposals sent from the server are displayed on the terminal. Users can review them and present them to corporate clients.

[1262] Proposal customization

[1263] The proposal document is customizable by the user. For example, the layout of the proposal document can be fine-tuned, or comments can be added.

[1264] Processing by robots in the factory

[1265] Dialogue

[1266] Factory robots interact with corporate clients. Through these interactions, they can collect additional data and offer immediate suggestions. Equipped with advanced sensors and voice recognition systems, the robots can delve deeper into the needs of corporate clients.

[1267] Specific example

[1268] Example 1: Proposal of IT solutions for the manufacturing industry

[1269] For corporate clients in the industrial manufacturing sector, the user inputs basic information into a terminal, and the server collects, preprocesses, and analyzes this information. Based on the analysis results, the introduction of IoT sensors is proposed, and a proposal document is generated. The proposal document is sent to the terminal, and the user customizes it and presents it to the corporate client.

[1270] Examples of prompts for generative AI models

[1271] "Based on the following company information, predict their needs and propose appropriate solutions. Company Information: Industry = Manufacturing, Number of Employees = 3000, Past Transaction History = High, Technology Implementation History = Medium"

[1272] Hardware and software

[1273] Server: Data collection, data preprocessing, data analysis, solution proposal, proposal generation.

[1274] Libraries used: Python, pandas, scikit-learn, requests

[1275] Terminals: User input, data transmission, proposal reception, proposal customization

[1276] Devices: Smartphones, tablets, PCs

[1277] Factory robots: Interacting with corporate clients, collecting data, and proposing solutions.

[1278] Technology: Advanced sensors, voice recognition systems

[1279] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1280] Step 1:

[1281] Data collection

[1282] The server collects summary information and historical transaction history data for corporate customers from API endpoints and internal databases. Specifically, it uses the requests module to retrieve data from external APIs in JSON format and converts it into a DataFrame using the pandas library.

[1283] Input: API endpoint URL, database connection information

[1284] Output: Dataframe of customer information and transaction history

[1285] Step 2:

[1286] Data preprocessing

[1287] The server imputes missing values ​​in the collected data and normalizes the data. Specifically, it uses pandas to impute missing values ​​with the mean and scikit-learn's StandardScaler to standardize the data.

[1288] Input: Dataframe of customer information and transaction history

[1289] Output: Preprocessed data frame

[1290] Step 3:

[1291] Data Analysis

[1292] The server applies a machine learning model to preprocessed data to predict future demand. Specifically, it uses a linear regression model from scikit-learn and makes predictions using the pre-trained model.

[1293] Input: Preprocessed data frame

[1294] Output: Demand forecast results (predicted values)

[1295] Step 4:

[1296] Solution proposal

[1297] The server proposes appropriate solutions based on the prediction results. For example, for high demand forecasts, it suggests the introduction of IoT sensors. The proposed solutions are obtained by searching for appropriate product information in the database.

[1298] Input: Demand forecast results

[1299] Output: Proposed solutions (list of products and services)

[1300] Step 5:

[1301] Proposal generation

[1302] The server automatically generates proposals. These proposals include an overview of the corporate client, projected demand, and details of the proposed products and services. Because they are generated based on templates, Python-docx and the ReportLab library are often used.

[1303] Input: Corporate customer overview information, projected demand, and details of proposed products and services.

[1304] Output: Proposal (PDF or DOCX format)

[1305] Step 6:

[1306] Sending a proposal

[1307] The server sends the generated proposal to the terminal. Here, it uses an email sending library (e.g., smtplib) to attach the proposal to the end user's terminal and send it.

[1308] Input: Proposal (PDF or DOCX format), end user's email address

[1309] Output: Notification of proposal submission to end users

[1310] Step 7:

[1311] User input

[1312] Users input basic information about corporate customers from their terminals. Specifically, they enter information using a form and submit it to the system.

[1313] Input: Basic information of the corporate customer (e.g., industry, number of employees, location)

[1314] Output: Sending input information to the system

[1315] Step 8:

[1316] Receiving and displaying proposals

[1317] The terminal receives the proposal sent from the server and displays it to the user. This allows the user to review the contents of the generated proposal.

[1318] Input: Proposal (PDF or DOCX format)

[1319] Output: Display of proposal

[1320] Step 9:

[1321] Proposal customization

[1322] Users can customize proposals generated on their devices. For example, they can modify the layout or add comments using a text editor or specific software.

[1323] Input: Proposal (PDF or DOCX format)

[1324] Output: Customized proposal

[1325] Step 10:

[1326] Dialogue between factory robots

[1327] The robots in the factory interact with corporate clients and collect additional data. Specifically, they use voice recognition systems and sensors to obtain customer feedback and send it to a server.

[1328] Input: Customer voice commands, robot sensor data

[1329] Output: Sending customer feedback to the server

[1330] This series of steps allows us to accurately predict the needs of corporate clients and efficiently propose appropriate solutions.

[1331] 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.

[1332] This invention relates to a system that combines a system for accurately predicting the needs of corporate clients and proposing corresponding solutions with an emotion engine that recognizes user emotions. It primarily functions through the interaction of a server, terminal, and user to perform processes such as data collection, data preprocessing, data analysis, emotion recognition, solution proposal, and automatic generation and transmission of proposal documents.

[1333] Server Processing

[1334] Data collection

[1335] The server first collects general information and past order data for corporate customers from internal databases and external APIs. For example, for a major manufacturing company, it retrieves information such as the number of employees, main product lines, location, and past transaction history.

[1336] Data preprocessing

[1337] Next, the server preprocesses the collected data. It fills in any incomplete parts of the data, corrects erroneous data, and filters out unnecessary information. It also unifies and normalizes data in different formats. For example, it fills in missing purchase amounts in past transaction data with the average value.

[1338] Data Analysis

[1339] The server uses pre-processed data to apply a trained machine learning model to predict future needs. For example, it might predict that a large manufacturing company is likely to implement IT infrastructure in the future.

[1340] emotion recognition

[1341] The server uses an emotion engine to analyze user emotions based on input data. The emotion engine detects emotions from user-entered text, for example, by using natural language processing techniques. For instance, it analyzes user-entered comments and feedback to identify emotions such as "satisfied" or "dissatisfied."

[1342] Solution proposal

[1343] The server searches the database for the optimal solution based on predicted needs and sentiment analysis. For example, it extracts our products and services, such as cloud services and security solutions, as potential suggestions. Taking the results of the sentiment analysis into account, the system is customized, for example, by providing more detailed suggestions if the user is "dissatisfied."

[1344] Proposal generation

[1345] Once a proposal is approved, the server selects a proposal template based on the sentiment analysis results and automatically generates the proposal. This proposal includes basic company information, anticipated needs, details of the proposed products or services, benefits of implementation, and implementation track record with similar companies. The final proposal is saved in PDF format and prepared for transmission to the terminal.

[1346] Terminal processing

[1347] User input

[1348] The terminal displays a screen for the user to input company information. The sales representative (user) enters basic information about the company they are scheduled to visit. For example, they might enter information such as "a company in the manufacturing industry with 5,000 employees."

[1349] Data transmission

[1350] The entered information is sent from the terminal to the server.

[1351] Receiving and displaying proposals

[1352] A proposal generated from the server is sent to the terminal and displayed to the user. The proposal details solutions that address anticipated needs.

[1353] Proposal customization

[1354] The device also provides editing capabilities, allowing users to customize proposals. For example, users can fine-tune the layout of the proposal or add comments.

[1355] Sending a proposal

[1356] The customized proposal is sent to the client. It is sent using email or cloud sharing features.

[1357] Specific example

[1358] Example 1: Proposal of IT solutions for the manufacturing industry

[1359] For a manufacturing company, a user (sales representative) inputs company information into a terminal. The terminal sends this information to a server, which analyzes past order data and market trends to predict that improvements to the manufacturing process are needed. Furthermore, an emotion engine detects a "somewhat disappointed" emotion from the user's input comments. The server selects our IoT sensor system as a proposal and generates a proposal document based on it. The generated proposal document is sent to the terminal, where the user customizes it and then sends it to the company.

[1360] Example 2: Proposal of environmental protection products for the food industry

[1361] For companies in the food industry, users input company information into a terminal. Based on this information, the server predicts that there is a growing need for food safety and environmental measures. Furthermore, an emotion engine detects a "highly interested" sentiment from the user's input comments. The server selects our eco-friendly packaging materials as a suggestion and generates a proposal based on this. The proposal is sent to the terminal, where the user customizes it to increase satisfaction and finally sends it to the company.

[1362] This system enables sales representatives to quickly and effectively make appropriate proposals to clients, resulting in increased efficiency in sales activities, improved customer satisfaction, and more personalized proposals that reflect user emotions.

[1363] The following describes the processing flow.

[1364] Step 1:

[1365] The server collects general information about corporate customers from internal databases and external APIs. This general information includes industry, company size, main business activities, and location.

[1366] Step 2:

[1367] The server collects past order data from its internal database. This data includes information such as the products and services purchased by the customer, the transaction amount, the date and time of the transaction, and the terms and conditions of the transaction.

[1368] Step 3:

[1369] The server preprocesses the collected general information and order performance data. Specifically, it performs tasks such as imputing missing values, correcting incorrect data, filtering out unnecessary information, and normalizing data in different formats.

[1370] Step 4:

[1371] The server applies a pre-trained machine learning model based on pre-processed data to predict future needs. For example, it might predict that a large manufacturing company is likely to implement IT infrastructure in the future.

[1372] Step 5:

[1373] The server searches the database for solutions that address anticipated needs. For example, it extracts our products and services, such as cloud services and security solutions, as potential suggestions.

[1374] Step 6:

[1375] The server analyzes the text data entered by the user using an emotion engine to recognize emotions. For example, it identifies emotions such as "satisfied" or "dissatisfied" from the user's input comments.

[1376] Step 7:

[1377] The server combines emotion recognition results with needs predictions to select the optimal solution. Based on the emotion recognition results, it fine-tunes the content of the proposal and the selection of templates.

[1378] Step 8:

[1379] The server automatically generates a proposal based on the selected solution. The proposal includes company information, anticipated needs, details of the proposed products and services, benefits of implementation, and past success stories.

[1380] Step 9:

[1381] The server sends the generated proposal to the terminal. The proposal is saved and sent in PDF or Word format.

[1382] Step 10:

[1383] The user enters information about the company they plan to visit into an input form displayed on the terminal screen. This information includes the company name, industry, and past transaction history.

[1384] Step 11:

[1385] The terminal sends the entered company information to the server. The data is securely transferred using an API endpoint.

[1386] Step 12:

[1387] The terminal receives the proposal sent from the server and displays it in a format that is easy for the user to understand.

[1388] Step 13:

[1389] Users review the received proposals and customize them as needed. For example, they can add comments to the proposal or adjust the layout.

[1390] Step 14:

[1391] The user sends the customized proposal to the client. The proposal is sent via email or cloud sharing.

[1392] Specific example

[1393] Example 1: Proposal of IT solutions for the manufacturing industry

[1394] Step 1: The server collects general information about major manufacturing customers from a database and external APIs.

[1395] Step 2: The server collects the customer's past order history data.

[1396] Step 3: The server preprocesses the collected data, performing missing value imputation and data normalization.

[1397] Step 4: The server applies a machine learning model based on the pre-processed data to predict that the customer is likely to need IT infrastructure in the future.

[1398] Step 5: The server searches the database for IT infrastructure-related solutions.

[1399] Step 6: The server analyzes the comments entered by the user on the device using an emotion engine and detects the emotion "slightly anxious".

[1400] Step 7: Based on emotion recognition and needs prediction, the server selects a detailed IT infrastructure proposal.

[1401] Step 8: The server automatically generates a proposal and customizes the content to reflect emotions.

[1402] Step 9: Send the completed proposal to your device.

[1403] Step 10: The user enters the company information of the company they are visiting into the terminal.

[1404] Step 11: The terminal sends information to the server.

[1405] Step 12: Receive proposals sent from the server and display them to the user.

[1406] Step 13: The user reviews the proposal and makes customizations as needed.

[1407] Step 14: The user sends the final proposal to the client.

[1408] Example 2: Proposal of environmental protection products for the food industry

[1409] Step 1: The server collects customer information from the food industry.

[1410] Step 2: The server collects past transaction history.

[1411] Step 3: The server performs data imputation and normalization.

[1412] Step 4: The server predicts future needs using a trained machine learning model. It predicts that the need for safety and environmental measures will increase.

[1413] Step 5: Search the database for related products such as eco-friendly packaging materials.

[1414] Step 6: The server detects a "very interested" sentiment from the user's input comments.

[1415] Step 7: Based on the emotion recognition results, select detailed proposals for eco-friendly packaging materials.

[1416] Step 8: Automatically generate a proposal that reflects emotions.

[1417] Step 9: Send the proposal to your device.

[1418] Step 10: The user enters company information for a food industry company.

[1419] Step 11: The terminal sends information to the server.

[1420] Step 12: Receive proposals from the server and display them to the user.

[1421] Step 13: The user customizes the proposal.

[1422] Step 14: Send the final proposal to the client.

[1423] This system allows sales representatives to not only make optimal proposals to clients quickly and effectively, but also to make personalized proposals based on the user's emotions.

[1424] (Example 2)

[1425] 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."

[1426] Traditional proposal systems for corporate clients predicted customer needs based on general information and past order data, but they were insufficient in considering user emotions. As a result, customer satisfaction decreased, and proposals lacked personalization. Furthermore, even in automated proposal generation, the selection of templates was uniform, making it difficult to respond flexibly to reflect user emotions. There is a need to solve these problems and provide more accurate and personalized proposals.

[1427] 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.

[1428] In this invention, the server includes means for collecting general information of corporate customers, means for collecting past order performance data, means for pre-processing the above general information and order performance data, means for applying a machine learning model based on the pre-processed data to predict future needs, means for recognizing emotions based on user input data, means for proposing solutions corresponding to the predicted needs and the results of emotion recognition, means for automatically generating a proposal document based on the proposed solutions, and means for transmitting the generated proposal document to a terminal. This enables personalized proposals that take into account the user's emotions, thereby improving customer satisfaction and streamlining sales activities.

[1429] "Corporate clients" refer to customers other than individuals, such as companies and organizations.

[1430] "General information" refers to basic attribute information about corporate clients, including the number of employees, location, and main business activities.

[1431] "Order performance data" refers to past transaction history and order-related data with corporate clients.

[1432] "Preprocessing" refers to the process of formatting collected data to make it analyzable, supplementing incomplete data, and unifying data in different formats.

[1433] A "machine learning model" refers to an algorithm that learns certain patterns and rules based on data and uses that knowledge to make predictions and classifications on new data.

[1434] "Emotion recognition" refers to the process of identifying and specifying emotions based on user input data.

[1435] A "solution" refers to a specific product or service, or a combination thereof, proposed to address anticipated needs or problems.

[1436] A "proposal" is a document that details the proposed solution, including basic company information, anticipated needs, proposed content, and its benefits.

[1437] A "terminal" refers to a hardware device used by a user to interact with a system, and includes personal computers, tablets, smartphones, and other similar devices.

[1438] Modes for carrying out the invention

[1439] This invention relates to a system that combines a system for accurately predicting the needs of corporate clients and proposing corresponding solutions with an emotion engine that recognizes user emotions. It primarily functions through the interaction of a server, terminal, and user to perform processes such as data collection, data preprocessing, data analysis, emotion recognition, solution proposal, and automatic generation and transmission of proposal documents.

[1440] Server Processing

[1441] The server first collects general information and past order history data for corporate customers from its internal database and external APIs. For example, for a manufacturing company, it retrieves information such as the number of employees, main product lines, location, and past transaction history from the "customer_db" database and the "https: / / api.example.com / company-data" API.

[1442] Next, the server preprocesses the collected data. It uses algorithms (such as Python's Pandas library) to impart incomplete data and fills missing values ​​with the mean. It unifies data in different formats (e.g., date formatting) and filters out unnecessary information.

[1443] Based on the pre-processed data, the server applies a pre-trained machine learning model (such as "future_needs_model.pkl") to predict future needs. For example, it might predict that a certain manufacturing company is likely to implement IT infrastructure in the future.

[1444] Furthermore, the server uses an emotion engine to recognize emotions based on user input data. Using natural language processing (NLP) techniques, it assigns emotion labels such as "satisfied" or "dissatisfied" to the text data provided by the user.

[1445] Next, the server searches the database (e.g., "solutions_db") for the optimal solution based on the predicted needs and sentiment recognition results. For example, products and services such as cloud services and security solutions may be extracted as suggested options. Taking the sentiment analysis results into consideration, more detailed suggestions are made if the user indicates dissatisfaction.

[1446] Once a proposal is approved, the server selects a proposal template (such as "template1.html") based on the sentiment analysis results and automatically generates the proposal. This proposal includes basic company information, predicted needs, details of the proposed products or services, implementation benefits, and implementation track record of similar companies. The final proposal is saved in PDF format (through the "pdfkit" library) and ready to be sent to the terminal.

[1447] Terminal processing

[1448] The terminal displays a screen for the user to input company information. The sales representative (user) enters basic information about the company they are scheduled to visit (e.g., manufacturing industry, 5,000 employees) into an HTML form. The entered information is sent from the terminal to the server in JSON format.

[1449] A proposal generated from the server is sent to the terminal and displayed to the user. The proposal details solutions that address anticipated needs. Users can further customize the proposal using the provided editing functions. For example, they can fine-tune the layout using a WYSIWYG editor or add comments.

[1450] The customized proposal is sent to the client via email or cloud sharing.

[1451] Specific example

[1452] Example 1: Proposal of IT solutions for the manufacturing industry

[1453] For a manufacturing company, a user (sales representative) inputs company information into a terminal. The terminal sends this information to a server, which analyzes past order data and market trends to predict that improvements to the manufacturing process are needed. Furthermore, an emotion engine detects a "somewhat disappointed" emotion from the user's input comments. The server selects our IoT sensor system as a proposal and generates a proposal document based on it. The generated proposal document is sent to the terminal, where the user customizes it and then sends it to the company.

[1454] Example 2: Proposal of environmental protection products for the food industry

[1455] For companies in the food industry, users input company information into a terminal. Based on this information, the server predicts that there is a growing need for food safety and environmental measures. Furthermore, an emotion engine detects a "highly interested" sentiment from the user's input comments. The server selects our eco-friendly packaging materials as a suggestion and generates a proposal based on this. The proposal is sent to the terminal, where the user customizes it to increase satisfaction and finally sends it to the company.

[1456] Examples of prompts for generative AI models

[1457] "Please describe the system's overview, which predicts the needs of corporate clients and proposes appropriate solutions. Please also provide details on the emotion recognition component based on user input data."

[1458] This prompt statement allows you to request specific explanations from the generated AI model.

[1459] This system enables sales representatives to quickly and effectively make appropriate proposals to clients, resulting in increased sales efficiency and improved customer satisfaction. It also allows for more personalized proposals that reflect user emotions.

[1460] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1461] Step 1:

[1462] The server collects general information about corporate customers from an internal database (customer_db) and an external API (https: / / api.example.com / company-data). Specifically, the server uses an API key to access the external API and retrieves data in JSON format. The input is the API endpoint and authentication information, and the output is general information data about corporate customers.

[1463] Step 2:

[1464] The server collects past order data from its internal database. It establishes a database connection and executes SQL queries such as "SELECT FROM orders WHERE customer_id = ?". The input is the customer ID, and the output is past order data.

[1465] Step 3:

[1466] The server preprocesses the collected general information and order history data. Specifically, it uses the Pandas library to impute missing values ​​with the mean, correct inaccuracies in the data, and filter out unnecessary information. The input is raw data, and the output is preprocessed, clean data.

[1467] Step 4:

[1468] The server uses preprocessed data to apply machine learning models and predict future needs. For example, it loads "future_needs_model.pkl" and uses the Scikit-learn library to predict future demand. The input is preprocessed data, and the output is the predicted needs.

[1469] Step 5:

[1470] The server recognizes emotions based on user input data. Using an NLP library (e.g., SpaCy), it identifies emotions from the input text and assigns emotion labels such as "satisfied" or "dissatisfied." The input is user text data, and the output is emotion labels.

[1471] Step 6:

[1472] The server searches for appropriate solutions based on predicted needs and sentiment recognition results. It uses SQL queries to retrieve the best solutions from the database (solutions_db) and extracts them in list format. The input is the predicted needs and sentiment labels, and the output is a list of suggested solutions.

[1473] Step 7:

[1474] The server automatically generates a proposal based on the proposed solution. Using a template engine (such as Jinja), it embeds data into the selected template to generate the final proposal. PDFkit is used to convert it to PDF format. The input is the proposed solution and template, and the output is a PDF file of the proposal.

[1475] Step 8:

[1476] The server sends the generated proposal to the user's terminal. It sends a PDF file as an HTTP response, which the user receives. The input is the generated PDF file, and the output is the transmission to the user's terminal.

[1477] Step 9:

[1478] The terminal displays the proposal to the user. It opens a PDF viewer to display the received PDF file. The input is the proposal PDF file, and the output is the proposal that the user can view.

[1479] Step 10:

[1480] The user customizes the proposal. They use the editing functions provided by the terminal (such as a WYSIWYG editor) to modify text and change the layout. The input is the received proposal and the user's edits, and the output is the customized proposal.

[1481] Step 11:

[1482] The user sends the customized proposal to the client. The proposal is sent via email or cloud sharing. The input is the customized proposal, and the output is the document sent to the client.

[1483] (Application Example 2)

[1484] 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."

[1485] Traditional corporate client solution proposal systems provided proposals based on predicted customer needs, but lacked personalization of proposals and customization of proposals to consider customer emotions. Therefore, it was difficult to improve customer satisfaction and maximize the effectiveness of proposals. In the security services sector in particular, there is a need to provide more accurate proposals that take customer emotions into consideration.

[1486] 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.

[1487] In this invention, the server includes means for collecting general information of corporate customers, means for collecting past order performance data, means for pre-processing the above general information and order performance data, means for applying a machine learning model based on the pre-processed data to predict future needs, means for recognizing emotions from user input data, means for proposing solutions based on the predicted needs and recognized emotions, means for automatically generating a proposal document based on the proposed solutions, and means for transmitting the generated proposal document to a terminal. This enables more personalized proposals that take into account the predicted needs and emotions of corporate customers.

[1488] A "corporate customer" refers to a specific corporation or company that is targeted as a recipient of products or services.

[1489] "General information" refers to basic attribute information of a company, including data such as industry, number of employees, and location.

[1490] "Order history data" refers to information about past transactions and contracts, specifically including historical data such as transaction amount, item, and transaction date.

[1491] "Preprocessing" refers to the process of preparing collected data into an analyzable format, and includes operations such as data cleaning, imputation of missing values, and normalization.

[1492] A "machine learning model" refers to an algorithm or model used to predict future outcomes based on data analysis, and is one that has learned past patterns using training data.

[1493] "Needs forecasting" is the process of predicting the products and services that corporate customers may need in the future, based on collected data and machine learning models.

[1494] "Emotion recognition" refers to the process of analyzing emotions from user input data and expressing the results numerically or in classifications, using technologies such as natural language processing.

[1495] "Solution proposal" refers to the act of selecting the most suitable products and services based on anticipated needs and perceived emotions, and proposing them to corporate clients.

[1496] "Automatic proposal generation" refers to the process of automatically creating a proposal based on a solution proposal, using templates to combine content.

[1497] "Sending to a device" refers to the process of electronically sending an automatically generated proposal to a device, specifically using methods such as email or cloud sharing.

[1498] This invention combines a system that accurately predicts the needs of corporate clients and proposes corresponding solutions with an emotion engine that recognizes user emotions. This maximizes the effectiveness of solution proposals to corporate clients.

[1499] Server Processing

[1500] The server performs various processes using the following methods.

[1501] 1. Data Collection

[1502] The server collects general information and past order history data of corporate customers from internal databases and external APIs. The collected data includes basic company data and transaction history.

[1503] 2. Data preprocessing

[1504] The collected data is cleaned and normalized using Pandas. Missing values ​​are imputed and the data is standardized, preparing it for analysis.

[1505] 3. Data Analysis

[1506] Based on pre-processed data, a pre-trained machine learning model (TensorFlow or PyTorch) is applied to predict future needs.

[1507] 4. Emotion recognition

[1508] The server uses the Google Cloud Natural Language API to analyze emotions from text entered by the user. This allows it to quantify and extract emotions such as "satisfied" or "dissatisfied."

[1509] 5. Solution Proposal

[1510] Based on predicted needs and sentiment analysis results, the system proposes the most suitable products and services from its solution database. This includes customization, such as providing detailed suggestions if the sentiment is "dissatisfied."

[1511] 6. Proposal generation

[1512] Based on the proposed content, a proposal document will be automatically generated. The proposal document will be generated using a selection of multiple templates and will include the proposed content, company information, and implementation benefits. It will be generated and saved in PDF format.

[1513] Terminal processing

[1514] The device interacts with the user using the following means:

[1515] 1. User input

[1516] The terminal provides an interface for entering corporate customer information. The user enters basic information about the company they plan to visit (for example, "a manufacturing company with 2,000 employees and 5 security incidents in the past 3 years").

[1517] 2. Data transmission

[1518] The entered information is sent from the terminal to the server.

[1519] 3. Receiving and displaying proposals

[1520] The proposal generated from the server is sent to the terminal and displayed to the user. The proposal details solutions that address anticipated needs.

[1521] 4. Customizing the proposal

[1522] The device provides editing capabilities that allow users to customize proposals, enabling fine-tuning of the layout and the addition of comments.

[1523] 5. Submission of proposal

[1524] The customized proposal is sent to the corporate client via email or cloud sharing.

[1525] Specific example

[1526] For a manufacturing company, a user inputs information into a terminal, such as "a manufacturing company with 2,000 employees and five security incidents in the past three years." The terminal sends the information to a server, which analyzes the data, recognizes emotions, and proposes the most suitable security measures.

[1527] Example of a prompt

[1528] "I would like to propose IT security measures for a manufacturing company with 2,000 employees that has experienced five security incidents in the past three years."

[1529] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1530] Step 1:

[1531] Users enter general and basic information about their corporate clients into the terminal. This includes industry, number of employees, location, and past security incidents. The entered data is temporarily stored in JSON format.

[1532] Step 2:

[1533] The terminal sends the entered information to the server. Data transmission is performed securely and encrypted using the HTTPS protocol. The server stores the received data in a database.

[1534] Step 3:

[1535] The server collects general information and past order performance data for corporate customers from internal databases and external APIs. Data collection involves obtaining data in JSON format from each API and integrating it into the database.

[1536] Step 4:

[1537] The server preprocesses the collected data using Pandas. This includes imputing missing values, filtering out unnecessary data, and normalizing the data. The preprocessed data is then converted into a format that is easy for machine learning models to handle.

[1538] Step 5:

[1539] The server applies a pre-trained machine learning model (TensorFlow or PyTorch) based on pre-processed data to predict the future needs of corporate customers. Specifically, if it determines that security measures are necessary, it outputs the details of those measures.

[1540] Step 6:

[1541] The server uses the Google Cloud Natural Language API to recognize emotions from user input data. It parses text data and extracts a numerical emotion score. The emotion score is used to customize suggestions.

[1542] Step 7:

[1543] Based on predicted needs and sentiment analysis results, the server suggests the most suitable products and services from its solution database. If the sentiment score indicates "dissatisfaction," the suggestions are adjusted to include more detailed explanations and additional options.

[1544] Step 8:

[1545] The server automatically generates proposals. These proposals, containing details such as the proposed solution, company information, and implementation benefits, are generated in PDF format. The PDFkit library is used for generation, and multiple templates are selected.

[1546] Step 9:

[1547] The server sends the generated proposal to the terminal. The terminal displays the received proposal to the user and provides an editing interface for customization. The user can fine-tune the content of the proposal and add comments.

[1548] Step 10:

[1549] The customized proposals, completed on the terminal, are sent to corporate clients via email or cloud sharing. This ensures that the proposals reach the clients and that effective security measures are proposed.

[1550] 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.

[1551] 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.

[1552] 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.

[1553] [Fourth Embodiment]

[1554] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1555] 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.

[1556] 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).

[1557] 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.

[1558] 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.

[1559] 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).

[1560] 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.

[1561] 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.

[1562] 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.

[1563] 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.

[1564] 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.

[1565] 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.

[1566] 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".

[1567] This invention relates to a system that accurately predicts the needs of corporate clients and proposes solutions to address them. It primarily functions through the interaction of a server, terminals, and users to perform processes such as data collection, data preprocessing, data analysis, solution proposal, and automatic generation and transmission of proposal documents.

[1568] Server Processing

[1569] Data collection

[1570] The server first collects general information and past order data for corporate customers from internal databases and external APIs. For example, for a major manufacturing company, it retrieves information such as the number of employees, main product lines, location, and past transaction history.

[1571] Data preprocessing

[1572] Next, the server preprocesses the collected data. It fills in any incomplete parts of the data, corrects erroneous data, and filters out unnecessary information. It also unifies and normalizes data in different formats. For example, it fills in missing purchase amounts in past transaction data with the average value.

[1573] Data Analysis

[1574] The server uses pre-processed data to apply a trained machine learning model to predict future needs. For example, it is predicted that large manufacturing companies are likely to adopt IoT technology in the future.

[1575] Solution proposal

[1576] The server searches the database for appropriate products and services to propose solutions that address anticipated needs. For example, it might suggest our IoT sensor system for a predicted IoT technology deployment.

[1577] Proposal generation

[1578] Once a proposal is approved, the server selects a proposal template and automatically generates the proposal based on it. This proposal includes basic company information, anticipated needs, details of the proposed products or services, benefits of implementation, and implementation track record with similar companies. The final proposal is saved in PDF format and prepared for transmission to the terminal.

[1579] Terminal processing

[1580] User input

[1581] The terminal displays a screen for the user to input company information. The sales representative (user) enters basic information about the company they are scheduled to visit. For example, they might enter information such as "a company in the manufacturing industry with 5,000 employees."

[1582] Data transmission

[1583] The entered information is sent from the terminal to the server.

[1584] Receiving and displaying proposals

[1585] A proposal generated from the server is sent to the terminal and displayed to the user. The proposal details solutions that address anticipated needs.

[1586] Proposal customization

[1587] The device also provides editing capabilities, allowing users to customize proposals. For example, users can fine-tune the layout of the proposal or add comments.

[1588] Sending a proposal

[1589] The customized proposal is sent to the client. It is sent using email or cloud sharing features.

[1590] Specific example

[1591] Example 1: Proposal of IT solutions for the manufacturing industry

[1592] For a manufacturing company called "ABC Manufacturing Co., Ltd.," a user (sales representative) inputs company information into a terminal. The terminal sends this information to a server, which analyzes past order data and market trends to predict that improvements to the manufacturing process are needed. The server selects our IoT sensor system as a proposal and generates a proposal document based on it. The generated proposal document is sent to the terminal, and the user customizes it before sending it to ABC Manufacturing Co., Ltd.

[1593] Example 2: Proposal of environmental protection products for the food industry

[1594] For the food industry company "XYZ Food Co., Ltd.", the user enters company information into a terminal. Based on this information, the server predicts that there is a growing need for food safety and environmental measures. The server selects our eco-friendly packaging materials as a suggestion and generates a proposal document based on it. The proposal document is sent to the terminal, the user customizes it, and finally sends it to XYZ Food Co., Ltd.

[1595] This system enables sales representatives to make appropriate proposals to clients quickly and effectively, resulting in increased efficiency in sales activities and improved customer satisfaction.

[1596] The following describes the processing flow.

[1597] Step 1:

[1598] The server collects general information about corporate customers from internal databases and external APIs. This general information includes industry information, company size, main business activities, and location.

[1599] Step 2:

[1600] The server collects past order data from its internal database. This data includes information such as the products and services purchased by the customer, the transaction amount, the date and time of the transaction, and the terms and conditions of the transaction.

[1601] Step 3:

[1602] The server preprocesses the collected general information and order performance data. This includes imputing missing values, correcting incorrect data, filtering out unnecessary information, and normalizing data in different formats.

[1603] Step 4:

[1604] The server applies machine learning models based on pre-processed data to predict the future needs of corporate customers. For example, it uses historical transaction data and market trends to predict increased demand for IT infrastructure.

[1605] Step 5:

[1606] The server searches the database for solutions that address anticipated needs. For example, it extracts our products and services, such as cloud services and security solutions, as potential suggestions.

[1607] Step 6:

[1608] The server selects the optimal solution from the proposed candidates. This selection takes into account factors such as cost-effectiveness, implementation benefits, and past performance.

[1609] Step 7:

[1610] The server automatically generates a proposal based on the selected solution. The proposal includes general company information, projected needs, details of the proposed products and services, benefits of implementation, and past success stories.

[1611] Step 8:

[1612] The server sends the generated proposal to the terminal. The proposal is saved and sent in PDF or Word format.

[1613] Step 9:

[1614] The user enters information about the company they plan to visit into an input form displayed on the terminal screen. This information includes the company name, industry, and past transaction history.

[1615] Step 10:

[1616] The terminal sends the entered company information to the server. The data is securely transferred using an API endpoint.

[1617] Step 11:

[1618] The terminal receives the proposal sent from the server. It saves the proposal and displays it in a user-friendly format.

[1619] Step 12:

[1620] Users review the received proposals and customize them as needed. For example, they can add comments to the proposal or adjust the layout.

[1621] Step 13:

[1622] The user sends the customized proposal to the client. The proposal is sent via email or cloud sharing.

[1623] (Example 1)

[1624] 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".

[1625] Traditional systems required significant time and effort to predict the needs of corporate clients and propose appropriate solutions. Furthermore, proposal creation and customization were often done manually, leading to inefficiencies and a high likelihood of errors. This hindered the efficiency of sales activities and improved customer satisfaction.

[1626] 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.

[1627] In this invention, the server includes means for collecting general information of corporate customers, means for collecting past order performance data, means for pre-processing the above general information and order performance data, means for applying a machine learning model based on the pre-processed data to predict future needs, means for proposing solutions that address the predicted needs, means for automatically generating a proposal document based on the proposed solutions, means for sending the generated proposal document to a terminal, means for the user to input company information using the terminal, means for sending the input information to the server, means for displaying the proposal document sent from the server on the terminal, means for customizing the proposal document on the terminal, and means for sending the customized proposal document to the client. This enables the generation and transmission of proposal documents that respond quickly and accurately to the requests of corporate customers, thereby improving the efficiency of sales activities and enhancing customer satisfaction.

[1628] A "corporate client" is a client that has a specific legal entity, such as a company or organization.

[1629] "General information" refers to basic information about corporate clients, including, for example, name, address, industry, and number of employees.

[1630] "Order history data" refers to historical information on transactions previously received from corporate clients, including, for example, the date and time of the transaction, the details of the transaction, the transaction amount, and product details.

[1631] "Preprocessing" refers to the process of supplementing incomplete parts of collected data, correcting errors, and filtering out unnecessary information.

[1632] A "machine learning model" is a statistical model that includes algorithms to learn specific patterns and relationships based on data and to make future predictions.

[1633] "Needs forecasting" refers to estimating the future needs and requirements of corporate customers using machine learning models.

[1634] "Solution proposal" means presenting appropriate products or services to meet anticipated needs.

[1635] A "proposal" is a document that details a solution and presents it to a corporate client.

[1636] "Automatic generation" refers to the process of automatically creating a document based on a series of data points using a program.

[1637] A "device" refers to a device operated by a user, such as a computer, tablet, or smartphone.

[1638] "Customization" refers to modifying and adjusting the generated proposal to meet the user's needs.

[1639] "Sending" refers to sending data from a server to a terminal, or from a terminal to a client.

[1640] "Client" refers to the recipient of the proposal, and usually refers to the person in charge at a corporate client company.

[1641] This invention relates to a system that accurately predicts the needs of corporate clients and proposes optimal solutions based on those predictions. It primarily functions through the interaction of a server, terminals, and users, via processes such as data collection, data preprocessing, data analysis, solution proposal, and automatic generation and transmission of proposal documents.

[1642] Hardware and software configuration

[1643] Server Configuration

[1644] Hardware: A server is a computer system equipped with a high-performance processor, sufficient memory, and large-capacity storage. It also includes hardware for database management systems (e.g., MySQL) and integration with external APIs.

[1645] Software: Python is used for data processing, MySQL for databases, the requests library for communication with external APIs, and various libraries such as pandas, numpy, scikit-learn, and TensorFlow for data analysis. Python-docx is used for document generation, and WeasyPrint is used for PDF conversion.

[1646] Device configuration

[1647] Hardware: A terminal is a device that a user accesses, and includes PCs, tablets, smartphones, etc.

[1648] Software: The user interface uses HTML and JavaScript, data transmission uses the JavaScript fetch API or Axios, and libraries such as PDF.js are used for displaying proposals.

[1649] Process details

[1650] Data collection

[1651] The server collects general information and past order data for corporate customers from an internal database and external APIs. Specifically, it retrieves information such as customer ID, order date, purchase amount, and purchased items from the internal database, and uses external APIs to retrieve additional data such as the company's location and industry classification.

[1652] Data preprocessing

[1653] The collected data is supplemented to fill in incomplete parts, erroneous data is corrected, and unnecessary information is filtered out. Python's pandas and numpy libraries are used to unify and normalize data in different formats. Additionally, sklearn.impute.SimpleImputer is used to impute missing values.

[1654] Data Analysis

[1655] Based on preprocessed data, machine learning models are applied to predict the future needs of corporate customers. The prediction models include scikit-learn's Random Forest and TensorFlow's deep learning models. Future demand is predicted by loading the models and calling the predict method.

[1656] Solution proposal

[1657] The system searches the database for the most suitable products and services to meet anticipated needs. For example, it might select an IoT sensor system as a proposal for "implementing IoT technology." It then executes an SQL query to retrieve product information.

[1658] Proposal generation

[1659] Load the proposal template and insert the anticipated needs and corresponding solutions. Create the proposal using Python-docx, convert it to PDF format using WeasyPrint, and save it. This proposal will then be sent to the terminal.

[1660] User input on the terminal

[1661] Users input basic corporate customer information through the terminal screen, and the terminal sends this information to the server. A JavaScript and HTML interface is used to enable efficient data entry.

[1662] Receiving and displaying proposals

[1663] The generated proposal is sent to the device and displayed to the user. PDF.js is used to display the contents of the proposal.

[1664] Proposal customization

[1665] Users can customize their proposals on their devices using a WYSIWYG editor (such as TinyMCE or CKEditor) to adjust the content and layout of the proposal.

[1666] Sending a proposal

[1667] Customized proposals are sent as email attachments or provided to clients via a generated cloud sharing link. They are sent using SMTP clients or cloud storage APIs (e.g., Google Drive API).

[1668] Specific example

[1669] A concrete example of a prompt is, "Based on transaction data and company information from the past five years, please tell me how to predict future needs and propose appropriate solutions." By inputting this prompt into the AI ​​generation model, effective suggestions can be obtained.

[1670] This system enables us to provide optimal solutions to corporate clients quickly and accurately, resulting in increased efficiency in sales activities and improved customer satisfaction.

[1671] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1672] The program's processing flow will be explained in steps.

[1673] Step 1:

[1674] Data collection

[1675] The server retrieves general information and past order data for corporate customers from its internal database. Furthermore, it calls external APIs (e.g., Google Places API) to obtain additional data such as company location information and industry classification.

[1676] Input fields: Company ID, number of employees, order date, purchase amount, purchased items, etc.

[1677] Output: A set of collected general information and past order performance data.

[1678] Specific operation: The server uses the Python requests library to collect data from an external API and executes SQL queries to retrieve the necessary information from an internal database.

[1679] Step 2:

[1680] Data preprocessing

[1681] The server preprocesses the collected data. Specifically, it uses the pandas library to impute missing data using appropriate methods (e.g., imputing with the mean), corrects incorrect data (e.g., negative purchase amounts), and filters out unnecessary information. It also normalizes data by unifying different formats.

[1682] Input: A set of collected general information and past order performance data.

[1683] Output: Preprocessed, clean dataset.

[1684] Specific operation: The server uses pandas and numpy to perform data cleaning and preprocessing.

[1685] Step 3:

[1686] Data Analysis

[1687] The server uses pre-processed data to train models such as scikit-learn's Random Forest and TensorFlow's deep learning models, predicting the future needs of corporate clients.

[1688] Input: A pre-processed, clean dataset.

[1689] Output: Predicted future needs (e.g., potential for IoT technology implementation).

[1690] Specific operation: The server loads the trained model, makes predictions using the predict method, and saves the results.

[1691] Step 4:

[1692] Solution proposal

[1693] The server searches the database for solutions that address anticipated needs and selects the most suitable products and services.

[1694] Input: Predicted future needs.

[1695] Output: A set of proposed solutions (e.g., IoT sensor system).

[1696] Specific operation: The server uses SQL queries to retrieve information about the best solution from the database.

[1697] Step 5:

[1698] Proposal generation

[1699] The server loads a proposal template, fills in the anticipated needs and corresponding solutions, and automatically generates the proposal. The generated proposal is converted to PDF format.

[1700] Input: A set of proposed solutions.

[1701] Output: Automatically generated proposal (PDF format).

[1702] Specific operation: The server uses Python-docx to create a proposal in DOCX format and converts it to PDF using WeasyPrint.

[1703] Step 6:

[1704] Sending a proposal

[1705] The server sends the generated proposal to the terminal.

[1706] Input: Automated proposal (PDF format).

[1707] Output: Send proposal data to the terminal.

[1708] Specific operation: The server sends the PDF data to the terminal as an HTTP response.

[1709] Step 7:

[1710] User input on the terminal

[1711] The user enters the company information they plan to visit on the terminal's input screen.

[1712] Input: Company information (e.g., company name, number of employees, industry, location).

[1713] Output: Sending corporate information from the terminal to the server.

[1714] Specific operation: The terminal uses HTML forms and JavaScript to build the user interface, validate the input data, and submit it.

[1715] Step 8:

[1716] Display of the proposal

[1717] The terminal displays the proposal sent from the server using a PDF viewer.

[1718] Input: Proposal submitted from the server (PDF format).

[1719] Output: The proposal displayed to the user.

[1720] Specific operation: The terminal uses the PDF.js library to display the contents of the PDF.

[1721] Step 9:

[1722] Proposal customization

[1723] Users customize proposals on their devices. For example, they can edit the layout and content of the proposal.

[1724] Input: The displayed proposal.

[1725] Output: Customized proposal.

[1726] Specific operation: The terminal allows users to customize proposals using WYSIWYG editors such as TinyMCE or CKEditor.

[1727] Step 10:

[1728] Sending a proposal

[1729] Customized proposals are sent to clients via email or cloud sharing.

[1730] Input: Customized proposal.

[1731] Output: Sending the proposal to the client.

[1732] Specific actions: The device will send emails using an SMTP client or share files using the Google Drive API, etc.

[1733] Examples of prompt statements that have been explicitly stated so far

[1734] A concrete example of a prompt is, "Based on transaction data and company information from the past five years, please tell me how to predict future needs and propose appropriate solutions." By inputting this prompt into the AI ​​generation model, effective suggestions can be obtained.

[1735] (Application Example 1)

[1736] 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".

[1737] Accurately predicting the needs of corporate clients and promptly proposing solutions that address those needs is crucial for improving the efficiency of sales activities and enhancing customer satisfaction. However, the systems required to achieve this are complex and necessitate the accurate processing and analysis of large amounts of data. Furthermore, customizing proposals and engaging with clients are also important elements. This invention aims to solve these problems and realize efficient and accurate solution proposals.

[1738] 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.

[1739] In this invention, the server includes means for collecting summary information of corporate customers, means for collecting past transaction history data, means for pre-processing the summary information and transaction history data, means for applying a machine learning model based on the pre-processed data to predict future demand, means for proposing solutions that correspond to the predicted demand, means for automatically generating a proposal document based on the proposed solutions, means for transmitting the generated proposal document to a terminal, means for customizing the proposal document, and means for interacting with corporate customers using robots in the factory. This makes it possible to accurately predict the needs of corporate customers and propose appropriate solutions efficiently and quickly.

[1740] "Corporate clients" refer to customers such as companies and organizations that have commercial or business needs.

[1741] "Summary information" refers to basic attribute information of corporate clients, including information such as industry, number of employees, location, and year of establishment.

[1742] "Transaction history data" refers to data related to past transactions with corporate customers, including information such as purchased items, transaction amount, number of transactions, and transaction date.

[1743] "Preprocessing" refers to a series of processes to prepare collected data for analysis, including imputation of missing values ​​and data normalization.

[1744] A "machine learning model" refers to an algorithm used to make specific predictions or classifications based on data, and is often trained on a training dataset.

[1745] "Demand forecasting" refers to predicting the products and services that corporate customers will need in the future, based on collected data.

[1746] A "solution" refers to a specific product or service proposed to address anticipated needs.

[1747] A "proposal" is a document that details a proposed solution and is submitted to corporate clients.

[1748] A "device" is a device used by a user to operate something, and includes smartphones, tablets, and personal computers.

[1749] "Customization" refers to modifying and adjusting a generated proposal to suit a specific corporate client.

[1750] "Robots in a factory" refers to mechanical devices that operate within a factory and perform designated tasks.

[1751] "Dialogue" refers to the act of robots in a factory communicating with corporate clients, and includes activities such as gathering information and proposing solutions.

[1752] This invention relates to the realization of a system that accurately predicts the needs of corporate clients and proposes solutions that address those needs. The specific operation and components of the system are described below.

[1753] Server Processing

[1754] Data collection

[1755] The server first collects summary information and historical transaction data for corporate customers. This is done using an internal database and an external API. The server retrieves data through API endpoints and uses the Python pandas library to organize it into a data frame.

[1756] Data preprocessing

[1757] The collected data is preprocessed by imputing missing values, normalizing the data, and filtering out unnecessary information. This is done using pandas and the StandardScaler class from scikit-learn.

[1758] Data Analysis

[1759] A machine learning model is applied based on preprocessed data. The prediction model is assumed to be a linear regression model using scikit-learn, which is trained on historical transaction data. This analysis predicts future demand.

[1760] Solution proposal

[1761] Based on the prediction results, appropriate solutions are proposed. For example, if high demand is predicted, the introduction of IoT sensors is suggested.

[1762] Proposal generation

[1763] Based on the proposed solutions, a proposal document is automatically generated. This proposal document includes an overview of the corporate client, projected demand, details of the proposed products and services, and the benefits of implementation. This process is carried out by converting the text data generated by Python into PDF format.

[1764] Sending a proposal

[1765] The generated proposal is sent to the terminal. An email sending library is used for this purpose. The proposal is sent via a cloud platform and is provided in real time.

[1766] Terminal processing

[1767] User input

[1768] The terminal displays a screen for the user to enter basic information about the corporate customer. The user (sales representative) enters the necessary information through this screen.

[1769] Data transmission

[1770] The information entered is sent from the terminal to the server. This allows the data to be collected on the server in real time.

[1771] Receiving and displaying proposals

[1772] Proposals sent from the server are displayed on the terminal. Users can review them and present them to corporate clients.

[1773] Proposal customization

[1774] The proposal document is customizable by the user. For example, the layout of the proposal document can be fine-tuned, or comments can be added.

[1775] Processing by robots in the factory

[1776] Dialogue

[1777] Factory robots interact with corporate clients. Through these interactions, they can collect additional data and offer immediate suggestions. Equipped with advanced sensors and voice recognition systems, the robots can delve deeper into the needs of corporate clients.

[1778] Specific example

[1779] Example 1: Proposal of IT solutions for the manufacturing industry

[1780] For corporate clients in the industrial manufacturing sector, the user inputs basic information into a terminal, and the server collects, preprocesses, and analyzes this information. Based on the analysis results, the introduction of IoT sensors is proposed, and a proposal document is generated. The proposal document is sent to the terminal, and the user customizes it and presents it to the corporate client.

[1781] Examples of prompts for generative AI models

[1782] "Based on the following company information, predict their needs and propose appropriate solutions. Company Information: Industry = Manufacturing, Number of Employees = 3000, Past Transaction History = High, Technology Implementation History = Medium"

[1783] Hardware and software

[1784] Server: Data collection, data preprocessing, data analysis, solution proposal, proposal generation.

[1785] Libraries used: Python, pandas, scikit-learn, requests

[1786] Terminals: User input, data transmission, proposal reception, proposal customization

[1787] Devices: Smartphones, tablets, PCs

[1788] Factory robots: Interacting with corporate clients, collecting data, and proposing solutions.

[1789] Technology: Advanced sensors, voice recognition systems

[1790] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1791] Step 1:

[1792] Data collection

[1793] The server collects summary information and historical transaction history data for corporate customers from API endpoints and internal databases. Specifically, it uses the requests module to retrieve data from external APIs in JSON format and converts it into a DataFrame using the pandas library.

[1794] Input: API endpoint URL, database connection information

[1795] Output: Dataframe of customer information and transaction history

[1796] Step 2:

[1797] Data preprocessing

[1798] The server imputes missing values ​​in the collected data and normalizes the data. Specifically, it uses pandas to impute missing values ​​with the mean and scikit-learn's StandardScaler to standardize the data.

[1799] Input: Dataframe of customer information and transaction history

[1800] Output: Preprocessed data frame

[1801] Step 3:

[1802] Data Analysis

[1803] The server applies a machine learning model to preprocessed data to predict future demand. Specifically, it uses a linear regression model from scikit-learn and makes predictions using the pre-trained model.

[1804] Input: Preprocessed data frame

[1805] Output: Demand forecast results (predicted values)

[1806] Step 4:

[1807] Solution proposal

[1808] The server proposes appropriate solutions based on the prediction results. For example, for high demand forecasts, it suggests the introduction of IoT sensors. The proposed solutions are obtained by searching for appropriate product information in the database.

[1809] Input: Demand forecast results

[1810] Output: Proposed solutions (list of products and services)

[1811] Step 5:

[1812] Proposal generation

[1813] The server automatically generates proposals. These proposals include an overview of the corporate client, projected demand, and details of the proposed products and services. Because they are generated based on templates, Python-docx and the ReportLab library are often used.

[1814] Input: Corporate customer overview information, projected demand, and details of proposed products and services.

[1815] Output: Proposal (PDF or DOCX format)

[1816] Step 6:

[1817] Sending a proposal

[1818] The server sends the generated proposal to the terminal. Here, it uses an email sending library (e.g., smtplib) to attach the proposal to the end user's terminal and send it.

[1819] Input: Proposal (PDF or DOCX format), end user's email address

[1820] Output: Notification of proposal submission to end users

[1821] Step 7:

[1822] User input

[1823] Users input basic information about corporate customers from their terminals. Specifically, they enter information using a form and submit it to the system.

[1824] Input: Basic information of the corporate customer (e.g., industry, number of employees, location)

[1825] Output: Sending input information to the system

[1826] Step 8:

[1827] Receiving and displaying proposals

[1828] The terminal receives the proposal sent from the server and displays it to the user. This allows the user to review the contents of the generated proposal.

[1829] Input: Proposal (PDF or DOCX format)

[1830] Output: Display of proposal

[1831] Step 9:

[1832] Proposal customization

[1833] Users can customize proposals generated on their devices. For example, they can modify the layout or add comments using a text editor or specific software.

[1834] Input: Proposal (PDF or DOCX format)

[1835] Output: Customized proposal

[1836] Step 10:

[1837] Dialogue between factory robots

[1838] The robots in the factory interact with corporate clients and collect additional data. Specifically, they use voice recognition systems and sensors to obtain customer feedback and send it to a server.

[1839] Input: Customer voice commands, robot sensor data

[1840] Output: Sending customer feedback to the server

[1841] This series of steps allows us to accurately predict the needs of corporate clients and efficiently propose appropriate solutions.

[1842] 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.

[1843] This invention relates to a system that combines a system for accurately predicting the needs of corporate clients and proposing corresponding solutions with an emotion engine that recognizes user emotions. It primarily functions through the interaction of a server, terminal, and user to perform processes such as data collection, data preprocessing, data analysis, emotion recognition, solution proposal, and automatic generation and transmission of proposal documents.

[1844] Server Processing

[1845] Data collection

[1846] The server first collects general information and past order data for corporate customers from internal databases and external APIs. For example, for a major manufacturing company, it retrieves information such as the number of employees, main product lines, location, and past transaction history.

[1847] Data preprocessing

[1848] Next, the server preprocesses the collected data. It fills in any incomplete parts of the data, corrects erroneous data, and filters out unnecessary information. It also unifies and normalizes data in different formats. For example, it fills in missing purchase amounts in past transaction data with the average value.

[1849] Data Analysis

[1850] The server uses pre-processed data to apply a trained machine learning model to predict future needs. For example, it might predict that a large manufacturing company is likely to implement IT infrastructure in the future.

[1851] emotion recognition

[1852] The server uses an emotion engine to analyze user emotions based on input data. The emotion engine detects emotions from user-entered text, for example, by using natural language processing techniques. For instance, it analyzes user-entered comments and feedback to identify emotions such as "satisfied" or "dissatisfied."

[1853] Solution proposal

[1854] The server searches the database for the optimal solution based on predicted needs and sentiment analysis. For example, it extracts our products and services, such as cloud services and security solutions, as potential suggestions. Taking the results of the sentiment analysis into account, the system is customized, for example, by providing more detailed suggestions if the user is "dissatisfied."

[1855] Proposal generation

[1856] Once a proposal is approved, the server selects a proposal template based on the sentiment analysis results and automatically generates the proposal. This proposal includes basic company information, anticipated needs, details of the proposed products or services, benefits of implementation, and implementation track record with similar companies. The final proposal is saved in PDF format and prepared for transmission to the terminal.

[1857] Terminal processing

[1858] User input

[1859] The terminal displays a screen for the user to input company information. The sales representative (user) enters basic information about the company they are scheduled to visit. For example, they might enter information such as "a company in the manufacturing industry with 5,000 employees."

[1860] Data transmission

[1861] The entered information is sent from the terminal to the server.

[1862] Receiving and displaying proposals

[1863] A proposal generated from the server is sent to the terminal and displayed to the user. The proposal details solutions that address anticipated needs.

[1864] Proposal customization

[1865] The device also provides editing capabilities, allowing users to customize proposals. For example, users can fine-tune the layout of the proposal or add comments.

[1866] Sending a proposal

[1867] The customized proposal is sent to the client. It is sent using email or cloud sharing features.

[1868] Specific example

[1869] Example 1: Proposal of IT solutions for the manufacturing industry

[1870] For a manufacturing company, a user (sales representative) inputs company information into a terminal. The terminal sends this information to a server, which analyzes past order data and market trends to predict that improvements to the manufacturing process are needed. Furthermore, an emotion engine detects a "somewhat disappointed" emotion from the user's input comments. The server selects our IoT sensor system as a proposal and generates a proposal document based on it. The generated proposal document is sent to the terminal, where the user customizes it and then sends it to the company.

[1871] Example 2: Proposal of environmental protection products for the food industry

[1872] For companies in the food industry, users input company information into a terminal. Based on this information, the server predicts that there is a growing need for food safety and environmental measures. Furthermore, an emotion engine detects a "highly interested" sentiment from the user's input comments. The server selects our eco-friendly packaging materials as a suggestion and generates a proposal based on this. The proposal is sent to the terminal, where the user customizes it to increase satisfaction and finally sends it to the company.

[1873] This system enables sales representatives to quickly and effectively make appropriate proposals to clients, resulting in increased efficiency in sales activities, improved customer satisfaction, and more personalized proposals that reflect user emotions.

[1874] The following describes the processing flow.

[1875] Step 1:

[1876] The server collects general information about corporate customers from internal databases and external APIs. This general information includes industry, company size, main business activities, and location.

[1877] Step 2:

[1878] The server collects past order data from its internal database. This data includes information such as the products and services purchased by the customer, the transaction amount, the date and time of the transaction, and the terms and conditions of the transaction.

[1879] Step 3:

[1880] The server preprocesses the collected general information and order performance data. Specifically, it performs tasks such as imputing missing values, correcting incorrect data, filtering out unnecessary information, and normalizing data in different formats.

[1881] Step 4:

[1882] The server applies a pre-trained machine learning model based on pre-processed data to predict future needs. For example, it might predict that a large manufacturing company is likely to implement IT infrastructure in the future.

[1883] Step 5:

[1884] The server searches the database for solutions that address anticipated needs. For example, it extracts our products and services, such as cloud services and security solutions, as potential suggestions.

[1885] Step 6:

[1886] The server analyzes the text data entered by the user using an emotion engine to recognize emotions. For example, it identifies emotions such as "satisfied" or "dissatisfied" from the user's input comments.

[1887] Step 7:

[1888] The server combines emotion recognition results with needs predictions to select the optimal solution. Based on the emotion recognition results, it fine-tunes the content of the proposal and the selection of templates.

[1889] Step 8:

[1890] The server automatically generates a proposal based on the selected solution. The proposal includes company information, anticipated needs, details of the proposed products and services, benefits of implementation, and past success stories.

[1891] Step 9:

[1892] The server sends the generated proposal to the terminal. The proposal is saved and sent in PDF or Word format.

[1893] Step 10:

[1894] The user enters information about the company they plan to visit into an input form displayed on the terminal screen. This information includes the company name, industry, and past transaction history.

[1895] Step 11:

[1896] The terminal sends the entered company information to the server. The data is securely transferred using an API endpoint.

[1897] Step 12:

[1898] The terminal receives the proposal sent from the server and displays it in a format that is easy for the user to understand.

[1899] Step 13:

[1900] Users review the received proposals and customize them as needed. For example, they can add comments to the proposal or adjust the layout.

[1901] Step 14:

[1902] The user sends the customized proposal to the client. The proposal is sent via email or cloud sharing.

[1903] Specific example

[1904] Example 1: Proposal of IT solutions for the manufacturing industry

[1905] Step 1: The server collects general information about major manufacturing customers from a database and external APIs.

[1906] Step 2: The server collects the customer's past order history data.

[1907] Step 3: The server preprocesses the collected data, performing missing value imputation and data normalization.

[1908] Step 4: The server applies a machine learning model based on the pre-processed data to predict that the customer is likely to need IT infrastructure in the future.

[1909] Step 5: The server searches the database for IT infrastructure-related solutions.

[1910] Step 6: The server analyzes the comments entered by the user on the device using an emotion engine and detects the emotion "slightly anxious".

[1911] Step 7: Based on emotion recognition and needs prediction, the server selects a detailed IT infrastructure proposal.

[1912] Step 8: The server automatically generates a proposal and customizes the content to reflect emotions.

[1913] Step 9: Send the completed proposal to your device.

[1914] Step 10: The user enters the company information of the company they are visiting into the terminal.

[1915] Step 11: The terminal sends information to the server.

[1916] Step 12: Receive proposals sent from the server and display them to the user.

[1917] Step 13: The user reviews the proposal and makes customizations as needed.

[1918] Step 14: The user sends the final proposal to the client.

[1919] Example 2: Proposal of environmental protection products for the food industry

[1920] Step 1: The server collects customer information from the food industry.

[1921] Step 2: The server collects past transaction history.

[1922] Step 3: The server performs data imputation and normalization.

[1923] Step 4: The server predicts future needs using a trained machine learning model. It predicts that the need for safety and environmental measures will increase.

[1924] Step 5: Search the database for related products such as eco-friendly packaging materials.

[1925] Step 6: The server detects a "very interested" sentiment from the user's input comments.

[1926] Step 7: Based on the emotion recognition results, select detailed proposals for eco-friendly packaging materials.

[1927] Step 8: Automatically generate a proposal that reflects emotions.

[1928] Step 9: Send the proposal to your device.

[1929] Step 10: The user enters company information for a food industry company.

[1930] Step 11: The terminal sends information to the server.

[1931] Step 12: Receive proposals from the server and display them to the user.

[1932] Step 13: The user customizes the proposal.

[1933] Step 14: Send the final proposal to the client.

[1934] This system allows sales representatives to not only make optimal proposals to clients quickly and effectively, but also to make personalized proposals based on the user's emotions.

[1935] (Example 2)

[1936] 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".

[1937] Traditional proposal systems for corporate clients predicted customer needs based on general information and past order data, but they were insufficient in considering user emotions. As a result, customer satisfaction decreased, and proposals lacked personalization. Furthermore, even in automated proposal generation, the selection of templates was uniform, making it difficult to respond flexibly to reflect user emotions. There is a need to solve these problems and provide more accurate and personalized proposals.

[1938] 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.

[1939] In this invention, the server includes means for collecting general information of corporate customers, means for collecting past order performance data, means for pre-processing the above general information and order performance data, means for applying a machine learning model based on the pre-processed data to predict future needs, means for recognizing emotions based on user input data, means for proposing solutions corresponding to the predicted needs and the results of emotion recognition, means for automatically generating a proposal document based on the proposed solutions, and means for transmitting the generated proposal document to a terminal. This enables personalized proposals that take into account the user's emotions, thereby improving customer satisfaction and streamlining sales activities.

[1940] "Corporate clients" refer to customers other than individuals, such as companies and organizations.

[1941] "General information" refers to basic attribute information about corporate clients, including the number of employees, location, and main business activities.

[1942] "Order performance data" refers to past transaction history and order-related data with corporate clients.

[1943] "Preprocessing" refers to the process of formatting collected data to make it analyzable, supplementing incomplete data, and unifying data in different formats.

[1944] A "machine learning model" refers to an algorithm that learns certain patterns and rules based on data and uses that knowledge to make predictions and classifications on new data.

[1945] "Emotion recognition" refers to the process of identifying and specifying emotions based on user input data.

[1946] A "solution" refers to a specific product or service, or a combination thereof, proposed to address anticipated needs or problems.

[1947] A "proposal" is a document that details the proposed solution, including basic company information, anticipated needs, proposed content, and its benefits.

[1948] A "terminal" refers to a hardware device used by a user to interact with a system, and includes personal computers, tablets, smartphones, and other similar devices.

[1949] Modes for carrying out the invention

[1950] This invention relates to a system that combines a system for accurately predicting the needs of corporate clients and proposing corresponding solutions with an emotion engine that recognizes user emotions. It primarily functions through the interaction of a server, terminal, and user to perform processes such as data collection, data preprocessing, data analysis, emotion recognition, solution proposal, and automatic generation and transmission of proposal documents.

[1951] Server Processing

[1952] The server first collects general information and past order history data for corporate customers from its internal database and external APIs. For example, for a manufacturing company, it retrieves information such as the number of employees, main product lines, location, and past transaction history from the "customer_db" database and the "https: / / api.example.com / company-data" API.

[1953] Next, the server preprocesses the collected data. It uses algorithms (such as Python's Pandas library) to impart incomplete data and fills missing values ​​with the mean. It unifies data in different formats (e.g., date formatting) and filters out unnecessary information.

[1954] Based on the pre-processed data, the server applies a pre-trained machine learning model (such as "future_needs_model.pkl") to predict future needs. For example, it might predict that a certain manufacturing company is likely to implement IT infrastructure in the future.

[1955] Furthermore, the server uses an emotion engine to recognize emotions based on user input data. Using natural language processing (NLP) techniques, it assigns emotion labels such as "satisfied" or "dissatisfied" to the text data provided by the user.

[1956] Next, the server searches the database (e.g., "solutions_db") for the optimal solution based on the predicted needs and sentiment recognition results. For example, products and services such as cloud services and security solutions may be extracted as suggested options. Taking the sentiment analysis results into consideration, more detailed suggestions are made if the user indicates dissatisfaction.

[1957] Once a proposal is approved, the server selects a proposal template (such as "template1.html") based on the sentiment analysis results and automatically generates the proposal. This proposal includes basic company information, predicted needs, details of the proposed products or services, implementation benefits, and implementation track record of similar companies. The final proposal is saved in PDF format (through the "pdfkit" library) and ready to be sent to the terminal.

[1958] Terminal processing

[1959] The terminal displays a screen for the user to input company information. The sales representative (user) enters basic information about the company they are scheduled to visit (e.g., manufacturing industry, 5,000 employees) into an HTML form. The entered information is sent from the terminal to the server in JSON format.

[1960] A proposal generated from the server is sent to the terminal and displayed to the user. The proposal details solutions that address anticipated needs. Users can further customize the proposal using the provided editing functions. For example, they can fine-tune the layout using a WYSIWYG editor or add comments.

[1961] The customized proposal is sent to the client via email or cloud sharing.

[1962] Specific example

[1963] Example 1: Proposal of IT solutions for the manufacturing industry

[1964] For a manufacturing company, a user (sales representative) inputs company information into a terminal. The terminal sends this information to a server, which analyzes past order data and market trends to predict that improvements to the manufacturing process are needed. Furthermore, an emotion engine detects a "somewhat disappointed" emotion from the user's input comments. The server selects our IoT sensor system as a proposal and generates a proposal document based on it. The generated proposal document is sent to the terminal, where the user customizes it and then sends it to the company.

[1965] Example 2: Proposal of environmental protection products for the food industry

[1966] For companies in the food industry, users input company information into a terminal. Based on this information, the server predicts that there is a growing need for food safety and environmental measures. Furthermore, an emotion engine detects a "highly interested" sentiment from the user's input comments. The server selects our eco-friendly packaging materials as a suggestion and generates a proposal based on this. The proposal is sent to the terminal, where the user customizes it to increase satisfaction and finally sends it to the company.

[1967] Examples of prompts for generative AI models

[1968] "Please describe the system's overview, which predicts the needs of corporate clients and proposes appropriate solutions. Please also provide details on the emotion recognition component based on user input data."

[1969] This prompt statement allows you to request specific explanations from the generated AI model.

[1970] This system enables sales representatives to quickly and effectively make appropriate proposals to clients, resulting in increased sales efficiency and improved customer satisfaction. It also allows for more personalized proposals that reflect user emotions.

[1971] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1972] Step 1:

[1973] The server collects general information about corporate customers from an internal database (customer_db) and an external API (https: / / api.example.com / company-data). Specifically, the server uses an API key to access the external API and retrieves data in JSON format. The input is the API endpoint and authentication information, and the output is general information data about corporate customers.

[1974] Step 2:

[1975] The server collects past order data from its internal database. It establishes a database connection and executes SQL queries such as "SELECT FROM orders WHERE customer_id = ?". The input is the customer ID, and the output is past order data.

[1976] Step 3:

[1977] The server preprocesses the collected general information and order history data. Specifically, it uses the Pandas library to impute missing values ​​with the mean, correct inaccuracies in the data, and filter out unnecessary information. The input is raw data, and the output is preprocessed, clean data.

[1978] Step 4:

[1979] The server uses preprocessed data to apply machine learning models and predict future needs. For example, it loads "future_needs_model.pkl" and uses the Scikit-learn library to predict future demand. The input is preprocessed data, and the output is the predicted needs.

[1980] Step 5:

[1981] The server recognizes emotions based on user input data. Using an NLP library (e.g., SpaCy), it identifies emotions from the input text and assigns emotion labels such as "satisfied" or "dissatisfied." The input is user text data, and the output is emotion labels.

[1982] Step 6:

[1983] The server searches for appropriate solutions based on predicted needs and sentiment recognition results. It uses SQL queries to retrieve the best solutions from the database (solutions_db) and extracts them in list format. The input is the predicted needs and sentiment labels, and the output is a list of suggested solutions.

[1984] Step 7:

[1985] The server automatically generates a proposal based on the proposed solution. Using a template engine (such as Jinja), it embeds data into the selected template to generate the final proposal. PDFkit is used to convert it to PDF format. The input is the proposed solution and template, and the output is a PDF file of the proposal.

[1986] Step 8:

[1987] The server sends the generated proposal to the user's terminal. It sends a PDF file as an HTTP response, which the user receives. The input is the generated PDF file, and the output is the transmission to the user's terminal.

[1988] Step 9:

[1989] The terminal displays the proposal to the user. It opens a PDF viewer to display the received PDF file. The input is the proposal PDF file, and the output is the proposal that the user can view.

[1990] Step 10:

[1991] The user customizes the proposal. They use the editing functions provided by the terminal (such as a WYSIWYG editor) to modify text and change the layout. The input is the received proposal and the user's edits, and the output is the customized proposal.

[1992] Step 11:

[1993] The user sends the customized proposal to the client. The proposal is sent via email or cloud sharing. The input is the customized proposal, and the output is the document sent to the client.

[1994] (Application Example 2)

[1995] 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".

[1996] Traditional corporate client solution proposal systems provided proposals based on predicted customer needs, but lacked personalization of proposals and customization of proposals to consider customer emotions. Therefore, it was difficult to improve customer satisfaction and maximize the effectiveness of proposals. In the security services sector in particular, there is a need to provide more accurate proposals that take customer emotions into consideration.

[1997] 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.

[1998] In this invention, the server includes means for collecting general information of corporate customers, means for collecting past order performance data, means for pre-processing the above general information and order performance data, means for applying a machine learning model based on the pre-processed data to predict future needs, means for recognizing emotions from user input data, means for proposing solutions based on the predicted needs and recognized emotions, means for automatically generating a proposal document based on the proposed solutions, and means for transmitting the generated proposal document to a terminal. This enables more personalized proposals that take into account the predicted needs and emotions of corporate customers.

[1999] A "corporate customer" refers to a specific corporation or company that is targeted as a recipient of products or services.

[2000] "General information" refers to basic attribute information of a company, including data such as industry, number of employees, and location.

[2001] "Order history data" refers to information about past transactions and contracts, specifically including historical data such as transaction amount, item, and transaction date.

[2002] "Preprocessing" refers to the process of preparing collected data into an analyzable format, and includes operations such as data cleaning, imputation of missing values, and normalization.

[2003] A "machine learning model" refers to an algorithm or model used to predict future outcomes based on data analysis, and is one that has learned past patterns using training data.

[2004] "Needs forecasting" is the process of predicting the products and services that corporate customers may need in the future, based on collected data and machine learning models.

[2005] "Emotion recognition" refers to the process of analyzing emotions from user input data and expressing the results numerically or in classifications, using technologies such as natural language processing.

[2006] "Solution proposal" refers to the act of selecting the most suitable products and services based on anticipated needs and perceived emotions, and proposing them to corporate clients.

[2007] "Automatic proposal generation" refers to the process of automatically creating a proposal based on a solution proposal, using templates to combine content.

[2008] "Sending to a device" refers to the process of electronically sending an automatically generated proposal to a device, specifically using methods such as email or cloud sharing.

[2009] This invention combines a system that accurately predicts the needs of corporate clients and proposes corresponding solutions with an emotion engine that recognizes user emotions. This maximizes the effectiveness of solution proposals to corporate clients.

[2010] Server Processing

[2011] The server performs various processes using the following methods.

[2012] 1. Data Collection

[2013] The server collects general information and past order history data of corporate customers from internal databases and external APIs. The collected data includes basic company data and transaction history.

[2014] 2. Data preprocessing

[2015] The collected data is cleaned and normalized using Pandas. Missing values ​​are imputed and the data is standardized, preparing it for analysis.

[2016] 3. Data Analysis

[2017] Based on pre-processed data, a pre-trained machine learning model (TensorFlow or PyTorch) is applied to predict future needs.

[2018] 4. Emotion recognition

[2019] The server uses the Google Cloud Natural Language API to analyze emotions from text entered by the user. This allows it to quantify and extract emotions such as "satisfied" or "dissatisfied."

[2020] 5. Solution Proposal

[2021] Based on predicted needs and sentiment analysis results, the system proposes the most suitable products and services from its solution database. This includes customization, such as providing detailed suggestions if the sentiment is "dissatisfied."

[2022] 6. Proposal generation

[2023] Based on the proposed content, a proposal document will be automatically generated. The proposal document will be generated using a selection of multiple templates and will include the proposed content, company information, and implementation benefits. It will be generated and saved in PDF format.

[2024] Terminal processing

[2025] The device interacts with the user using the following means:

[2026] 1. User input

[2027] The terminal provides an interface for entering corporate customer information. The user enters basic information about the co...

Claims

1. Means of collecting general information from corporate customers, Methods for collecting past order performance data, A means for pre-processing the above general information and order performance data, A means of applying a machine learning model based on preprocessed data to predict future needs, A means of proposing solutions that address anticipated needs, A means of automatically generating a proposal based on the proposed solution, A means for sending the generated proposal to a terminal, A system that includes this.

2. The system according to claim 1, wherein the machine learning model is trained based on past order performance data.

3. The system according to claim 1, wherein the aforementioned proposal is generated based on a template selected from a plurality of templates.

Images