System

Abstract

A system is provided.SOLUTION: A system comprising: means for obtaining basic information of a counterparty; means for collecting additional counterparty data from external data sources; means for cleaning and consolidating the collected data; means for calculating a counterparty credit risk score; means for visually displaying the calculated credit risk score; and means for generating strategic advice based on market data and competitive analysis results.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] Traditional corporate credit operations and strategic decision-making require significant time and cost, placing a heavy burden on both small and medium-sized enterprises (SMEs) and large corporations. Lack of resources and expertise makes it difficult for SMEs to effectively conduct credit operations, making it difficult to assess client risk and determine funding. Large corporations, too, face the challenge of processing massive amounts of data and complex organizational structures, which require efficient business operations and rapid decision-making. Furthermore, the lack of a system that comprehensively forecasts future market trends, analyzes competitors, and provides strategic advice makes it difficult for companies to determine their business direction. Given these circumstances, there is a need for a system that streamlines data-driven decision-making and provides reliable information. [Means for solving the problem]

[0005] The present invention provides a system that includes a means for acquiring basic information about counterparty companies, a means for collecting additional counterparty data from external data sources, a means for cleaning and integrating the collected data, a means for calculating a counterparty credit risk score, a means for visually displaying the calculated credit risk score, and a means for generating strategic advice based on market data and competitive analysis results. The system further includes a means for customizing strategic advice based on a user's industry profile and a means for collecting and analyzing user feedback and updating the model. This streamlines corporate credit operations and strategic decision-making, and provides accurate and timely information. This can solve the challenges faced by small and medium-sized enterprises and large corporations, contributing to business growth and improved competitiveness.

[0006] "Basic information about business partners" refers to company identification and basic business information such as the name, address, industry, and annual sales of the business partner.

[0007] "External data source" refers to a data source managed by a third party other than the user, such as an external public database or API.

[0008] "Business partner data" refers to financial information, past transaction history, credit information, etc., regarding business partners.

[0009] "Data cleaning" refers to the process of removing missing or inaccurate data from collected data to improve the quality of the data.

[0010] "Data integration" refers to the process of combining data obtained from multiple data sources into one and managing it in a consistent manner.

[0011] A "credit risk score" is a numerical indicator of a trading partner's ability to repay and creditworthiness.

[0012] "Visual display means" refers to a method of presenting analysis results and data to users in a visually easy-to-understand format, such as graphs and charts.

[0013] "Market data" refers to information about a particular market or industry that indicates trends and movements.

[0014] "Competitive analysis results" refers to information that summarizes the results of analyzing competitors' business situations and strategies.

[0015] "Strategic advice" refers to proposals based on collected data and analysis results to guide corporate decision-making and action plans.

[0016] An "industry profile" refers to the characteristics and trends of a particular industry and attribute information of a group of companies.

[0017] "Feedback" refers to information such as evaluations and opinions provided by users, and actual operational results.

[0018] "Means for updating the model" refers to methods for improving analytical models and algorithms based on collected feedback and reflecting this in subsequent processing. [Brief explanation of the drawings]

[0019] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6]FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0020] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0021] First, the terms used in the following description will be explained.

[0022] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0023] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0024] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0025] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0026] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0027] [First embodiment]

[0028] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0029] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0030] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0031] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0032] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0033] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0034] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0035] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0036] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0037] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0038] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0039] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0040] This invention relates to "Business AI Navigator," a new AI data service for businesses. It is a system that streamlines credit approval procedures for clients and supports strategic decision-making. A specific embodiment of this system is shown below.

[0041] Specific program behavior

[0042] 1. Data Collection

[0043] User: Enter basic information about the business partner into the terminal, such as the company name, address, industry, and annual sales.

[0044] Terminal: Sends the information entered by the user to the server.

[0045] Server: Based on the received information, collects additional customer data (e.g., financial information and transaction history) from external data sources.

[0046] 2. Data Analysis

[0047] Server: Performs data cleaning based on the collected data, such as filling in missing values, removing inaccurate data, and standardizing formats.

[0048] Server: Consolidates the cleaned data and calculates the credit risk score for each counterparty, using machine learning algorithms and statistical models.

[0049] 3. Displaying the results

[0050] Server: Visualizes the calculated credit risk scores and generates graphs and dashboards.

[0051] Terminal: Presents the generated visualization data to the user for easy understanding.

[0052] 4. Generating strategic advice

[0053] Server: Generates strategic advice based on market data and competitive analysis, such as whether to enter new markets or improve contract terms with business partners.

[0054] Server: Customizes the generated strategic advice based on the user's industry profile.

[0055] 5. Providing advice

[0056] Server: Sends customized advice to the device.

[0057] On the device: Provide user-visible advice and list required action items.

[0058] 6. Incorporating feedback

[0059] User: Enters feedback and results on the advice provided into the device.

[0060] Terminal: Sends user feedback to the server.

[0061] Server: Updates the analysis model based on the feedback, improving the accuracy of the next analysis.

[0062] Specific examples

[0063] Example 1: Credit decision

[0064] User: A small business owner considering signing a new contract with a business partner. The user uses the device to enter basic information about the business partner (company name, address, etc.).

[0065] Terminal: Sends the entered information to the server.

[0066] Server: Collects additional financial data and transaction history from external data sources, performs data cleaning and analysis, calculates credit risk scores, and visualizes the results.

[0067] Terminal: Presents visualized risk assessment results to management and provides strategic advice on whether to proceed with a transaction and risk countermeasures.

[0068] Example 2: Market trend forecasting and strategic advice

[0069] User: A marketing manager from a large company looking to enter a new market. He enters the market conditions of interest into the terminal.

[0070] Terminal: Sends entered market conditions to the server.

[0071] Server: Predicts future market trends based on market data and competitive analysis, and generates strategic advice customized based on the user's industry profile.

[0072] Terminal: Presents strategic advice and relevant action items to the Marketing Manager.

[0073] In this way, Business AI Navigator is a system that supports corporate credit operations and strategic decision-making by providing efficient and accurate information.

[0074] The processing flow will be explained below.

[0075] Step 1:

[0076] The user enters basic information about the business partner (company name, address, industry, annual sales, etc.) into the terminal.

[0077] Step 2:

[0078] The terminal formats the basic information entered by the user and sends it to the server.

[0079] Step 3:

[0080] Based on the basic information received, the server collects additional customer data from external data sources, such as financial information, past transaction history, credit information, etc.

[0081] Step 4:

[0082] The server performs data cleaning on the collected raw data, such as filling in missing values, removing inaccurate data, and standardizing the format.

[0083] Step 5:

[0084] The server integrates the cleaned data and manages it as a single dataset.

[0085] Step 6:

[0086] The server uses machine learning algorithms and statistical models to calculate the credit risk score of the client based on the integrated data.

[0087] Step 7:

[0088] The server visualizes the calculated credit risk score and related analytical results by generating graphs and charts in a visually easy-to-understand format.

[0089] Step 8:

[0090] The server sends the visualized data to the terminal so that it can be presented to the user in an easy-to-understand format.

[0091] Step 9:

[0092] The terminal displays the visualized data sent from the server to the user, allowing the user to easily understand the data.

[0093] Step 10:

[0094] The server generates strategic advice based on market data and competitive analysis, such as whether to enter a new market or suggest improvements to contract terms.

[0095] Step 11:

[0096] The server customizes the generated strategic advice based on the user's industry profile.

[0097] Step 12:

[0098] The server sends customized strategy advice to the terminal.

[0099] Step 13:

[0100] The device displays strategic advice and related action items to the user, helping them create a concrete action plan.

[0101] Step 14:

[0102] The user inputs feedback on the provided advice into the terminal, including actual operational results and evaluations.

[0103] Step 15:

[0104] The terminal transmits the feedback from the user to the server.

[0105] Step 16:

[0106] The server updates the analysis model based on the collected feedback to improve the accuracy of future analyses. Model training is performed to reflect new data.

[0107] Example 1

[0108] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0109] Conventional credit decision-making systems had problems with complex credit risk assessment of counterparties and time-consuming data integration and cleaning. They also lacked a mechanism for effectively utilizing market data and competitive analysis results necessary for strategic decision-making. Furthermore, there was an insufficient means for appropriately collecting user feedback and incorporating it into the analysis model, which could lead to a decline in analysis accuracy. There was a need to solve these problems and provide information efficiently and accurately.

[0110] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0111] In this invention, the server includes means for acquiring basic information about counterparty companies, means for collecting additional counterparty data from external data sources, means for cleaning and integrating the collected data, means for calculating counterparty credit risk scores, means for visually displaying the calculated credit risk scores, means for generating strategic advice based on market data and competitive analysis results, means for customizing the generated advice and presenting it on a terminal, means for collecting user feedback to update the analytical model, and means for securely transmitting the feedback, thereby enabling more efficient and accurate credit risk assessment and strategic decision support.

[0112] "Basic information about a business partner" refers to basic identifying information about the business partner, such as company name, location, industry, and annual sales.

[0113] "External Data Sources" refers to external databases or information services that provide additional information about business partners.

[0114] "Data cleaning" refers to the process of correcting, completing, and standardizing missing values, inaccurate data, and improper formats from collected data.

[0115] "Synthesis" refers to the process of bringing together data collected from different sources into a single data set.

[0116] "Credit risk score" refers to a numerical evaluation value of a trading partner's credit risk, and is used to evaluate the trading partner's creditworthiness.

[0117] "Visually displaying" refers to displaying credit risk scores and other analytical results in a visual format, such as a graph or dashboard.

[0118] "Market data" refers to information regarding market movements, trends, the competitive environment, etc.

[0119] "Competitive analysis results" refers to the analysis results regarding competitors' strategies, performance, market share, etc.

[0120] "Strategic advice" refers to providing advice on trading strategies with business partners and entry into new markets based on market data and competitive analysis results.

[0121] An "industry profile" refers to information that compiles standard characteristics, trends, trade practices, etc. related to a particular industry.

[0122] "Gathering feedback" refers to collecting information such as opinions, ratings, and results from users.

[0123] "Updating the analytical model" refers to readjusting machine learning algorithms and statistical models based on newly collected feedback to improve prediction and analytical accuracy.

[0124] "Securely" refers to measures to ensure safe communications and data processing and prevent information leaks and unauthorized access.

[0125] This invention relates to "Business AI Navigator," a new AI data service for businesses, which is a system that streamlines credit approval operations for clients and supports strategic decision-making. This system is implemented using the following hardware and software.

[0126] Hardware and software used

[0127] Server: Data processing and storage are performed using cloud services (e.g., Amazon Web Services, Microsoft® Azure®).

[0128] Device: A desktop PC, laptop, tablet, etc. A web browser (examples: GOOGLE CHROME (registered trademark), Firefox) is used as the user interface.

[0129] Software: Data visualization tools (e.g., Tableau, Power BI) and machine learning algorithms (e.g., Scikit-learn, TENSORFLOW (registered trademark)) are used.

[0130] System Operation Overview

[0131] 1. Data Collection

[0132] User: Enter basic information about a business partner into the terminal. For example, enter the company name "ABC Co., Ltd.", address "Chiyoda-ku, Tokyo," industry "manufacturing," and annual sales of "10 billion yen."

[0133] Terminal: Sends the entered information to the server. The terminal uses a secure communication protocol (HTTPS).

[0134] Server: Collect additional counterparty data (financial information, transaction history, etc.) from external data sources, using APIs.

[0135] 2. Data Analysis

[0136] Server: Cleans the collected data, imputes missing values, removes inaccurate data, and standardizes the format.

[0137] Server: Integrates clean data and calculates credit risk scores for clients. Random forest and logistic regression models are used. Machine learning algorithms include Scikit-learn and TensorFlow.

[0138] 3. Displaying the results

[0139] Server: Visualize the calculated credit risk scores and generate graphs and dashboards using Tableau or Power BI as data visualization tools.

[0140] Terminal: Presents the generated visualization data to the user in an easy-to-understand format.

[0141] 4. Generating strategic advice

[0142] Server: Generates strategic advice based on market data and competitive analysis results, such as proposing whether to enter a new market or improve contract terms with business partners.

[0143] Server: Customizes generated strategic advice based on the user's industry profile.

[0144] 5. Providing advice

[0145] Server: Sends customized advice to the device.

[0146] Terminal: Provides advice to the user and lists required action items.

[0147] 6. Incorporating feedback

[0148] User: Enters feedback and results on the advice provided into the device.

[0149] Terminal: Sends user feedback to the server.

[0150] Server: Updates the analysis model based on the feedback, improving the accuracy of the next analysis.

[0151] Specific examples

[0152] Example 1: Credit decision

[0153] User: A small business owner considering signing a new contract with a business partner. He enters basic information about the business partner into the terminal.

[0154] Terminal: Sends the entered information to the server.

[0155] Server: Collects additional financial data and transaction history from external data sources, performs data cleaning and analysis, calculates credit risk scores, and visualizes the results.

[0156] Terminal: Presents visualized risk assessment results and provides strategic advice on whether to trade and how to mitigate risks.

[0157] Example 2: Market trend forecasting and strategic advice

[0158] User: A marketing manager from a large company looking to enter a new market. He enters the market conditions of interest into the terminal.

[0159] Terminal: Sends entered market conditions to the server.

[0160] Server: Predicts future market trends based on market data and competitive analysis, and generates strategic advice customized based on the user's industry profile.

[0161] Terminal: Presents strategic advice and relevant action items to the Marketing Manager.

[0162] Prompt Sentence Examples

[0163] "Based on the company information below, please assess the credit risk of your business partner and provide strategic advice."

[0164] 1. Company Name: ABC Co., Ltd.

[0165] 2. Address: Chiyoda-ku, Tokyo

[0166] 3. Industry: Manufacturing

[0167] 4. Annual sales: 10 billion yen

[0168] In this way, the Business AI Navigator supports corporate credit operations and strategic decision-making by providing efficient and accurate information. The above is an embodiment of the present invention.

[0169] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0170] Step 1:

[0171] Data collection

[0172] User: Enters basic information about a business partner into the terminal. The entered information is company name "ABC Co., Ltd.", address "Chiyoda-ku, Tokyo," industry "manufacturing," and annual sales of "10 billion yen."

[0173] Terminal: The input information is sent to the server using a secure communication protocol (HTTPS). The input data is transferred to the server in JSON format.

[0174] Server: Collects additional customer data (financial information, transaction history, etc.) from external data sources via API. The collected data is stored in a database in an organized format by field.

[0175] Step 2:

[0176] Data Cleaning

[0177] Server: Cleans the collected data by filling in missing values, removing inaccurate data, and standardizing date and currency formats. For example, if there is missing annual sales data, it fills in the approximate value using historical data or other relevant data. It also corrects inaccurate entries (e.g., company names with trailing spaces).

[0178] Input: The raw data collected.

[0179] Output: Cleaned consolidated data.

[0180] Step 3:

[0181] Data Integration

[0182] Server: Consolidates the cleaned data into a single dataset. This process involves combining information from multiple data sources using a common key (e.g., company ID). For example, combining financial data and transaction history based on company ID.

[0183] Input: Multiple cleaned datasets.

[0184] Output: A consolidated dataset.

[0185] Step 4:

[0186] Credit risk score calculation

[0187] Server: Calculates a credit risk score based on the integrated data. This uses a machine learning algorithm (e.g., random forest, logistic regression). The integrated dataset is input into the machine learning model, and a credit risk score is output in numerical form.

[0188] Input: Unified dataset.

[0189] Output: Credit risk score.

[0190] Step 5:

[0191] Visualizing the results

[0192] Server: Generate visualization data using the calculated credit risk scores. Use a data visualization tool (e.g., Tableau, Power BI) to convert the data into a format that is easy to display as graphs or dashboards.

[0193] Terminal: Receives the visualization data sent from the server and displays it to the user through a user interface, presenting the results in the form of easy-to-understand graphs and charts.

[0194] Input: Credit Risk Score.

[0195] Output: Visualized data (graphs, dashboards).

[0196] Step 6:

[0197] Generating Strategic Advice

[0198] Server: Generates strategic advice based on market data and competitive analysis results. Using AI models, it analyzes, for example, whether to enter a new market or how to improve transaction terms, and makes specific strategic proposals.

[0199] Server: Customizes the generated advice based on the user's industry profile. Provides strategic advice suited to the user's industry characteristics.

[0200] Inputs: Market data, competitive analysis results, and your industry profile.

[0201] Output: Strategy advice.

[0202] Step 7:

[0203] Providing advice

[0204] Server: Sends customized advice to the device.

[0205] On the device: Providing advice to users and listing necessary action items, such as specific bulleted steps for entering a new market.

[0206] Enter: strategic advice.

[0207] Output: Displayed advice and action items.

[0208] Step 8:

[0209] Gathering and implementing feedback

[0210] User: Enters feedback and results on the advice provided into the device.

[0211] Terminal: Sends user feedback to the server, which then forwards the feedback to the server in an appropriate format.

[0212] Server: Updates the analysis model based on the feedback to improve the accuracy of future analyses. Retrains the machine learning model based on new feedback data.

[0213] Input: User feedback.

[0214] Output: Updated analytical model.

[0215] (Application example 1)

[0216] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0217] While existing AI data services for businesses were capable of collecting basic information on business partners and calculating credit risk, they had issues with not being able to update analytical models that take into account individualized content recommendations or user feedback. Furthermore, they were unable to provide personalized content recommendations based on users' viewing history and interests, making it difficult to improve the user experience.

[0218] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0219] In this invention, the server includes means for acquiring basic information about trading companies, means for collecting additional trading data from external sources, means for cleaning and integrating the collected data, means for assessing trading credit risk, means for visually displaying the calculated credit risk, means for generating strategic advice based on market information and competitive analysis results, means for making personalized content recommendations based on the user's viewing history and genres of interest, means for presenting the generated recommendations to the user, and means for collecting feedback on the provided recommendations and updating the analytical model. This makes it possible to improve the analytical model based on the personalized content recommendations and feedback, in addition to assessing the credit risk of trading partners.

[0220] A "trading enterprise" is a legal entity or organization that transacts business with other organizations or individuals in the course of commercial activities.

[0221] "External sources" refers to additional data that can be accessed both inside and outside the company, such as from public or private organizations or industry reports.

[0222] "Trade credit risk" is an assessment criterion that indicates the credit risk regarding the financial soundness of a trading partner and the stability of transactions in commercial activities.

[0223] "Cleaning" refers to the process of removing inaccurate data, filling in missing values, and standardizing the format prior to data analysis.

[0224] "Synthesis" refers to the process of bringing together data sets collected from different sources and converting them into a format that can be used for analysis.

[0225] "Strategic advice" refers to advice and suggestions to support business decision-making based on market information and competitive analysis results.

[0226] "Viewing history" refers to information such as the content a user has viewed in the past, the viewing time, and ratings, and is used to provide personalized services.

[0227] A "genre" is a classification of content, referring to a category that indicates a particular theme or style of film, music, book, etc.

[0228] "Feedback" refers to the ratings, opinions, and experiences you provide regarding the services and recommendations you receive.

[0229] An "analytical model" refers to a computational model that uses mathematical and statistical techniques to find patterns and make predictions or classifications based on collected data.

[0230] "Content recommendation" refers to a service that individually suggests appropriate content based on a user's interests and history information.

[0231] This system is realized through data interaction between the server, the terminal, and the user.

[0232] Data collection

[0233] 1. User: Enters information about viewing history and genres of interest into the device. For example, the user enters their recent movie viewing history and movie genres of interest (action, drama, etc.).

[0234] 2. Terminal: Sends the entered information to the server.

[0235] 3. Server: Collects additional data from external sources (e.g., viewing history and reputation data of other users).

[0236] Data analysis

[0237] 1. Server: Cleans the collected data, removing inaccurate data, filling in missing values, and standardizing the data format.

[0238] 2. Server: Train an analytics model (e.g., RandomForestClassifier) ​​based on the cleaned data. The analytics model is used to recommend personalized content based on viewing history and genre information.

[0239] Displaying the results

[0240] 1. Server: Generates graphs and dashboards to visually present the calculated recommended content.

[0241] 2. Terminal: Displays the generated visualization data and content recommendation results to the user.

[0242] Strategic Advice Generation

[0243] 1. Server: Generates personalized strategic advice based on market intelligence and competitive analysis, such as recommendations for new movies or the next series to watch.

[0244] 2. Server: Customizes strategic advice based on the user's profile.

[0245] Providing advice

[0246] 1. Server: Sends customized strategy advice to devices.

[0247] 2. On the device: Provide user-visible advice and a list of what to watch next.

[0248] Reflecting feedback

[0249] 1. User: Enters feedback into the device regarding the advice and recommended content provided.

[0250] 2. Terminal: Sends feedback information to the server.

[0251] 3. Server: Updates the analysis model based on user feedback to improve the accuracy of future content recommendations.

[0252] Examples and prompts

[0253] Example: A scenario in which a user receives recommendations for the next movie to watch based on the genre and ratings of the movies they have watched. For example, recommendations for highly rated new action movies.

[0254] Example prompt:

[0255] "Recommend a personalized list of movies based on the genres and ratings of movies the user has watched. Also provide details about the latest top-rated action movies."

[0256] This system allows users to receive personalized content quickly and accurately, improving the viewing experience, while also providing optimal recommendations by updating the analysis model based on user feedback.

[0257] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0258] Step 1:

[0259] User: The user inputs information about their viewing history and genres of interest into the device. For example, they input their recent movie viewing history and movie genres of interest (action, drama, etc.). This is the input information.

[0260] Step 2:

[0261] Device: The device sends the information entered by the user to the server, such as data about viewing history and genres of interest.

[0262] Step 3:

[0263] Server: The server collects additional data from external sources (viewing history and reputation data from other users), thereby supplementing the input information to create a comprehensive dataset.

[0264] Step 4:

[0265] Server: Cleans the collected data. Ensures consistency of the dataset by removing inaccurate data and filling in missing information. The cleaned data becomes the output information.

[0266] Step 5:

[0267] Server: Trains a generative AI model (e.g., RandomForestClassifier) ​​based on the cleaned data. This model is used to recommend personalized content based on viewing history and genre information. The trained model is the output.

[0268] Step 6:

[0269] Server: Using the trained model, the server predicts the best content recommendations for the user. This prediction is performed using input data such as viewing history and genre information. The prediction results are the output information.

[0270] Step 7:

[0271] Server: Generates graphs and dashboards to visually present the calculated recommended content. The visualized data is the output information.

[0272] Step 8:

[0273] Terminal: The terminal displays the generated visualization data and content recommendation results to the user, allowing the user to check the recommended content.

[0274] Step 9:

[0275] User: Enters feedback on the provided recommended content into the device. The feedback data becomes input information.

[0276] Step 10:

[0277] Terminal: The terminal sends the feedback from the user to the server, which passes this feedback information to the server.

[0278] Step 11:

[0279] Server: The server updates the analytical model based on user feedback, improving the accuracy of future content recommendations. The updated model becomes the output information.

[0280] This process allows users to receive personalized content quickly and accurately, improving their viewing experience. Furthermore, by updating the analysis model based on feedback, the system can consistently provide optimal recommendations.

[0281] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0282] This invention relates to a system that combines an emotion engine with "Business AI Navigator," a new AI data service for businesses. This system not only streamlines credit approval procedures for clients and supports strategic decision-making, but also provides more personalized services by recognizing users' emotions. A specific embodiment of this system is shown below.

[0283] Specific program behavior

[0284] 1. Data Collection

[0285] The user enters basic information about the business partner into the terminal, such as the company name, address, industry, and annual sales.

[0286] The terminal formats the basic information entered by the user and sends it to the server.

[0287] Based on the information received, the server collects additional customer data from external data sources, such as financial information, past transaction history, credit information, etc.

[0288] 2. Data Analysis

[0289] The server performs data cleaning on the collected raw data, such as filling in missing values, removing inaccurate data, and standardizing the format.

[0290] The server integrates the cleaned data and manages it as a single dataset.

[0291] The server uses machine learning algorithms and statistical models to calculate the credit risk score of the client based on the integrated data.

[0292] 3. Displaying the results

[0293] The server visualizes the calculated credit risk score and related analytical results, generating graphs and dashboards.

[0294] The terminal presents the generated visualization data to the user for easy understanding.

[0295] 4. Generating strategic advice

[0296] The server generates strategic advice based on market data and competitive analysis results, such as whether to enter new markets or propose improvements to contract terms.

[0297] The server customizes the generated strategic advice based on the user's industry profile.

[0298] 5. Emotion recognition

[0299] The device detects the user's emotional state in real time. For example, the emotion engine analyzes subtle changes in facial expressions and voice through the camera and microphone.

[0300] The server retrieves the recognized emotion data and adjusts the content of the strategic advice based on the emotion.

[0301] 6. Providing advice

[0302] The server sends customized strategy advice to the terminal.

[0303] The device displays strategic advice and related action items to the user, helping them create a concrete action plan.

[0304] 7. Incorporating feedback

[0305] The user inputs feedback on the provided advice into the terminal, including actual operational results and evaluations.

[0306] The terminal transmits the feedback from the user to the server.

[0307] The server updates the analysis model based on the collected feedback to improve the accuracy of future analyses. Model training is performed to reflect new data.

[0308] Specific examples

[0309] Example 1: Credit decision

[0310] User: A small business owner considering signing a new contract with a business partner. The user uses the device to enter basic information about the business partner (company name, address, etc.).

[0311] Terminal: Sends the entered information to the server.

[0312] Server: Collects additional financial data and transaction history from external data sources, performs data cleaning and analysis, calculates credit risk scores, and visualizes the results.

[0313] Terminal: Presents visualized risk assessment results to management and provides strategic advice on whether to proceed with a transaction and risk countermeasures.

[0314] Device: The emotion engine analyzes the manager's facial expressions and tone of voice, and the advice is adjusted appropriately based on the manager's emotions.

[0315] Example 2: Market trend forecasting and strategic advice

[0316] User: A marketing manager from a large company looking to enter a new market. He enters the market conditions of interest into the terminal.

[0317] Terminal: Sends entered market conditions to the server.

[0318] Server: Predicts future market trends based on market data and competitive analysis, and generates strategic advice customized based on the user's industry profile.

[0319] Terminal: The emotion engine analyzes the marketing manager's emotions in real time and reflects them in strategic advice.

[0320] Terminal: Presents strategic advice and relevant action items to the Marketing Manager.

[0321] In this way, the Business AI Navigator is a system that supports corporate decision-making by providing more personalized information and advice by taking into account the user's real-time emotions.

[0322] The processing flow will be explained below.

[0323] Step 1:

[0324] The user enters basic information about the business partner (company name, address, industry, annual sales, etc.) into the terminal.

[0325] Step 2:

[0326] The terminal formats the basic information entered by the user and sends it to the server.

[0327] Step 3:

[0328] Based on the basic information received by the server, additional customer data (such as financial information, past transaction history, credit information, etc.) is collected from external data sources.

[0329] Step 4:

[0330] The server performs data cleaning on the collected raw data, such as filling in missing values, removing inaccurate data, and standardizing the format.

[0331] Step 5:

[0332] The server integrates the cleaned data and manages it as a single dataset.

[0333] Step 6:

[0334] The server uses machine learning algorithms and statistical models to calculate the credit risk score of the client based on the integrated data.

[0335] Step 7:

[0336] The server visualizes the calculated credit risk score and related analytical results, generating graphs and dashboards.

[0337] Step 8:

[0338] The server sends the visualized data to the terminal so that the user can view it in an easily understandable format.

[0339] Step 9:

[0340] The terminal displays the visualized data sent from the server to the user, making it easy to understand.

[0341] Step 10:

[0342] The server generates strategic advice based on market data and competitive analysis results, such as whether to enter a new market or propose improvements to contract terms with business partners.

[0343] Step 11:

[0344] The server customizes the generated strategic advice based on the user's industry profile.

[0345] Step 12:

[0346] The device detects the user's emotional state in real time. For example, the emotion engine analyzes subtle changes in facial expressions and voice through the camera and microphone.

[0347] Step 13:

[0348] The server retrieves the recognized emotion data and adjusts the content of the strategic advice based on the emotion.

[0349] Step 14:

[0350] The server sends customized strategy advice to the terminal.

[0351] Step 15:

[0352] The device displays strategic advice and related action items to the user, helping them create a concrete action plan.

[0353] Step 16:

[0354] The user inputs feedback on the provided advice into the terminal, including actual operational results and evaluations.

[0355] Step 17:

[0356] The terminal transmits the feedback from the user to the server.

[0357] Step 18:

[0358] The server updates the analysis model based on the collected feedback to improve the accuracy of future analyses. Model training is performed to reflect new data.

[0359] Example 2

[0360] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0361] When companies make credit assessments or strategic decisions on their business partners, they need to collect and analyze massive amounts of data. However, existing systems require time and effort to clean and integrate data and calculate credit risk scores, making efficient decision-making difficult. Furthermore, they do not provide personalized advice that takes into account the user's emotional state, making it difficult to select the optimal course of action. To address these issues, a system is needed that can consistently perform everything from collecting business partner data to calculating credit risk scores, generating strategic advice, and even adjusting that advice based on the user's emotional state.

[0362] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0363] In this invention, the server includes means for acquiring basic information about trading partners, means for collecting additional data from external data sources, means for cleaning and integrating the collected data, means for calculating a credit risk score, means for visually displaying the calculated credit risk score, means for generating strategic advice based on market data and competitive analysis results, means for detecting a user's emotional state, and means for adjusting the strategic advice based on the detected emotional state, thereby enabling efficient cleaning and integration of data, calculation of a reliable credit risk score, and provision of personalized strategic advice that takes user emotions into consideration.

[0364] "Basic information" refers to basic information such as the business partner's company name, address, industry, and annual sales.

[0365] "External Data Source" refers to an external source from which the server can gather additional data about a business partner, such as a financial database or credit bureau.

[0366] "Data cleaning" refers to the process of filling in missing values ​​from collected data, removing inaccurate data, and standardizing the format.

[0367] "Data integration" refers to the process of integrating cleaned data and managing it as a single data set.

[0368] "Credit Risk Score" means a quantitative value calculated to assess the credit risk of a counterparty.

[0369] "Visually displaying" refers to visually representing the calculated credit risk score and related analytical results in the form of graphs, dashboards, etc.

[0370] "Generating strategic advice" refers to creating strategic advice such as whether to enter a new market or proposing improvements to contract terms based on market data and competitive analysis results.

[0371] "Emotional state detection" refers to analyzing the user's facial expressions and voice through a camera or microphone using an emotion engine to grasp the user's real-time emotional state.

[0372] "Adjusting strategic advice" refers to dynamically changing the content of strategic advice based on the detected emotional state of the user.

[0373] This invention relates to a new data analysis system for businesses, and presents an embodiment of a system called "Business AI Navigator" that streamlines credit risk assessment and strategic advice provision for business partners. This system has the ability to recognize the emotional state of the user and provide more personalized services.

[0374] Hardware and software used

[0375] Terminal: Input and display device (PC, tablet, smartphone, etc.)

[0376] Server: a high-performance computer for data processing and analysis

[0377] Database: Data storage for storing business partner information and data from external data sources

[0378] External data sources: financial databases and credit bureaus

[0379] Machine learning algorithms: Models for calculating credit risk scores (e.g., random forests, logistic regression)

[0380] Data visualization tools: D3.js, Tableau, etc.

[0381] Emotion recognition engine: Google(R) Cloud Vision API, Amazon Rekognition

[0382] Processing steps

[0383] Step 1. Data collection

[0384] The user inputs basic information about the business partner (company name, address, industry, annual turnover, etc.) into the terminal. As an example, let's say the owner of a small or medium-sized business inputs information about a business partner with whom he or she wants to start a new transaction.

[0385] The device formats the basic information entered and sends it to the server in a format such as JSON or XML, using HTTP or HTTPS as the transmission protocol.

[0386] Based on the basic information received by the server, additional customer data is collected from external data sources (e.g., financial information databases, credit bureaus).

[0387] Step 2. Data analysis

[0388] The raw data collected by the server is subjected to data cleaning, which involves filling in missing financial information, removing inaccurate data, and standardizing the format.

[0389] The server consolidates the cleaned data and manages it as a single dataset, for example by eliminating duplicate data from different data sources.

[0390] The server uses the integrated data to calculate the client's credit risk score using machine learning algorithms (e.g., random forest, logistic regression model).

[0391] Step 3. View the results

[0392] The server visualizes the calculated credit risk scores and related analytical results, generating graphs and dashboards using tools such as D3.js and Tableau.

[0393] The device presents the generated visualizations to the user in an easy-to-understand manner, such as line graphs or heat maps showing rising or falling risk scores.

[0394] Step 4. Generate strategic advice

[0395] The server generates strategic advice based on market data and competitive analysis, including recommendations on whether to enter new markets and improving contract terms.

[0396] The server customizes the generated strategic advice based on the user's industry profile, including industry-specific competitive information for users in the automotive industry.

[0397] Step 5. Emotion Recognition

[0398] The device detects the user's emotional state in real time using a camera and microphone, and facial and voice data is analyzed by an emotion engine (e.g., Google Cloud Vision API, Amazon Rekognition).

[0399] The server acquires the emotion data and adjusts the strategic advice based on the user's emotions. If the user is feeling stressed, the advice will be changed to help alleviate that stress.

[0400] Step 6. Providing advice

[0401] The server sends customized strategy advice to the device, using a secure communication protocol (e.g., HTTPS).

[0402] The device will display strategic advice and related action items to the user to help them plan specific actions, for example in the form of a to-do list.

[0403] Step 7. Incorporate feedback

[0404] The user inputs feedback (actual operational results and evaluation) on the advice into the terminal. For example, the user inputs comments such as "The advice was effective" or "Improvement is needed."

[0405] The device sends the feedback to the server, which formats it in the appropriate schema and stores it in a database.

[0406] The server analyzes the feedback and updates the analysis model to improve accuracy in future runs. Model training uses machine learning frameworks such as TensorFlow and PyTorch.

[0407] In this way, the Business AI Navigator is a system that supports corporate decision-making by taking into account users' real-time emotions and providing personalized information and advice.

[0408] Prompt Sentence Examples

[0409] "Please provide a strategy for entering a new market. The market conditions are as follows: the region is North America, the target customer base is small and medium-sized businesses, and the competitors are Company A and Company B."

[0410] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0411] Step 1:

[0412] Enter basic information about your business partner

[0413] The user enters basic information about the business partner (company name, address, industry, annual sales, etc.) into the terminal.

[0414] Input: Basic information such as company name, address, industry, annual sales, etc.

[0415] The terminal formats this information and sends it to the server.

[0416] Output: Basic formatted information sent to the server.

[0417] Step 2:

[0418] External Data Collection

[0419] Based on the basic information received by the server, additional customer data is collected from external data sources (e.g., financial databases, credit bureaus).

[0420] Input: Basic formatted information

[0421] The server uses APIs to communicate with external data sources to obtain additional data (financial information, past transaction history, credit information, etc.).

[0422] Output: The additional customer data collected is stored on the server.

[0423] Step 3:

[0424] Data Cleaning

[0425] The server performs data cleaning on the collected raw data.

[0426] Input: Additional Account Data

[0427] Data cleaning includes imputing missing values, removing inaccurate data, and standardizing formats.

[0428] Output: Cleaned data is obtained.

[0429] Step 4:

[0430] Data Integration

[0431] The server integrates the cleaned data and manages it as a single dataset.

[0432] Input: Cleaned data

[0433] Eliminate duplicate data and standardize data formats.

[0434] Output: A consolidated dataset is created.

[0435] Step 5:

[0436] Credit risk score calculation

[0437] The server uses the integrated data to calculate the client's credit risk score using machine learning algorithms (e.g., random forest, logistic regression model).

[0438] Input: Integrated dataset

[0439] An algorithm analyzes the data and calculates a credit risk score.

[0440] Output: Calculated credit risk score

[0441] Step 6:

[0442] Visualizing the results

[0443] The server visualizes the calculated credit risk score and related analytical results, generating graphs and dashboards.

[0444] Input: Credit risk scores and associated analysis results

[0445] Use tools such as D3.js or Tableau to visualize the results.

[0446] Output: Visualized data

[0447] Step 7:

[0448] Presenting Visualized Data

[0449] The terminal presents the generated visualization data to the user and displays it in an easy-to-understand manner.

[0450] Input: Visualized data

[0451] Displays line graphs and heat maps showing increases or decreases in risk scores.

[0452] Output: Visualized data presented to the user

[0453] Step 8:

[0454] Strategic Advice Generation

[0455] The server generates strategic advice based on market data and competitive analysis results.

[0456] Input: Market data, competitive analysis results

[0457] This includes whether to enter new markets and proposals for improving contract terms.

[0458] Output: Strategic advice

[0459] Step 9:

[0460] Customization

[0461] The server customizes the generated strategic advice based on the user's industry profile.

[0462] Input: Strategic advice, user industry profile

[0463] For users in the automotive industry, industry-specific competitive information is added.

[0464] Output: Customized strategy advice

[0465] Step 10:

[0466] emotion recognition

[0467] The device detects the user's emotional state in real time.

[0468] Input: User's facial expressions and voice data

[0469] Using the camera and microphone, the data is analyzed by an emotion engine (e.g., Google Cloud Vision API, Amazon Rekognition).

[0470] Output: Recognized emotion data

[0471] Step 11:

[0472] Advice adjustment

[0473] The server retrieves the recognized emotion data and adjusts the content of the strategic advice.

[0474] Input: Recognized sentiment data, customized strategy advice

[0475] If you are feeling stressed, change the advice to something that will relieve you.

[0476] Output: Tailored strategy advice

[0477] Step 12:

[0478] Providing advice

[0479] The server sends the adjusted strategy advice to the terminal.

[0480] Input: Tailored Strategy Advice

[0481] Use secure communication protocols (e.g. HTTPS).

[0482] Output: Strategy advice sent to the terminal

[0483] Step 13:

[0484] Providing advice

[0485] The device displays strategic advice and related action items to the user to assist with specific action planning.

[0486] Input: Strategy Advice

[0487] Display in ToDo list format, etc.

[0488] Output: A concrete action item presented to the user

[0489] Step 14:

[0490] Enter your feedback

[0491] The user inputs feedback (actual operational results and evaluation) on the advice into the terminal.

[0492] Input: Feedback

[0493] Record the effectiveness of the advice and areas for improvement.

[0494] Output: Input feedback data

[0495] Step 15:

[0496] Send Feedback

[0497] The device sends the feedback to the server, which formats it into the appropriate schema.

[0498] Input: Feedback data

[0499] The feedback is stored in a database.

[0500] Output: Feedback data sent to the server

[0501] Step 16:

[0502] Updating a Model

[0503] The server analyzes the feedback and updates the analysis model to improve the accuracy of the analysis from the next time onwards.

[0504] Input: Submitted feedback data

[0505] Machine learning frameworks such as TensorFlow and PyTorch are used for model training.

[0506] Output: Updated analytical model

[0507] (Application example 2)

[0508] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0509] Conventional security systems determine alert levels and countermeasures based on the target's behavior and belongings. However, because there is no mechanism for analyzing and reflecting the target's emotional state in real time, appropriate responses are often delayed, which can result in increased security risks. Furthermore, conventional systems have the problem of low response accuracy because the advice they determine is uniform and lacks personalization.

[0510] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for acquiring basic information about trading partner companies, means for collecting additional trading partner data from external data sources, means for cleaning and integrating the collected data, means for calculating trading partner credit risk scores, means for visually displaying the calculated credit risk scores, means for generating strategic advice based on market data and competitive analysis results, means for analyzing the user's real-time emotional state, and means for adjusting the strategic advice based on the analyzed emotional state. This makes it possible to analyze and reflect the user's emotional state in real time and provide personalized security countermeasures.

[0511] "Means for obtaining basic information about business partners" refers to methods for collecting basic information such as the name, address, industry, and sales of business partners.

[0512] "Means for collecting additional business partner data from external data sources" refers to methods for obtaining financial information, past transaction history, credit information, etc. of business partners from external data sources.

[0513] "Means for cleaning and integrating collected data" refers to methods for preparing collected data by filling in missing values, removing inaccurate data, and standardizing the format.

[0514] The "means for calculating a customer credit risk score" is a method for calculating a score for evaluating the credit risk of a customer based on the collected and cleaned customer data.

[0515] The "means for visually displaying the calculated credit risk score" is a method for visually displaying the counterparty credit risk score as a graph or dashboard.

[0516] The "means for generating strategic advice based on market data and competitor analysis results" is a method for generating effective strategic advice based on market data and competitor analysis results.

[0517] "Means for analyzing the user's real-time emotional state" refers to a method for analyzing the user's facial expressions and tone of voice through a camera or microphone to grasp the user's emotional state in real time.

[0518] The "means for adjusting strategic advice based on the analyzed emotional state" is a method for appropriately adjusting the strategic advice provided based on the results of the user's emotional analysis.

[0519] This invention relates to a system that combines "Security Navigator," a new AI data service for businesses, with an emotion engine to analyze real-time emotional processes and support strategic security responses. This system uses smart glasses to collect visual and audio information from the target, which is then sent to a server for analysis. The following describes in detail the mode for carrying out the invention.

[0520] Hardware Configuration

[0521] 1. Smart Glasses

[0522] Visual information collection: high-resolution cameras

[0523] Audio information collection: High-sensitivity microphone

[0524] Display: Built-in display for visual display

[0525] Processor: Initial processing and formatting of data

[0526] 2. Server

[0527] High-performance cloud servers (e.g., AWS (registered trademark), Google Cloud)

[0528] 3. Means of communication

[0529] Real-time communication: WebSocket, MQTT

[0530] Software Configuration

[0531] 1. Emotion Recognition Engine

[0532] Google Cloud Vision API

[0533] OpenAI® Sentiment Analysis Model

[0534] 2. Data Analysis Tools

[0535] Python

[0536] TensorFlow

[0537] PyTorch

[0538] Processing flow (system program)

[0539] The server implements the invention using the following means.

[0540] 1. Data Collection

[0541] Smart glasses collect visual and audio information from the wearer.

[0542] Video data is acquired through a camera, and audio data is acquired through a microphone.

[0543] This data is formatted by a processor within the smart glasses and sent to a server in real time.

[0544] 2. Data Analysis

[0545] The server analyzes the received video and audio data and recognizes the emotional state using an emotion engine.

[0546] The emotion recognition engine uses Google Cloud Vision API and OpenAI's emotion analysis model.

[0547] The analysis results are fed back visually and audibly in real time.

[0548] 3. Displaying the results

[0549] The analysis results, showing the emotional state of the target person, are displayed on the smart glasses' display.

[0550] Depending on the emotional state, a security level such as alert or normal is indicated.

[0551] 4. Generating strategic advice

[0552] The server generates optimal responses based on the emotional state.

[0553] If the subject appears suspicious, security guards will provide real-time advice on whether increased vigilance is needed or whether further investigation is required.

[0554] 5. Providing advice

[0555] Smart glasses display real-time strategic advice.

[0556] Countermeasures will also be provided through audio guidance, allowing the wearer to respond smoothly.

[0557] 6. Incorporating feedback

[0558] Staff feedback is entered into the smart glasses via voice input or a touch panel.

[0559] This feedback is sent to the server and used to update and improve the sentiment analysis model.

[0560] Specific examples

[0561] Example 1: Commercial Intruder Screening

[0562] 1. A security guard wearing smart glasses during an intruder inspection at a commercial facility.

[0563] 2. The smart glasses send visual and audio data to the server.

[0564] 3. The server recognizes through emotion analysis that the intruder is nervous and sets the alert level to high.

[0565] 4. The smart glasses display will show "High alert state" and a voice prompt will say "Prompt for fixation confirmation."

[0566] Prompt Sentence Examples

[0567] "Based on this video and audio data, please analyze the subject's emotional state and generate an appropriate response accordingly."

[0568] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0569] Step 1:

[0570] The smart glasses collect the wearer's visual and audio information. The video data is acquired through a camera, and the audio data is acquired through a microphone. The smart glasses receive the video and audio data from the camera and microphone as input, format these data, and send them to a server in real time through a processor within the smart glasses. The output is the formatted video and audio data.

[0571] Step 2:

[0572] The server analyzes the received video and audio data. As input, it receives the video and audio data sent from the smart glasses and performs data cleaning. Specifically, it removes unnecessary noise and extracts only the necessary information. As output, it obtains the cleaned video and audio data.

[0573] Step 3:

[0574] The server uses the cleaned data to analyze the emotional state using an emotion engine. It receives the cleaned video and audio data as input, and uses Google Cloud Vision API and OpenAI emotion analysis models to recognize the user's emotional state from their facial expressions and tone of voice. The output is an analysis result that represents the user's emotional state.

[0575] Step 4:

[0576] The server generates countermeasures based on the analyzed emotion results. As input, it receives the analysis results that represent the emotional state and generates optimal security countermeasures based on them. For example, if it detects a state of tension, it sets the alert level to high. The generated countermeasures are obtained as output.

[0577] Step 5:

[0578] The results and countermeasures are displayed and provided to the wearer in real time through the smart glasses' display and audio guide. Countermeasures sent from the server are received as input and displayed on the display. Specifically, instructions such as "High alert state" and "Encourage fixation confirmation" are displayed, and the same information is provided via audio guide. Countermeasures are delivered to the wearer visually and audibly as output.

[0579] Step 6:

[0580] The wearer performs the countermeasure and provides feedback on the results through the smart glasses. As input, the wearer's feedback data is received, which is collected through the touch panel or voice input. As output, the collected feedback data is sent to the server.

[0581] Step 7:

[0582] The server analyzes the received feedback data and updates the sentiment analysis model. It takes the feedback data as input and trains the model based on it, specifically using new data to improve the sentiment analysis algorithm. The output is an updated sentiment analysis model.

[0583] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0584] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0585] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0586] [Second embodiment]

[0587] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0588] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0589] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0590] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0591] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0592] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0593] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0594] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0595] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0596] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0597] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0598] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0599] This invention relates to "Business AI Navigator," a new AI data service for businesses. It is a system that streamlines credit approval procedures for clients and supports strategic decision-making. A specific embodiment of this system is shown below.

[0600] Specific program behavior

[0601] 1. Data Collection

[0602] User: Enter basic information about the business partner into the terminal, such as the company name, address, industry, and annual sales.

[0603] Terminal: Sends the information entered by the user to the server.

[0604] Server: Based on the received information, collects additional customer data (e.g., financial information and transaction history) from external data sources.

[0605] 2. Data Analysis

[0606] Server: Performs data cleaning based on the collected data, such as filling in missing values, removing inaccurate data, and standardizing formats.

[0607] Server: Consolidates the cleaned data and calculates the credit risk score for each counterparty, using machine learning algorithms and statistical models.

[0608] 3. Displaying the results

[0609] Server: Visualizes the calculated credit risk scores and generates graphs and dashboards.

[0610] Terminal: Presents the generated visualization data to the user for easy understanding.

[0611] 4. Generating strategic advice

[0612] Server: Generates strategic advice based on market data and competitive analysis, such as whether to enter new markets or improve contract terms with business partners.

[0613] Server: Customizes the generated strategic advice based on the user's industry profile.

[0614] 5. Providing advice

[0615] Server: Sends customized advice to the device.

[0616] On the device: Provide user-visible advice and list required action items.

[0617] 6. Incorporating feedback

[0618] User: Enters feedback and results on the advice provided into the device.

[0619] Terminal: Sends user feedback to the server.

[0620] Server: Updates the analysis model based on the feedback, improving the accuracy of the next analysis.

[0621] Specific examples

[0622] Example 1: Credit decision

[0623] User: A small business owner considering signing a new contract with a business partner. The user uses the device to enter basic information about the business partner (company name, address, etc.).

[0624] Terminal: Sends the entered information to the server.

[0625] Server: Collects additional financial data and transaction history from external data sources, performs data cleaning and analysis, calculates credit risk scores, and visualizes the results.

[0626] Terminal: Presents visualized risk assessment results to management and provides strategic advice on whether to proceed with a transaction and risk countermeasures.

[0627] Example 2: Market trend forecasting and strategic advice

[0628] User: A marketing manager from a large company looking to enter a new market. He enters the market conditions of interest into the terminal.

[0629] Terminal: Sends entered market conditions to the server.

[0630] Server: Predicts future market trends based on market data and competitive analysis, and generates strategic advice customized based on the user's industry profile.

[0631] Terminal: Presents strategic advice and relevant action items to the Marketing Manager.

[0632] In this way, Business AI Navigator is a system that supports corporate credit operations and strategic decision-making by providing efficient and accurate information.

[0633] The processing flow will be explained below.

[0634] Step 1:

[0635] The user enters basic information about the business partner (company name, address, industry, annual sales, etc.) into the terminal.

[0636] Step 2:

[0637] The terminal formats the basic information entered by the user and sends it to the server.

[0638] Step 3:

[0639] Based on the basic information received, the server collects additional customer data from external data sources, such as financial information, past transaction history, credit information, etc.

[0640] Step 4:

[0641] The server performs data cleaning on the collected raw data, such as filling in missing values, removing inaccurate data, and standardizing the format.

[0642] Step 5:

[0643] The server integrates the cleaned data and manages it as a single dataset.

[0644] Step 6:

[0645] The server uses machine learning algorithms and statistical models to calculate the credit risk score of the client based on the integrated data.

[0646] Step 7:

[0647] The server visualizes the calculated credit risk score and related analytical results by generating graphs and charts in a visually easy-to-understand format.

[0648] Step 8:

[0649] The server sends the visualized data to the terminal so that it can be presented to the user in an easy-to-understand format.

[0650] Step 9:

[0651] The terminal displays the visualized data sent from the server to the user, allowing the user to easily understand the data.

[0652] Step 10:

[0653] The server generates strategic advice based on market data and competitive analysis, such as whether to enter a new market or suggest improvements to contract terms.

[0654] Step 11:

[0655] The server customizes the generated strategic advice based on the user's industry profile.

[0656] Step 12:

[0657] The server sends customized strategy advice to the terminal.

[0658] Step 13:

[0659] The device displays strategic advice and related action items to the user, helping them create a concrete action plan.

[0660] Step 14:

[0661] The user inputs feedback on the provided advice into the terminal, including actual operational results and evaluations.

[0662] Step 15:

[0663] The terminal transmits the feedback from the user to the server.

[0664] Step 16:

[0665] The server updates the analysis model based on the collected feedback to improve the accuracy of future analyses. Model training is performed to reflect new data.

[0666] Example 1

[0667] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0668] Conventional credit decision-making systems had problems with complex credit risk assessment of counterparties and time-consuming data integration and cleaning. They also lacked a mechanism for effectively utilizing market data and competitive analysis results necessary for strategic decision-making. Furthermore, there was an insufficient means for appropriately collecting user feedback and incorporating it into the analysis model, which could lead to a decline in analysis accuracy. There was a need to solve these problems and provide information efficiently and accurately.

[0669] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0670] In this invention, the server includes means for acquiring basic information about counterparty companies, means for collecting additional counterparty data from external data sources, means for cleaning and integrating the collected data, means for calculating counterparty credit risk scores, means for visually displaying the calculated credit risk scores, means for generating strategic advice based on market data and competitive analysis results, means for customizing the generated advice and presenting it on a terminal, means for collecting user feedback to update the analytical model, and means for securely transmitting the feedback, thereby enabling more efficient and accurate credit risk assessment and strategic decision support.

[0671] "Basic information about a business partner" refers to basic identifying information about the business partner, such as company name, location, industry, and annual sales.

[0672] "External Data Sources" refers to external databases or information services that provide additional information about business partners.

[0673] "Data cleaning" refers to the process of correcting, completing, and standardizing missing values, inaccurate data, and improper formats from collected data.

[0674] "Synthesis" refers to the process of bringing together data collected from different sources into a single data set.

[0675] "Credit risk score" refers to a numerical evaluation value of a trading partner's credit risk, and is used to evaluate the trading partner's creditworthiness.

[0676] "Visually displaying" refers to displaying credit risk scores and other analytical results in a visual format, such as a graph or dashboard.

[0677] "Market data" refers to information regarding market movements, trends, the competitive environment, etc.

[0678] "Competitive analysis results" refers to the analysis results regarding competitors' strategies, performance, market share, etc.

[0679] "Strategic advice" refers to providing advice on trading strategies with business partners and entry into new markets based on market data and competitive analysis results.

[0680] An "industry profile" refers to information that compiles standard characteristics, trends, trade practices, etc. related to a particular industry.

[0681] "Gathering feedback" refers to collecting information such as opinions, ratings, and results from users.

[0682] "Updating the analytical model" refers to readjusting machine learning algorithms and statistical models based on newly collected feedback to improve prediction and analytical accuracy.

[0683] "Securely" refers to measures to ensure safe communications and data processing and prevent information leaks and unauthorized access.

[0684] This invention relates to "Business AI Navigator," a new AI data service for businesses, which is a system that streamlines credit approval operations for clients and supports strategic decision-making. This system is implemented using the following hardware and software.

[0685] Hardware and software used

[0686] Server: Uses cloud services (e.g. Amazon Web Services, Microsoft Azure) for data processing and storage.

[0687] Device: Desktop PC, laptop, tablet, etc. The user interface is a web browser (e.g., Google Chrome, Firefox).

[0688] Software: Data visualization tools (e.g., Tableau, Power BI) and machine learning algorithms (e.g., Scikit-learn, TensorFlow) will be used.

[0689] System Operation Overview

[0690] 1. Data Collection

[0691] User: Enter basic information about a business partner into the terminal. For example, enter the company name "ABC Co., Ltd.", address "Chiyoda-ku, Tokyo," industry "manufacturing," and annual sales of "10 billion yen."

[0692] Terminal: Sends the entered information to the server. The terminal uses a secure communication protocol (HTTPS).

[0693] Server: Collect additional counterparty data (financial information, transaction history, etc.) from external data sources, using APIs.

[0694] 2. Data Analysis

[0695] Server: Cleans the collected data, imputes missing values, removes inaccurate data, and standardizes the format.

[0696] Server: Integrates clean data and calculates credit risk scores for clients. Random forest and logistic regression models are used. Machine learning algorithms include Scikit-learn and TensorFlow.

[0697] 3. Displaying the results

[0698] Server: Visualize the calculated credit risk scores and generate graphs and dashboards using Tableau or Power BI as data visualization tools.

[0699] Terminal: Presents the generated visualization data to the user in an easy-to-understand format.

[0700] 4. Generating strategic advice

[0701] Server: Generates strategic advice based on market data and competitive analysis results, such as proposing whether to enter a new market or improve contract terms with business partners.

[0702] Server: Customizes generated strategic advice based on the user's industry profile.

[0703] 5. Providing advice

[0704] Server: Sends customized advice to the device.

[0705] Terminal: Provides advice to the user and lists required action items.

[0706] 6. Incorporating feedback

[0707] User: Enters feedback and results on the advice provided into the device.

[0708] Terminal: Sends user feedback to the server.

[0709] Server: Updates the analysis model based on the feedback, improving the accuracy of the next analysis.

[0710] Specific examples

[0711] Example 1: Credit decision

[0712] User: A small business owner considering signing a new contract with a business partner. He enters basic information about the business partner into the terminal.

[0713] Terminal: Sends the entered information to the server.

[0714] Server: Collects additional financial data and transaction history from external data sources, performs data cleaning and analysis, calculates credit risk scores, and visualizes the results.

[0715] Terminal: Presents visualized risk assessment results and provides strategic advice on whether to trade and how to mitigate risks.

[0716] Example 2: Market trend forecasting and strategic advice

[0717] User: A marketing manager from a large company looking to enter a new market. He enters the market conditions of interest into the terminal.

[0718] Terminal: Sends entered market conditions to the server.

[0719] Server: Predicts future market trends based on market data and competitive analysis, and generates strategic advice customized based on the user's industry profile.

[0720] Terminal: Presents strategic advice and relevant action items to the Marketing Manager.

[0721] Prompt Sentence Examples

[0722] "Based on the company information below, please assess the credit risk of your business partner and provide strategic advice."

[0723] 1. Company Name: ABC Co., Ltd.

[0724] 2. Address: Chiyoda-ku, Tokyo

[0725] 3. Industry: Manufacturing

[0726] 4. Annual sales: 10 billion yen

[0727] In this way, the Business AI Navigator supports corporate credit operations and strategic decision-making by providing efficient and accurate information. The above is an embodiment of the present invention.

[0728] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0729] Step 1:

[0730] Data collection

[0731] User: Enters basic information about a business partner into the terminal. The entered information is company name "ABC Co., Ltd.", address "Chiyoda-ku, Tokyo," industry "manufacturing," and annual sales of "10 billion yen."

[0732] Terminal: The input information is sent to the server using a secure communication protocol (HTTPS). The input data is transferred to the server in JSON format.

[0733] Server: Collects additional customer data (financial information, transaction history, etc.) from external data sources via API. The collected data is stored in a database in an organized format by field.

[0734] Step 2:

[0735] Data Cleaning

[0736] Server: Cleans the collected data by filling in missing values, removing inaccurate data, and standardizing date and currency formats. For example, if there is missing annual sales data, it fills in the approximate value using historical data or other relevant data. It also corrects inaccurate entries (e.g., company names with trailing spaces).

[0737] Input: The raw data collected.

[0738] Output: Cleaned consolidated data.

[0739] Step 3:

[0740] Data Integration

[0741] Server: Consolidates the cleaned data into a single dataset. This process involves combining information from multiple data sources using a common key (e.g., company ID). For example, combining financial data and transaction history based on company ID.

[0742] Input: Multiple cleaned datasets.

[0743] Output: A consolidated dataset.

[0744] Step 4:

[0745] Credit risk score calculation

[0746] Server: Calculates a credit risk score based on the integrated data. This uses a machine learning algorithm (e.g., random forest, logistic regression). The integrated dataset is input into the machine learning model, and a credit risk score is output in numerical form.

[0747] Input: Unified dataset.

[0748] Output: Credit risk score.

[0749] Step 5:

[0750] Visualizing the results

[0751] Server: Generate visualization data using the calculated credit risk scores. Use a data visualization tool (e.g., Tableau, Power BI) to convert the data into a format that is easy to display as graphs or dashboards.

[0752] Terminal: Receives the visualization data sent from the server and displays it to the user through a user interface, presenting the results in the form of easy-to-understand graphs and charts.

[0753] Input: Credit Risk Score.

[0754] Output: Visualized data (graphs, dashboards).

[0755] Step 6:

[0756] Generating Strategic Advice

[0757] Server: Generates strategic advice based on market data and competitive analysis results. Using AI models, it analyzes, for example, whether to enter a new market or how to improve transaction terms, and makes specific strategic proposals.

[0758] Server: Customizes the generated advice based on the user's industry profile. Provides strategic advice suited to the user's industry characteristics.

[0759] Inputs: Market data, competitive analysis results, and your industry profile.

[0760] Output: Strategy advice.

[0761] Step 7:

[0762] Providing advice

[0763] Server: Sends customized advice to the device.

[0764] On the device: Providing advice to users and listing necessary action items, such as specific bulleted steps for entering a new market.

[0765] Enter: strategic advice.

[0766] Output: Displayed advice and action items.

[0767] Step 8:

[0768] Gathering and implementing feedback

[0769] User: Enters feedback and results on the advice provided into the device.

[0770] Terminal: Sends user feedback to the server, which then forwards the feedback to the server in an appropriate format.

[0771] Server: Updates the analysis model based on the feedback to improve the accuracy of future analyses. Retrains the machine learning model based on new feedback data.

[0772] Input: User feedback.

[0773] Output: Updated analytical model.

[0774] (Application example 1)

[0775] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0776] While existing AI data services for businesses were capable of collecting basic information on business partners and calculating credit risk, they had issues with not being able to update analytical models that take into account individualized content recommendations or user feedback. Furthermore, they were unable to provide personalized content recommendations based on users' viewing history and interests, making it difficult to improve the user experience.

[0777] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0778] In this invention, the server includes means for acquiring basic information about trading companies, means for collecting additional trading data from external sources, means for cleaning and integrating the collected data, means for assessing trading credit risk, means for visually displaying the calculated credit risk, means for generating strategic advice based on market information and competitive analysis results, means for making personalized content recommendations based on the user's viewing history and genres of interest, means for presenting the generated recommendations to the user, and means for collecting feedback on the provided recommendations and updating the analytical model. This makes it possible to improve the analytical model based on the personalized content recommendations and feedback, in addition to assessing the credit risk of trading partners.

[0779] A "trading enterprise" is a legal entity or organization that transacts business with other organizations or individuals in the course of commercial activities.

[0780] "External sources" refers to additional data that can be accessed both inside and outside the company, such as from public or private organizations or industry reports.

[0781] "Trade credit risk" is an assessment criterion that indicates the credit risk regarding the financial soundness of a trading partner and the stability of transactions in commercial activities.

[0782] "Cleaning" refers to the process of removing inaccurate data, filling in missing values, and standardizing the format prior to data analysis.

[0783] "Synthesis" refers to the process of bringing together data sets collected from different sources and converting them into a format that can be used for analysis.

[0784] "Strategic advice" refers to advice and suggestions to support business decision-making based on market information and competitive analysis results.

[0785] "Viewing history" refers to information such as the content a user has viewed in the past, the viewing time, and ratings, and is used to provide personalized services.

[0786] A "genre" is a classification of content, referring to a category that indicates a particular theme or style of film, music, book, etc.

[0787] "Feedback" refers to the ratings, opinions, and experiences you provide regarding the services and recommendations you receive.

[0788] An "analytical model" refers to a computational model that uses mathematical and statistical techniques to find patterns and make predictions or classifications based on collected data.

[0789] "Content recommendation" refers to a service that individually suggests appropriate content based on a user's interests and history information.

[0790] This system is realized through data interaction between the server, the terminal, and the user.

[0791] Data collection

[0792] 1. User: Enters information about viewing history and genres of interest into the device. For example, the user enters their recent movie viewing history and movie genres of interest (action, drama, etc.).

[0793] 2. Terminal: Sends the entered information to the server.

[0794] 3. Server: Collects additional data from external sources (e.g., viewing history and reputation data of other users).

[0795] Data analysis

[0796] 1. Server: Cleans the collected data, removing inaccurate data, filling in missing values, and standardizing the data format.

[0797] 2. Server: Train an analytics model (e.g., RandomForestClassifier) ​​based on the cleaned data. The analytics model is used to recommend personalized content based on viewing history and genre information.

[0798] Displaying the results

[0799] 1. Server: Generates graphs and dashboards to visually present the calculated recommended content.

[0800] 2. Terminal: Displays the generated visualization data and content recommendation results to the user.

[0801] Strategic Advice Generation

[0802] 1. Server: Generates personalized strategic advice based on market intelligence and competitive analysis, such as recommendations for new movies or the next series to watch.

[0803] 2. Server: Customizes strategic advice based on the user's profile.

[0804] Providing advice

[0805] 1. Server: Sends customized strategy advice to devices.

[0806] 2. On the device: Provide user-visible advice and a list of what to watch next.

[0807] Reflecting feedback

[0808] 1. User: Enters feedback into the device regarding the advice and recommended content provided.

[0809] 2. Terminal: Sends feedback information to the server.

[0810] 3. Server: Updates the analysis model based on user feedback to improve the accuracy of future content recommendations.

[0811] Examples and prompts

[0812] Example: A scenario in which a user receives recommendations for the next movie to watch based on the genre and ratings of the movies they have watched. For example, recommendations for highly rated new action movies.

[0813] Example prompt:

[0814] "Recommend a personalized list of movies based on the genres and ratings of movies the user has watched. Also provide details about the latest top-rated action movies."

[0815] This system allows users to receive personalized content quickly and accurately, improving the viewing experience, while also providing optimal recommendations by updating the analysis model based on user feedback.

[0816] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0817] Step 1:

[0818] User: The user inputs information about their viewing history and genres of interest into the device. For example, they input their recent movie viewing history and movie genres of interest (action, drama, etc.). This is the input information.

[0819] Step 2:

[0820] Device: The device sends the information entered by the user to the server, such as data about viewing history and genres of interest.

[0821] Step 3:

[0822] Server: The server collects additional data from external sources (viewing history and reputation data from other users), thereby supplementing the input information to create a comprehensive dataset.

[0823] Step 4:

[0824] Server: Cleans the collected data. Ensures consistency of the dataset by removing inaccurate data and filling in missing information. The cleaned data becomes the output information.

[0825] Step 5:

[0826] Server: Trains a generative AI model (e.g., RandomForestClassifier) ​​based on the cleaned data. This model is used to recommend personalized content based on viewing history and genre information. The trained model is the output.

[0827] Step 6:

[0828] Server: Using the trained model, the server predicts the best content recommendations for the user. This prediction is performed using input data such as viewing history and genre information. The prediction results are the output information.

[0829] Step 7:

[0830] Server: Generates graphs and dashboards to visually present the calculated recommended content. The visualized data is the output information.

[0831] Step 8:

[0832] Terminal: The terminal displays the generated visualization data and content recommendation results to the user, allowing the user to check the recommended content.

[0833] Step 9:

[0834] User: Enters feedback on the provided recommended content into the device. The feedback data becomes input information.

[0835] Step 10:

[0836] Terminal: The terminal sends the feedback from the user to the server, which passes this feedback information to the server.

[0837] Step 11:

[0838] Server: The server updates the analytical model based on user feedback, improving the accuracy of future content recommendations. The updated model becomes the output information.

[0839] This process allows users to receive personalized content quickly and accurately, improving their viewing experience. Furthermore, by updating the analysis model based on feedback, the system can consistently provide optimal recommendations.

[0840] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0841] This invention relates to a system that combines an emotion engine with "Business AI Navigator," a new AI data service for businesses. This system not only streamlines credit approval procedures for clients and supports strategic decision-making, but also provides more personalized services by recognizing users' emotions. A specific embodiment of this system is shown below.

[0842] Specific program behavior

[0843] 1. Data Collection

[0844] The user enters basic information about the business partner into the terminal, such as the company name, address, industry, and annual sales.

[0845] The terminal formats the basic information entered by the user and sends it to the server.

[0846] Based on the information received, the server collects additional customer data from external data sources, such as financial information, past transaction history, credit information, etc.

[0847] 2. Data Analysis

[0848] The server performs data cleaning on the collected raw data, such as filling in missing values, removing inaccurate data, and standardizing the format.

[0849] The server integrates the cleaned data and manages it as a single dataset.

[0850] The server uses machine learning algorithms and statistical models to calculate the credit risk score of the client based on the integrated data.

[0851] 3. Displaying the results

[0852] The server visualizes the calculated credit risk score and related analytical results, generating graphs and dashboards.

[0853] The terminal presents the generated visualization data to the user for easy understanding.

[0854] 4. Generating strategic advice

[0855] The server generates strategic advice based on market data and competitive analysis results, such as whether to enter new markets or propose improvements to contract terms.

[0856] The server customizes the generated strategic advice based on the user's industry profile.

[0857] 5. Emotion recognition

[0858] The device detects the user's emotional state in real time. For example, the emotion engine analyzes subtle changes in facial expressions and voice through the camera and microphone.

[0859] The server retrieves the recognized emotion data and adjusts the content of the strategic advice based on the emotion.

[0860] 6. Providing advice

[0861] The server sends customized strategy advice to the terminal.

[0862] The device displays strategic advice and related action items to the user, helping them create a concrete action plan.

[0863] 7. Incorporating feedback

[0864] The user inputs feedback on the provided advice into the terminal, including actual operational results and evaluations.

[0865] The terminal transmits the feedback from the user to the server.

[0866] The server updates the analysis model based on the collected feedback to improve the accuracy of future analyses. Model training is performed to reflect new data.

[0867] Specific examples

[0868] Example 1: Credit decision

[0869] User: A small business owner considering signing a new contract with a business partner. The user uses the device to enter basic information about the business partner (company name, address, etc.).

[0870] Terminal: Sends the entered information to the server.

[0871] Server: Collects additional financial data and transaction history from external data sources, performs data cleaning and analysis, calculates credit risk scores, and visualizes the results.

[0872] Terminal: Presents visualized risk assessment results to management and provides strategic advice on whether to proceed with a transaction and risk countermeasures.

[0873] Device: The emotion engine analyzes the manager's facial expressions and tone of voice, and the advice is adjusted appropriately based on the manager's emotions.

[0874] Example 2: Market trend forecasting and strategic advice

[0875] User: A marketing manager from a large company looking to enter a new market. He enters the market conditions of interest into the terminal.

[0876] Terminal: Sends entered market conditions to the server.

[0877] Server: Predicts future market trends based on market data and competitive analysis, and generates strategic advice customized based on the user's industry profile.

[0878] Terminal: The emotion engine analyzes the marketing manager's emotions in real time and reflects them in strategic advice.

[0879] Terminal: Presents strategic advice and relevant action items to the Marketing Manager.

[0880] In this way, the Business AI Navigator is a system that supports corporate decision-making by providing more personalized information and advice by taking into account the user's real-time emotions.

[0881] The processing flow will be explained below.

[0882] Step 1:

[0883] The user enters basic information about the business partner (company name, address, industry, annual sales, etc.) into the terminal.

[0884] Step 2:

[0885] The terminal formats the basic information entered by the user and sends it to the server.

[0886] Step 3:

[0887] Based on the basic information received by the server, additional customer data (such as financial information, past transaction history, credit information, etc.) is collected from external data sources.

[0888] Step 4:

[0889] The server performs data cleaning on the collected raw data, such as filling in missing values, removing inaccurate data, and standardizing the format.

[0890] Step 5:

[0891] The server integrates the cleaned data and manages it as a single dataset.

[0892] Step 6:

[0893] The server uses machine learning algorithms and statistical models to calculate the credit risk score of the client based on the integrated data.

[0894] Step 7:

[0895] The server visualizes the calculated credit risk score and related analytical results, generating graphs and dashboards.

[0896] Step 8:

[0897] The server sends the visualized data to the terminal so that the user can view it in an easily understandable format.

[0898] Step 9:

[0899] The terminal displays the visualized data sent from the server to the user, making it easy to understand.

[0900] Step 10:

[0901] The server generates strategic advice based on market data and competitive analysis results, such as whether to enter a new market or propose improvements to contract terms with business partners.

[0902] Step 11:

[0903] The server customizes the generated strategic advice based on the user's industry profile.

[0904] Step 12:

[0905] The device detects the user's emotional state in real time. For example, the emotion engine analyzes subtle changes in facial expressions and voice through the camera and microphone.

[0906] Step 13:

[0907] The server retrieves the recognized emotion data and adjusts the content of the strategic advice based on the emotion.

[0908] Step 14:

[0909] The server sends customized strategy advice to the terminal.

[0910] Step 15:

[0911] The device displays strategic advice and related action items to the user, helping them create a concrete action plan.

[0912] Step 16:

[0913] The user inputs feedback on the provided advice into the terminal, including actual operational results and evaluations.

[0914] Step 17:

[0915] The terminal transmits the feedback from the user to the server.

[0916] Step 18:

[0917] The server updates the analysis model based on the collected feedback to improve the accuracy of future analyses. Model training is performed to reflect new data.

[0918] Example 2

[0919] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0920] When companies make credit assessments or strategic decisions on their business partners, they need to collect and analyze massive amounts of data. However, existing systems require time and effort to clean and integrate data and calculate credit risk scores, making efficient decision-making difficult. Furthermore, they do not provide personalized advice that takes into account the user's emotional state, making it difficult to select the optimal course of action. To address these issues, a system is needed that can consistently perform everything from collecting business partner data to calculating credit risk scores, generating strategic advice, and even adjusting that advice based on the user's emotional state.

[0921] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0922] In this invention, the server includes means for acquiring basic information about trading partners, means for collecting additional data from external data sources, means for cleaning and integrating the collected data, means for calculating a credit risk score, means for visually displaying the calculated credit risk score, means for generating strategic advice based on market data and competitive analysis results, means for detecting a user's emotional state, and means for adjusting the strategic advice based on the detected emotional state, thereby enabling efficient cleaning and integration of data, calculation of a reliable credit risk score, and provision of personalized strategic advice that takes user emotions into consideration.

[0923] "Basic information" refers to basic information such as the business partner's company name, address, industry, and annual sales.

[0924] "External Data Source" refers to an external source from which the server can gather additional data about a business partner, such as a financial database or credit bureau.

[0925] "Data cleaning" refers to the process of filling in missing values ​​from collected data, removing inaccurate data, and standardizing the format.

[0926] "Data integration" refers to the process of integrating cleaned data and managing it as a single data set.

[0927] "Credit Risk Score" means a quantitative value calculated to assess the credit risk of a counterparty.

[0928] "Visually displaying" refers to visually representing the calculated credit risk score and related analytical results in the form of graphs, dashboards, etc.

[0929] "Generating strategic advice" refers to creating strategic advice such as whether to enter a new market or proposing improvements to contract terms based on market data and competitive analysis results.

[0930] "Emotional state detection" refers to analyzing the user's facial expressions and voice through a camera or microphone using an emotion engine to grasp the user's real-time emotional state.

[0931] "Adjusting strategic advice" refers to dynamically changing the content of strategic advice based on the detected emotional state of the user.

[0932] This invention relates to a new data analysis system for businesses, and presents an embodiment of a system called "Business AI Navigator" that streamlines credit risk assessment and strategic advice provision for business partners. This system has the ability to recognize the emotional state of the user and provide more personalized services.

[0933] Hardware and software used

[0934] Terminal: Input and display device (PC, tablet, smartphone, etc.)

[0935] Server: a high-performance computer for data processing and analysis

[0936] Database: Data storage for storing business partner information and data from external data sources

[0937] External data sources: financial databases and credit bureaus

[0938] Machine learning algorithms: Models for calculating credit risk scores (e.g., random forests, logistic regression)

[0939] Data visualization tools: D3.js, Tableau, etc.

[0940] Emotion recognition engine: Google Cloud Vision API, Amazon Rekognition

[0941] Processing steps

[0942] Step 1. Data collection

[0943] The user inputs basic information about the business partner (company name, address, industry, annual turnover, etc.) into the terminal. As an example, let's say the owner of a small or medium-sized business inputs information about a business partner with whom he or she wants to start a new transaction.

[0944] The device formats the basic information entered and sends it to the server in a format such as JSON or XML, using HTTP or HTTPS as the transmission protocol.

[0945] Based on the basic information received by the server, additional customer data is collected from external data sources (e.g., financial information databases, credit bureaus).

[0946] Step 2. Data analysis

[0947] The raw data collected by the server is subjected to data cleaning, which involves filling in missing financial information, removing inaccurate data, and standardizing the format.

[0948] The server consolidates the cleaned data and manages it as a single dataset, for example by eliminating duplicate data from different data sources.

[0949] The server uses the integrated data to calculate the client's credit risk score using machine learning algorithms (e.g., random forest, logistic regression model).

[0950] Step 3. View the results

[0951] The server visualizes the calculated credit risk scores and related analytical results, generating graphs and dashboards using tools such as D3.js and Tableau.

[0952] The device presents the generated visualizations to the user in an easy-to-understand manner, such as line graphs or heat maps showing rising or falling risk scores.

[0953] Step 4. Generate strategic advice

[0954] The server generates strategic advice based on market data and competitive analysis, including recommendations on whether to enter new markets and improving contract terms.

[0955] The server customizes the generated strategic advice based on the user's industry profile, including industry-specific competitive information for users in the automotive industry.

[0956] Step 5. Emotion Recognition

[0957] The device detects the user's emotional state in real time using a camera and microphone, and facial and voice data is analyzed by an emotion engine (e.g., Google Cloud Vision API, Amazon Rekognition).

[0958] The server acquires the emotion data and adjusts the strategic advice based on the user's emotions. If the user is feeling stressed, the advice will be changed to help alleviate that stress.

[0959] Step 6. Providing advice

[0960] The server sends customized strategy advice to the device, using a secure communication protocol (e.g., HTTPS).

[0961] The device will display strategic advice and related action items to the user to help them plan specific actions, for example in the form of a to-do list.

[0962] Step 7. Incorporate feedback

[0963] The user inputs feedback (actual operational results and evaluation) on the advice into the terminal. For example, the user inputs comments such as "The advice was effective" or "Improvement is needed."

[0964] The device sends the feedback to the server, which formats it in the appropriate schema and stores it in a database.

[0965] The server analyzes the feedback and updates the analysis model to improve accuracy in future runs. Model training uses machine learning frameworks such as TensorFlow and PyTorch.

[0966] In this way, the Business AI Navigator is a system that supports corporate decision-making by taking into account users' real-time emotions and providing personalized information and advice.

[0967] Prompt Sentence Examples

[0968] "Please provide a strategy for entering a new market. The market conditions are as follows: the region is North America, the target customer base is small and medium-sized businesses, and the competitors are Company A and Company B."

[0969] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0970] Step 1:

[0971] Enter basic information about your business partner

[0972] The user enters basic information about the business partner (company name, address, industry, annual sales, etc.) into the terminal.

[0973] Input: Basic information such as company name, address, industry, annual sales, etc.

[0974] The terminal formats this information and sends it to the server.

[0975] Output: Basic formatted information sent to the server.

[0976] Step 2:

[0977] External Data Collection

[0978] Based on the basic information received by the server, additional customer data is collected from external data sources (e.g., financial databases, credit bureaus).

[0979] Input: Basic formatted information

[0980] The server uses APIs to communicate with external data sources to obtain additional data (financial information, past transaction history, credit information, etc.).

[0981] Output: The additional customer data collected is stored on the server.

[0982] Step 3:

[0983] Data Cleaning

[0984] The server performs data cleaning on the collected raw data.

[0985] Input: Additional Account Data

[0986] Data cleaning includes imputing missing values, removing inaccurate data, and standardizing formats.

[0987] Output: Cleaned data is obtained.

[0988] Step 4:

[0989] Data Integration

[0990] The server integrates the cleaned data and manages it as a single dataset.

[0991] Input: Cleaned data

[0992] Eliminate duplicate data and standardize data formats.

[0993] Output: A consolidated dataset is created.

[0994] Step 5:

[0995] Credit risk score calculation

[0996] The server uses the integrated data to calculate the client's credit risk score using machine learning algorithms (e.g., random forest, logistic regression model).

[0997] Input: Integrated dataset

[0998] An algorithm analyzes the data and calculates a credit risk score.

[0999] Output: Calculated credit risk score

[1000] Step 6:

[1001] Visualizing the results

[1002] The server visualizes the calculated credit risk score and related analytical results, generating graphs and dashboards.

[1003] Input: Credit risk scores and associated analysis results

[1004] Use tools such as D3.js or Tableau to visualize the results.

[1005] Output: Visualized data

[1006] Step 7:

[1007] Presenting Visualized Data

[1008] The terminal presents the generated visualization data to the user and displays it in an easy-to-understand manner.

[1009] Input: Visualized data

[1010] Displays line graphs and heat maps showing increases or decreases in risk scores.

[1011] Output: Visualized data presented to the user

[1012] Step 8:

[1013] Strategic Advice Generation

[1014] The server generates strategic advice based on market data and competitive analysis results.

[1015] Input: Market data, competitive analysis results

[1016] This includes whether to enter new markets and proposals for improving contract terms.

[1017] Output: Strategic advice

[1018] Step 9:

[1019] Customization

[1020] The server customizes the generated strategic advice based on the user's industry profile.

[1021] Input: Strategic advice, user industry profile

[1022] For users in the automotive industry, industry-specific competitive information is added.

[1023] Output: Customized strategy advice

[1024] Step 10:

[1025] emotion recognition

[1026] The device detects the user's emotional state in real time.

[1027] Input: User's facial expressions and voice data

[1028] Using the camera and microphone, the data is analyzed by an emotion engine (e.g., Google Cloud Vision API, Amazon Rekognition).

[1029] Output: Recognized emotion data

[1030] Step 11:

[1031] Advice adjustment

[1032] The server retrieves the recognized emotion data and adjusts the content of the strategic advice.

[1033] Input: Recognized sentiment data, customized strategy advice

[1034] If you are feeling stressed, change the advice to something that will relieve you.

[1035] Output: Tailored strategy advice

[1036] Step 12:

[1037] Providing advice

[1038] The server sends the adjusted strategy advice to the terminal.

[1039] Input: Tailored Strategy Advice

[1040] Use secure communication protocols (e.g. HTTPS).

[1041] Output: Strategy advice sent to the terminal

[1042] Step 13:

[1043] Providing advice

[1044] The device displays strategic advice and related action items to the user to assist with specific action planning.

[1045] Input: Strategy Advice

[1046] Display in ToDo list format, etc.

[1047] Output: A concrete action item presented to the user

[1048] Step 14:

[1049] Enter your feedback

[1050] The user inputs feedback (actual operational results and evaluation) on the advice into the terminal.

[1051] Input: Feedback

[1052] Record the effectiveness of the advice and areas for improvement.

[1053] Output: Input feedback data

[1054] Step 15:

[1055] Send Feedback

[1056] The device sends the feedback to the server, which formats it into the appropriate schema.

[1057] Input: Feedback data

[1058] The feedback is stored in a database.

[1059] Output: Feedback data sent to the server

[1060] Step 16:

[1061] Updating a Model

[1062] The server analyzes the feedback and updates the analysis model to improve the accuracy of the analysis from the next time onwards.

[1063] Input: Submitted feedback data

[1064] Machine learning frameworks such as TensorFlow and PyTorch are used for model training.

[1065] Output: Updated analytical model

[1066] (Application example 2)

[1067] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[1068] Conventional security systems determine alert levels and countermeasures based on the target's behavior and belongings. However, because there is no mechanism for analyzing and reflecting the target's emotional state in real time, appropriate responses are often delayed, which can result in increased security risks. Furthermore, conventional systems have the problem of low response accuracy because the advice they determine is uniform and lacks personalization.

[1069] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for acquiring basic information about trading partner companies, means for collecting additional trading partner data from external data sources, means for cleaning and integrating the collected data, means for calculating trading partner credit risk scores, means for visually displaying the calculated credit risk scores, means for generating strategic advice based on market data and competitive analysis results, means for analyzing the user's real-time emotional state, and means for adjusting the strategic advice based on the analyzed emotional state. This makes it possible to analyze and reflect the user's emotional state in real time and provide personalized security countermeasures.

[1070] "Means for obtaining basic information about business partners" refers to methods for collecting basic information such as the name, address, industry, and sales of business partners.

[1071] "Means for collecting additional business partner data from external data sources" refers to methods for obtaining financial information, past transaction history, credit information, etc. of business partners from external data sources.

[1072] "Means for cleaning and integrating collected data" refers to methods for preparing collected data by filling in missing values, removing inaccurate data, and standardizing the format.

[1073] The "means for calculating a customer credit risk score" is a method for calculating a score for evaluating the credit risk of a customer based on the collected and cleaned customer data.

[1074] The "means for visually displaying the calculated credit risk score" is a method for visually displaying the counterparty credit risk score as a graph or dashboard.

[1075] The "means for generating strategic advice based on market data and competitor analysis results" is a method for generating effective strategic advice based on market data and competitor analysis results.

[1076] "Means for analyzing the user's real-time emotional state" refers to a method for analyzing the user's facial expressions and tone of voice through a camera or microphone to grasp the user's emotional state in real time.

[1077] The "means for adjusting strategic advice based on the analyzed emotional state" is a method for appropriately adjusting the strategic advice provided based on the results of the user's emotional analysis.

[1078] This invention relates to a system that combines "Security Navigator," a new AI data service for businesses, with an emotion engine to analyze real-time emotional processes and support strategic security responses. This system uses smart glasses to collect visual and audio information from the target, which is then sent to a server for analysis. The following describes in detail the mode for carrying out the invention.

[1079] Hardware Configuration

[1080] 1. Smart Glasses

[1081] Visual information collection: high-resolution cameras

[1082] Audio information collection: High-sensitivity microphone

[1083] Display: Built-in display for visual display

[1084] Processor: Initial processing and formatting of data

[1085] 2. Server

[1086] High-performance cloud servers (e.g. AWS, Google Cloud)

[1087] 3. Means of communication

[1088] Real-time communication: WebSocket, MQTT

[1089] Software Configuration

[1090] 1. Emotion Recognition Engine

[1091] Google Cloud Vision API

[1092] OpenAI sentiment analysis model

[1093] 2. Data Analysis Tools

[1094] Python

[1095] TensorFlow

[1096] PyTorch

[1097] Processing flow (system program)

[1098] The server implements the invention using the following means.

[1099] 1. Data Collection

[1100] Smart glasses collect visual and audio information from the wearer.

[1101] Video data is acquired through a camera, and audio data is acquired through a microphone.

[1102] This data is formatted by a processor within the smart glasses and sent to a server in real time.

[1103] 2. Data Analysis

[1104] The server analyzes the received video and audio data and recognizes the emotional state using an emotion engine.

[1105] The emotion recognition engine uses Google Cloud Vision API and OpenAI's emotion analysis model.

[1106] The analysis results are fed back visually and audibly in real time.

[1107] 3. Displaying the results

[1108] The analysis results, showing the emotional state of the target person, are displayed on the smart glasses' display.

[1109] Depending on the emotional state, a security level such as alert or normal is indicated.

[1110] 4. Generating strategic advice

[1111] The server generates optimal responses based on the emotional state.

[1112] If the subject appears suspicious, security guards will provide real-time advice on whether increased vigilance is needed or whether further investigation is required.

[1113] 5. Providing advice

[1114] Smart glasses display real-time strategic advice.

[1115] Countermeasures will also be provided through audio guidance, allowing the wearer to respond smoothly.

[1116] 6. Incorporating feedback

[1117] Staff feedback is entered into the smart glasses via voice input or a touch panel.

[1118] This feedback is sent to the server and used to update and improve the sentiment analysis model.

[1119] Specific examples

[1120] Example 1: Commercial Intruder Screening

[1121] 1. A security guard wearing smart glasses during an intruder inspection at a commercial facility.

[1122] 2. The smart glasses send visual and audio data to the server.

[1123] 3. The server recognizes through emotion analysis that the intruder is nervous and sets the alert level to high.

[1124] 4. The smart glasses display will show "High alert state" and a voice prompt will say "Prompt for fixation confirmation."

[1125] Prompt Sentence Examples

[1126] "Based on this video and audio data, please analyze the subject's emotional state and generate an appropriate response accordingly."

[1127] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1128] Step 1:

[1129] The smart glasses collect the wearer's visual and audio information. The video data is acquired through a camera, and the audio data is acquired through a microphone. The smart glasses receive the video and audio data from the camera and microphone as input, format these data, and send them to a server in real time through a processor within the smart glasses. The output is the formatted video and audio data.

[1130] Step 2:

[1131] The server analyzes the received video and audio data. As input, it receives the video and audio data sent from the smart glasses and performs data cleaning. Specifically, it removes unnecessary noise and extracts only the necessary information. As output, it obtains the cleaned video and audio data.

[1132] Step 3:

[1133] The server uses the cleaned data to analyze the emotional state using an emotion engine. It receives the cleaned video and audio data as input, and uses Google Cloud Vision API and OpenAI emotion analysis models to recognize the user's emotional state from their facial expressions and tone of voice. The output is an analysis result that represents the user's emotional state.

[1134] Step 4:

[1135] The server generates countermeasures based on the analyzed emotion results. As input, it receives the analysis results that represent the emotional state and generates optimal security countermeasures based on them. For example, if it detects a state of tension, it sets the alert level to high. The generated countermeasures are obtained as output.

[1136] Step 5:

[1137] The results and countermeasures are displayed and provided to the wearer in real time through the smart glasses' display and audio guide. Countermeasures sent from the server are received as input and displayed on the display. Specifically, instructions such as "High alert state" and "Encourage fixation confirmation" are displayed, and the same information is provided via audio guide. Countermeasures are delivered to the wearer visually and audibly as output.

[1138] Step 6:

[1139] The wearer performs the countermeasure and provides feedback on the results through the smart glasses. As input, the wearer's feedback data is received, which is collected through the touch panel or voice input. As output, the collected feedback data is sent to the server.

[1140] Step 7:

[1141] The server analyzes the received feedback data and updates the sentiment analysis model. It takes the feedback data as input and trains the model based on it, specifically using new data to improve the sentiment analysis algorithm. The output is an updated sentiment analysis model.

[1142] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1143] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1144] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[1145] [Third embodiment]

[1146] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1147] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[1148] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1149] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[1150] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1151] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1152] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1153] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1154] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1155] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1156] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1157] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[1158] This invention relates to "Business AI Navigator," a new AI data service for businesses. It is a system that streamlines credit approval procedures for clients and supports strategic decision-making. A specific embodiment of this system is shown below.

[1159] Specific program behavior

[1160] 1. Data Collection

[1161] User: Enter basic information about the business partner into the terminal, such as the company name, address, industry, and annual sales.

[1162] Terminal: Sends the information entered by the user to the server.

[1163] Server: Based on the received information, collects additional customer data (e.g., financial information and transaction history) from external data sources.

[1164] 2. Data Analysis

[1165] Server: Performs data cleaning based on the collected data, such as filling in missing values, removing inaccurate data, and standardizing formats.

[1166] Server: Consolidates the cleaned data and calculates the credit risk score for each counterparty, using machine learning algorithms and statistical models.

[1167] 3. Displaying the results

[1168] Server: Visualizes the calculated credit risk scores and generates graphs and dashboards.

[1169] Terminal: Presents the generated visualization data to the user for easy understanding.

[1170] 4. Generating strategic advice

[1171] Server: Generates strategic advice based on market data and competitive analysis, such as whether to enter new markets or improve contract terms with business partners.

[1172] Server: Customizes the generated strategic advice based on the user's industry profile.

[1173] 5. Providing advice

[1174] Server: Sends customized advice to the device.

[1175] On the device: Provide user-visible advice and list required action items.

[1176] 6. Incorporating feedback

[1177] User: Enters feedback and results on the advice provided into the device.

[1178] Terminal: Sends user feedback to the server.

[1179] Server: Updates the analysis model based on the feedback, improving the accuracy of the next analysis.

[1180] Specific examples

[1181] Example 1: Credit decision

[1182] User: A small business owner considering signing a new contract with a business partner. The user uses the device to enter basic information about the business partner (company name, address, etc.).

[1183] Terminal: Sends the entered information to the server.

[1184] Server: Collects additional financial data and transaction history from external data sources, performs data cleaning and analysis, calculates credit risk scores, and visualizes the results.

[1185] Terminal: Presents visualized risk assessment results to management and provides strategic advice on whether to proceed with a transaction and risk countermeasures.

[1186] Example 2: Market trend forecasting and strategic advice

[1187] User: A marketing manager from a large company looking to enter a new market. He enters the market conditions of interest into the terminal.

[1188] Terminal: Sends entered market conditions to the server.

[1189] Server: Predicts future market trends based on market data and competitive analysis, and generates strategic advice customized based on the user's industry profile.

[1190] Terminal: Presents strategic advice and relevant action items to the Marketing Manager.

[1191] In this way, Business AI Navigator is a system that supports corporate credit operations and strategic decision-making by providing efficient and accurate information.

[1192] The processing flow will be explained below.

[1193] Step 1:

[1194] The user enters basic information about the business partner (company name, address, industry, annual sales, etc.) into the terminal.

[1195] Step 2:

[1196] The terminal formats the basic information entered by the user and sends it to the server.

[1197] Step 3:

[1198] Based on the basic information received, the server collects additional customer data from external data sources, such as financial information, past transaction history, credit information, etc.

[1199] Step 4:

[1200] The server performs data cleaning on the collected raw data, such as filling in missing values, removing inaccurate data, and standardizing the format.

[1201] Step 5:

[1202] The server integrates the cleaned data and manages it as a single dataset.

[1203] Step 6:

[1204] The server uses machine learning algorithms and statistical models to calculate the credit risk score of the client based on the integrated data.

[1205] Step 7:

[1206] The server visualizes the calculated credit risk score and related analytical results by generating graphs and charts in a visually easy-to-understand format.

[1207] Step 8:

[1208] The server sends the visualized data to the terminal so that it can be presented to the user in an easy-to-understand format.

[1209] Step 9:

[1210] The terminal displays the visualized data sent from the server to the user, allowing the user to easily understand the data.

[1211] Step 10:

[1212] The server generates strategic advice based on market data and competitive analysis, such as whether to enter a new market or suggest improvements to contract terms.

[1213] Step 11:

[1214] The server customizes the generated strategic advice based on the user's industry profile.

[1215] Step 12:

[1216] The server sends customized strategy advice to the terminal.

[1217] Step 13:

[1218] The device displays strategic advice and related action items to the user, helping them create a concrete action plan.

[1219] Step 14:

[1220] The user inputs feedback on the provided advice into the terminal, including actual operational results and evaluations.

[1221] Step 15:

[1222] The terminal transmits the feedback from the user to the server.

[1223] Step 16:

[1224] The server updates the analysis model based on the collected feedback to improve the accuracy of future analyses. Model training is performed to reflect new data.

[1225] Example 1

[1226] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1227] Conventional credit decision-making systems had problems with complex credit risk assessment of counterparties and time-consuming data integration and cleaning. They also lacked a mechanism for effectively utilizing market data and competitive analysis results necessary for strategic decision-making. Furthermore, there was an insufficient means for appropriately collecting user feedback and incorporating it into the analysis model, which could lead to a decline in analysis accuracy. There was a need to solve these problems and provide information efficiently and accurately.

[1228] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1229] In this invention, the server includes means for acquiring basic information about counterparty companies, means for collecting additional counterparty data from external data sources, means for cleaning and integrating the collected data, means for calculating counterparty credit risk scores, means for visually displaying the calculated credit risk scores, means for generating strategic advice based on market data and competitive analysis results, means for customizing the generated advice and presenting it on a terminal, means for collecting user feedback to update the analytical model, and means for securely transmitting the feedback, thereby enabling more efficient and accurate credit risk assessment and strategic decision support.

[1230] "Basic information about a business partner" refers to basic identifying information about the business partner, such as company name, location, industry, and annual sales.

[1231] "External Data Sources" refers to external databases or information services that provide additional information about business partners.

[1232] "Data cleaning" refers to the process of correcting, completing, and standardizing missing values, inaccurate data, and improper formats from collected data.

[1233] "Synthesis" refers to the process of bringing together data collected from different sources into a single data set.

[1234] "Credit risk score" refers to a numerical evaluation value of a trading partner's credit risk, and is used to evaluate the trading partner's creditworthiness.

[1235] "Visually displaying" refers to displaying credit risk scores and other analytical results in a visual format, such as a graph or dashboard.

[1236] "Market data" refers to information regarding market movements, trends, the competitive environment, etc.

[1237] "Competitive analysis results" refers to the analysis results regarding competitors' strategies, performance, market share, etc.

[1238] "Strategic advice" refers to providing advice on trading strategies with business partners and entry into new markets based on market data and competitive analysis results.

[1239] An "industry profile" refers to information that compiles standard characteristics, trends, trade practices, etc. related to a particular industry.

[1240] "Gathering feedback" refers to collecting information such as opinions, ratings, and results from users.

[1241] "Updating the analytical model" refers to readjusting machine learning algorithms and statistical models based on newly collected feedback to improve prediction and analytical accuracy.

[1242] "Securely" refers to measures to ensure safe communications and data processing and prevent information leaks and unauthorized access.

[1243] This invention relates to "Business AI Navigator," a new AI data service for businesses, which is a system that streamlines credit approval operations for clients and supports strategic decision-making. This system is implemented using the following hardware and software.

[1244] Hardware and software used

[1245] Server: Uses cloud services (e.g. Amazon Web Services, Microsoft Azure) for data processing and storage.

[1246] Device: Desktop PC, laptop, tablet, etc. The user interface is a web browser (e.g., Google Chrome, Firefox).

[1247] Software: Data visualization tools (e.g., Tableau, Power BI) and machine learning algorithms (e.g., Scikit-learn, TensorFlow) will be used.

[1248] System Operation Overview

[1249] 1. Data Collection

[1250] User: Enter basic information about a business partner into the terminal. For example, enter the company name "ABC Co., Ltd.", address "Chiyoda-ku, Tokyo," industry "manufacturing," and annual sales of "10 billion yen."

[1251] Terminal: Sends the entered information to the server. The terminal uses a secure communication protocol (HTTPS).

[1252] Server: Collect additional counterparty data (financial information, transaction history, etc.) from external data sources, using APIs.

[1253] 2. Data Analysis

[1254] Server: Cleans the collected data, imputes missing values, removes inaccurate data, and standardizes the format.

[1255] Server: Integrates clean data and calculates credit risk scores for clients. Random forest and logistic regression models are used. Machine learning algorithms include Scikit-learn and TensorFlow.

[1256] 3. Displaying the results

[1257] Server: Visualize the calculated credit risk scores and generate graphs and dashboards using Tableau or Power BI as data visualization tools.

[1258] Terminal: Presents the generated visualization data to the user in an easy-to-understand format.

[1259] 4. Generating strategic advice

[1260] Server: Generates strategic advice based on market data and competitive analysis results, such as proposing whether to enter a new market or improve contract terms with business partners.

[1261] Server: Customizes generated strategic advice based on the user's industry profile.

[1262] 5. Providing advice

[1263] Server: Sends customized advice to the device.

[1264] Terminal: Provides advice to the user and lists required action items.

[1265] 6. Incorporating feedback

[1266] User: Enters feedback and results on the advice provided into the device.

[1267] Terminal: Sends user feedback to the server.

[1268] Server: Updates the analysis model based on the feedback, improving the accuracy of the next analysis.

[1269] Specific examples

[1270] Example 1: Credit decision

[1271] User: A small business owner considering signing a new contract with a business partner. He enters basic information about the business partner into the terminal.

[1272] Terminal: Sends the entered information to the server.

[1273] Server: Collects additional financial data and transaction history from external data sources, performs data cleaning and analysis, calculates credit risk scores, and visualizes the results.

[1274] Terminal: Presents visualized risk assessment results and provides strategic advice on whether to trade and how to mitigate risks.

[1275] Example 2: Market trend forecasting and strategic advice

[1276] User: A marketing manager from a large company looking to enter a new market. He enters the market conditions of interest into the terminal.

[1277] Terminal: Sends entered market conditions to the server.

[1278] Server: Predicts future market trends based on market data and competitive analysis, and generates strategic advice customized based on the user's industry profile.

[1279] Terminal: Presents strategic advice and relevant action items to the Marketing Manager.

[1280] Prompt Sentence Examples

[1281] "Based on the company information below, please assess the credit risk of your business partner and provide strategic advice."

[1282] 1. Company Name: ABC Co., Ltd.

[1283] 2. Address: Chiyoda-ku, Tokyo

[1284] 3. Industry: Manufacturing

[1285] 4. Annual sales: 10 billion yen

[1286] In this way, the Business AI Navigator supports corporate credit operations and strategic decision-making by providing efficient and accurate information. The above is an embodiment of the present invention.

[1287] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1288] Step 1:

[1289] Data collection

[1290] User: Enters basic information about a business partner into the terminal. The entered information is company name "ABC Co., Ltd.", address "Chiyoda-ku, Tokyo," industry "manufacturing," and annual sales of "10 billion yen."

[1291] Terminal: The input information is sent to the server using a secure communication protocol (HTTPS). The input data is transferred to the server in JSON format.

[1292] Server: Collects additional customer data (financial information, transaction history, etc.) from external data sources via API. The collected data is stored in a database in an organized format by field.

[1293] Step 2:

[1294] Data Cleaning

[1295] Server: Cleans the collected data by filling in missing values, removing inaccurate data, and standardizing date and currency formats. For example, if there is missing annual sales data, it fills in the approximate value using historical data or other relevant data. It also corrects inaccurate entries (e.g., company names with trailing spaces).

[1296] Input: The raw data collected.

[1297] Output: Cleaned consolidated data.

[1298] Step 3:

[1299] Data Integration

[1300] Server: Consolidates the cleaned data into a single dataset. This process involves combining information from multiple data sources using a common key (e.g., company ID). For example, combining financial data and transaction history based on company ID.

[1301] Input: Multiple cleaned datasets.

[1302] Output: A consolidated dataset.

[1303] Step 4:

[1304] Credit risk score calculation

[1305] Server: Calculates a credit risk score based on the integrated data. This uses a machine learning algorithm (e.g., random forest, logistic regression). The integrated dataset is input into the machine learning model, and a credit risk score is output in numerical form.

[1306] Input: Unified dataset.

[1307] Output: Credit risk score.

[1308] Step 5:

[1309] Visualizing the results

[1310] Server: Generate visualization data using the calculated credit risk scores. Use a data visualization tool (e.g., Tableau, Power BI) to convert the data into a format that is easy to display as graphs or dashboards.

[1311] Terminal: Receives the visualization data sent from the server and displays it to the user through a user interface, presenting the results in the form of easy-to-understand graphs and charts.

[1312] Input: Credit Risk Score.

[1313] Output: Visualized data (graphs, dashboards).

[1314] Step 6:

[1315] Generating Strategic Advice

[1316] Server: Generates strategic advice based on market data and competitive analysis results. Using AI models, it analyzes, for example, whether to enter a new market or how to improve transaction terms, and makes specific strategic proposals.

[1317] Server: Customizes the generated advice based on the user's industry profile. Provides strategic advice suited to the user's industry characteristics.

[1318] Inputs: Market data, competitive analysis results, and your industry profile.

[1319] Output: Strategy advice.

[1320] Step 7:

[1321] Providing advice

[1322] Server: Sends customized advice to the device.

[1323] On the device: Providing advice to users and listing necessary action items, such as specific bulleted steps for entering a new market.

[1324] Enter: strategic advice.

[1325] Output: Displayed advice and action items.

[1326] Step 8:

[1327] Gathering and implementing feedback

[1328] User: Enters feedback and results on the advice provided into the device.

[1329] Terminal: Sends user feedback to the server, which then forwards the feedback to the server in an appropriate format.

[1330] Server: Updates the analysis model based on the feedback to improve the accuracy of future analyses. Retrains the machine learning model based on new feedback data.

[1331] Input: User feedback.

[1332] Output: Updated analytical model.

[1333] (Application example 1)

[1334] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1335] While existing AI data services for businesses were capable of collecting basic information on business partners and calculating credit risk, they had issues with not being able to update analytical models that take into account individualized content recommendations or user feedback. Furthermore, they were unable to provide personalized content recommendations based on users' viewing history and interests, making it difficult to improve the user experience.

[1336] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1337] In this invention, the server includes means for acquiring basic information about trading companies, means for collecting additional trading data from external sources, means for cleaning and integrating the collected data, means for assessing trading credit risk, means for visually displaying the calculated credit risk, means for generating strategic advice based on market information and competitive analysis results, means for making personalized content recommendations based on the user's viewing history and genres of interest, means for presenting the generated recommendations to the user, and means for collecting feedback on the provided recommendations and updating the analytical model. This makes it possible to improve the analytical model based on the personalized content recommendations and feedback, in addition to assessing the credit risk of trading partners.

[1338] A "trading enterprise" is a legal entity or organization that transacts business with other organizations or individuals in the course of commercial activities.

[1339] "External sources" refers to additional data that can be accessed both inside and outside the company, such as from public or private organizations or industry reports.

[1340] "Trade credit risk" is an assessment criterion that indicates the credit risk regarding the financial soundness of a trading partner and the stability of transactions in commercial activities.

[1341] "Cleaning" refers to the process of removing inaccurate data, filling in missing values, and standardizing the format prior to data analysis.

[1342] "Synthesis" refers to the process of bringing together data sets collected from different sources and converting them into a format that can be used for analysis.

[1343] "Strategic advice" refers to advice and suggestions to support business decision-making based on market information and competitive analysis results.

[1344] "Viewing history" refers to information such as the content a user has viewed in the past, the viewing time, and ratings, and is used to provide personalized services.

[1345] A "genre" is a classification of content, referring to a category that indicates a particular theme or style of film, music, book, etc.

[1346] "Feedback" refers to the ratings, opinions, and experiences you provide regarding the services and recommendations you receive.

[1347] An "analytical model" refers to a computational model that uses mathematical and statistical techniques to find patterns and make predictions or classifications based on collected data.

[1348] "Content recommendation" refers to a service that individually suggests appropriate content based on a user's interests and history information.

[1349] This system is realized through data interaction between the server, the terminal, and the user.

[1350] Data collection

[1351] 1. User: Enters information about viewing history and genres of interest into the device. For example, the user enters their recent movie viewing history and movie genres of interest (action, drama, etc.).

[1352] 2. Terminal: Sends the entered information to the server.

[1353] 3. Server: Collects additional data from external sources (e.g., viewing history and reputation data of other users).

[1354] Data analysis

[1355] 1. Server: Cleans the collected data, removing inaccurate data, filling in missing values, and standardizing the data format.

[1356] 2. Server: Train an analytics model (e.g., RandomForestClassifier) ​​based on the cleaned data. The analytics model is used to recommend personalized content based on viewing history and genre information.

[1357] Displaying the results

[1358] 1. Server: Generates graphs and dashboards to visually present the calculated recommended content.

[1359] 2. Terminal: Displays the generated visualization data and content recommendation results to the user.

[1360] Strategic Advice Generation

[1361] 1. Server: Generates personalized strategic advice based on market intelligence and competitive analysis, such as recommendations for new movies or the next series to watch.

[1362] 2. Server: Customizes strategic advice based on the user's profile.

[1363] Providing advice

[1364] 1. Server: Sends customized strategy advice to devices.

[1365] 2. On the device: Provide user-visible advice and a list of what to watch next.

[1366] Reflecting feedback

[1367] 1. User: Enters feedback into the device regarding the advice and recommended content provided.

[1368] 2. Terminal: Sends feedback information to the server.

[1369] 3. Server: Updates the analysis model based on user feedback to improve the accuracy of future content recommendations.

[1370] Examples and prompts

[1371] Example: A scenario in which a user receives recommendations for the next movie to watch based on the genre and ratings of the movies they have watched. For example, recommendations for highly rated new action movies.

[1372] Example prompt:

[1373] "Recommend a personalized list of movies based on the genres and ratings of movies the user has watched. Also provide details about the latest top-rated action movies."

[1374] This system allows users to receive personalized content quickly and accurately, improving the viewing experience, while also providing optimal recommendations by updating the analysis model based on user feedback.

[1375] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1376] Step 1:

[1377] User: The user inputs information about their viewing history and genres of interest into the device. For example, they input their recent movie viewing history and movie genres of interest (action, drama, etc.). This is the input information.

[1378] Step 2:

[1379] Device: The device sends the information entered by the user to the server, such as data about viewing history and genres of interest.

[1380] Step 3:

[1381] Server: The server collects additional data from external sources (viewing history and reputation data from other users), thereby supplementing the input information to create a comprehensive dataset.

[1382] Step 4:

[1383] Server: Cleans the collected data. Ensures consistency of the dataset by removing inaccurate data and filling in missing information. The cleaned data becomes the output information.

[1384] Step 5:

[1385] Server: Trains a generative AI model (e.g., RandomForestClassifier) ​​based on the cleaned data. This model is used to recommend personalized content based on viewing history and genre information. The trained model is the output.

[1386] Step 6:

[1387] Server: Using the trained model, the server predicts the best content recommendations for the user. This prediction is performed using input data such as viewing history and genre information. The prediction results are the output information.

[1388] Step 7:

[1389] Server: Generates graphs and dashboards to visually present the calculated recommended content. The visualized data is the output information.

[1390] Step 8:

[1391] Terminal: The terminal displays the generated visualization data and content recommendation results to the user, allowing the user to check the recommended content.

[1392] Step 9:

[1393] User: Enters feedback on the provided recommended content into the device. The feedback data becomes input information.

[1394] Step 10:

[1395] Terminal: The terminal sends the feedback from the user to the server, which passes this feedback information to the server.

[1396] Step 11:

[1397] Server: The server updates the analytical model based on user feedback, improving the accuracy of future content recommendations. The updated model becomes the output information.

[1398] This process allows users to receive personalized content quickly and accurately, improving their viewing experience. Furthermore, by updating the analysis model based on feedback, the system can consistently provide optimal recommendations.

[1399] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1400] This invention relates to a system that combines an emotion engine with "Business AI Navigator," a new AI data service for businesses. This system not only streamlines credit approval procedures for clients and supports strategic decision-making, but also provides more personalized services by recognizing users' emotions. A specific embodiment of this system is shown below.

[1401] Specific program behavior

[1402] 1. Data Collection

[1403] The user enters basic information about the business partner into the terminal, such as the company name, address, industry, and annual sales.

[1404] The terminal formats the basic information entered by the user and sends it to the server.

[1405] Based on the information received, the server collects additional customer data from external data sources, such as financial information, past transaction history, credit information, etc.

[1406] 2. Data Analysis

[1407] The server performs data cleaning on the collected raw data, such as filling in missing values, removing inaccurate data, and standardizing the format.

[1408] The server integrates the cleaned data and manages it as a single dataset.

[1409] The server uses machine learning algorithms and statistical models to calculate the credit risk score of the client based on the integrated data.

[1410] 3. Displaying the results

[1411] The server visualizes the calculated credit risk score and related analytical results, generating graphs and dashboards.

[1412] The terminal presents the generated visualization data to the user for easy understanding.

[1413] 4. Generating strategic advice

[1414] The server generates strategic advice based on market data and competitive analysis results, such as whether to enter new markets or propose improvements to contract terms.

[1415] The server customizes the generated strategic advice based on the user's industry profile.

[1416] 5. Emotion recognition

[1417] The device detects the user's emotional state in real time. For example, the emotion engine analyzes subtle changes in facial expressions and voice through the camera and microphone.

[1418] The server retrieves the recognized emotion data and adjusts the content of the strategic advice based on the emotion.

[1419] 6. Providing advice

[1420] The server sends customized strategy advice to the terminal.

[1421] The device displays strategic advice and related action items to the user, helping them create a concrete action plan.

[1422] 7. Incorporating feedback

[1423] The user inputs feedback on the provided advice into the terminal, including actual operational results and evaluations.

[1424] The terminal transmits the feedback from the user to the server.

[1425] The server updates the analysis model based on the collected feedback to improve the accuracy of future analyses. Model training is performed to reflect new data.

[1426] Specific examples

[1427] Example 1: Credit decision

[1428] User: A small business owner considering signing a new contract with a business partner. The user uses the device to enter basic information about the business partner (company name, address, etc.).

[1429] Terminal: Sends the entered information to the server.

[1430] Server: Collects additional financial data and transaction history from external data sources, performs data cleaning and analysis, calculates credit risk scores, and visualizes the results.

[1431] Terminal: Presents visualized risk assessment results to management and provides strategic advice on whether to proceed with a transaction and risk countermeasures.

[1432] Device: The emotion engine analyzes the manager's facial expressions and tone of voice, and the advice is adjusted appropriately based on the manager's emotions.

[1433] Example 2: Market trend forecasting and strategic advice

[1434] User: A marketing manager from a large company looking to enter a new market. He enters the market conditions of interest into the terminal.

[1435] Terminal: Sends entered market conditions to the server.

[1436] Server: Predicts future market trends based on market data and competitive analysis, and generates strategic advice customized based on the user's industry profile.

[1437] Terminal: The emotion engine analyzes the marketing manager's emotions in real time and reflects them in strategic advice.

[1438] Terminal: Presents strategic advice and relevant action items to the Marketing Manager.

[1439] In this way, the Business AI Navigator is a system that supports corporate decision-making by providing more personalized information and advice by taking into account the user's real-time emotions.

[1440] The processing flow will be explained below.

[1441] Step 1:

[1442] The user enters basic information about the business partner (company name, address, industry, annual sales, etc.) into the terminal.

[1443] Step 2:

[1444] The terminal formats the basic information entered by the user and sends it to the server.

[1445] Step 3:

[1446] Based on the basic information received by the server, additional customer data (such as financial information, past transaction history, credit information, etc.) is collected from external data sources.

[1447] Step 4:

[1448] The server performs data cleaning on the collected raw data, such as filling in missing values, removing inaccurate data, and standardizing the format.

[1449] Step 5:

[1450] The server integrates the cleaned data and manages it as a single dataset.

[1451] Step 6:

[1452] The server uses machine learning algorithms and statistical models to calculate the credit risk score of the client based on the integrated data.

[1453] Step 7:

[1454] The server visualizes the calculated credit risk score and related analytical results, generating graphs and dashboards.

[1455] Step 8:

[1456] The server sends the visualized data to the terminal so that the user can view it in an easily understandable format.

[1457] Step 9:

[1458] The terminal displays the visualized data sent from the server to the user, making it easy to understand.

[1459] Step 10:

[1460] The server generates strategic advice based on market data and competitive analysis results, such as whether to enter a new market or propose improvements to contract terms with business partners.

[1461] Step 11:

[1462] The server customizes the generated strategic advice based on the user's industry profile.

[1463] Step 12:

[1464] The device detects the user's emotional state in real time. For example, the emotion engine analyzes subtle changes in facial expressions and voice through the camera and microphone.

[1465] Step 13:

[1466] The server retrieves the recognized emotion data and adjusts the content of the strategic advice based on the emotion.

[1467] Step 14:

[1468] The server sends customized strategy advice to the terminal.

[1469] Step 15:

[1470] The device displays strategic advice and related action items to the user, helping them create a concrete action plan.

[1471] Step 16:

[1472] The user inputs feedback on the provided advice into the terminal, including actual operational results and evaluations.

[1473] Step 17:

[1474] The terminal transmits the feedback from the user to the server.

[1475] Step 18:

[1476] The server updates the analysis model based on the collected feedback to improve the accuracy of future analyses. Model training is performed to reflect new data.

[1477] Example 2

[1478] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1479] When companies make credit assessments or strategic decisions on their business partners, they need to collect and analyze massive amounts of data. However, existing systems require time and effort to clean and integrate data and calculate credit risk scores, making efficient decision-making difficult. Furthermore, they do not provide personalized advice that takes into account the user's emotional state, making it difficult to select the optimal course of action. To address these issues, a system is needed that can consistently perform everything from collecting business partner data to calculating credit risk scores, generating strategic advice, and even adjusting that advice based on the user's emotional state.

[1480] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1481] In this invention, the server includes means for acquiring basic information about trading partners, means for collecting additional data from external data sources, means for cleaning and integrating the collected data, means for calculating a credit risk score, means for visually displaying the calculated credit risk score, means for generating strategic advice based on market data and competitive analysis results, means for detecting a user's emotional state, and means for adjusting the strategic advice based on the detected emotional state, thereby enabling efficient cleaning and integration of data, calculation of a reliable credit risk score, and provision of personalized strategic advice that takes user emotions into consideration.

[1482] "Basic information" refers to basic information such as the business partner's company name, address, industry, and annual sales.

[1483] "External Data Source" refers to an external source from which the server can gather additional data about a business partner, such as a financial database or credit bureau.

[1484] "Data cleaning" refers to the process of filling in missing values ​​from collected data, removing inaccurate data, and standardizing the format.

[1485] "Data integration" refers to the process of integrating cleaned data and managing it as a single data set.

[1486] "Credit Risk Score" means a quantitative value calculated to assess the credit risk of a counterparty.

[1487] "Visually displaying" refers to visually representing the calculated credit risk score and related analytical results in the form of graphs, dashboards, etc.

[1488] "Generating strategic advice" refers to creating strategic advice such as whether to enter a new market or proposing improvements to contract terms based on market data and competitive analysis results.

[1489] "Emotional state detection" refers to analyzing the user's facial expressions and voice through a camera or microphone using an emotion engine to grasp the user's real-time emotional state.

[1490] "Adjusting strategic advice" refers to dynamically changing the content of strategic advice based on the detected emotional state of the user.

[1491] This invention relates to a new data analysis system for businesses, and presents an embodiment of a system called "Business AI Navigator" that streamlines credit risk assessment and strategic advice provision for business partners. This system has the ability to recognize the emotional state of the user and provide more personalized services.

[1492] Hardware and software used

[1493] Terminal: Input and display device (PC, tablet, smartphone, etc.)

[1494] Server: a high-performance computer for data processing and analysis

[1495] Database: Data storage for storing business partner information and data from external data sources

[1496] External data sources: financial databases and credit bureaus

[1497] Machine learning algorithms: Models for calculating credit risk scores (e.g., random forests, logistic regression)

[1498] Data visualization tools: D3.js, Tableau, etc.

[1499] Emotion recognition engine: Google Cloud Vision API, Amazon Rekognition

[1500] Processing steps

[1501] Step 1. Data collection

[1502] The user inputs basic information about the business partner (company name, address, industry, annual turnover, etc.) into the terminal. As an example, let's say the owner of a small or medium-sized business inputs information about a business partner with whom he or she wants to start a new transaction.

[1503] The device formats the basic information entered and sends it to the server in a format such as JSON or XML, using HTTP or HTTPS as the transmission protocol.

[1504] Based on the basic information received by the server, additional customer data is collected from external data sources (e.g., financial information databases, credit bureaus).

[1505] Step 2. Data analysis

[1506] The raw data collected by the server is subjected to data cleaning, which involves filling in missing financial information, removing inaccurate data, and standardizing the format.

[1507] The server consolidates the cleaned data and manages it as a single dataset, for example by eliminating duplicate data from different data sources.

[1508] The server uses the integrated data to calculate the client's credit risk score using machine learning algorithms (e.g., random forest, logistic regression model).

[1509] Step 3. View the results

[1510] The server visualizes the calculated credit risk scores and related analytical results, generating graphs and dashboards using tools such as D3.js and Tableau.

[1511] The device presents the generated visualizations to the user in an easy-to-understand manner, such as line graphs or heat maps showing rising or falling risk scores.

[1512] Step 4. Generate strategic advice

[1513] The server generates strategic advice based on market data and competitive analysis, including recommendations on whether to enter new markets and improving contract terms.

[1514] The server customizes the generated strategic advice based on the user's industry profile, including industry-specific competitive information for users in the automotive industry.

[1515] Step 5. Emotion Recognition

[1516] The device detects the user's emotional state in real time using a camera and microphone, and facial and voice data is analyzed by an emotion engine (e.g., Google Cloud Vision API, Amazon Rekognition).

[1517] The server acquires the emotion data and adjusts the strategic advice based on the user's emotions. If the user is feeling stressed, the advice will be changed to help alleviate that stress.

[1518] Step 6. Providing advice

[1519] The server sends customized strategy advice to the device, using a secure communication protocol (e.g., HTTPS).

[1520] The device will display strategic advice and related action items to the user to help them plan specific actions, for example in the form of a to-do list.

[1521] Step 7. Incorporate feedback

[1522] The user inputs feedback (actual operational results and evaluation) on the advice into the terminal. For example, the user inputs comments such as "The advice was effective" or "Improvement is needed."

[1523] The device sends the feedback to the server, which formats it in the appropriate schema and stores it in a database.

[1524] The server analyzes the feedback and updates the analysis model to improve accuracy in future runs. Model training uses machine learning frameworks such as TensorFlow and PyTorch.

[1525] In this way, the Business AI Navigator is a system that supports corporate decision-making by taking into account users' real-time emotions and providing personalized information and advice.

[1526] Prompt Sentence Examples

[1527] "Please provide a strategy for entering a new market. The market conditions are as follows: the region is North America, the target customer base is small and medium-sized businesses, and the competitors are Company A and Company B."

[1528] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1529] Step 1:

[1530] Enter basic information about your business partner

[1531] The user enters basic information about the business partner (company name, address, industry, annual sales, etc.) into the terminal.

[1532] Input: Basic information such as company name, address, industry, annual sales, etc.

[1533] The terminal formats this information and sends it to the server.

[1534] Output: Basic formatted information sent to the server.

[1535] Step 2:

[1536] External Data Collection

[1537] Based on the basic information received by the server, additional customer data is collected from external data sources (e.g., financial databases, credit bureaus).

[1538] Input: Basic formatted information

[1539] The server uses APIs to communicate with external data sources to obtain additional data (financial information, past transaction history, credit information, etc.).

[1540] Output: The additional customer data collected is stored on the server.

[1541] Step 3:

[1542] Data Cleaning

[1543] The server performs data cleaning on the collected raw data.

[1544] Input: Additional Account Data

[1545] Data cleaning includes imputing missing values, removing inaccurate data, and standardizing formats.

[1546] Output: Cleaned data is obtained.

[1547] Step 4:

[1548] Data Integration

[1549] The server integrates the cleaned data and manages it as a single dataset.

[1550] Input: Cleaned data

[1551] Eliminate duplicate data and standardize data formats.

[1552] Output: A consolidated dataset is created.

[1553] Step 5:

[1554] Credit risk score calculation

[1555] The server uses the integrated data to calculate the client's credit risk score using machine learning algorithms (e.g., random forest, logistic regression model).

[1556] Input: Integrated dataset

[1557] An algorithm analyzes the data and calculates a credit risk score.

[1558] Output: Calculated credit risk score

[1559] Step 6:

[1560] Visualizing the results

[1561] The server visualizes the calculated credit risk score and related analytical results, generating graphs and dashboards.

[1562] Input: Credit risk scores and associated analysis results

[1563] Use tools such as D3.js or Tableau to visualize the results.

[1564] Output: Visualized data

[1565] Step 7:

[1566] Presenting Visualized Data

[1567] The terminal presents the generated visualization data to the user and displays it in an easy-to-understand manner.

[1568] Input: Visualized data

[1569] Displays line graphs and heat maps showing increases or decreases in risk scores.

[1570] Output: Visualized data presented to the user

[1571] Step 8:

[1572] Strategic Advice Generation

[1573] The server generates strategic advice based on market data and competitive analysis results.

[1574] Input: Market data, competitive analysis results

[1575] This includes whether to enter new markets and proposals for improving contract terms.

[1576] Output: Strategic advice

[1577] Step 9:

[1578] Customization

[1579] The server customizes the generated strategic advice based on the user's industry profile.

[1580] Input: Strategic advice, user industry profile

[1581] For users in the automotive industry, industry-specific competitive information is added.

[1582] Output: Customized strategy advice

[1583] Step 10:

[1584] emotion recognition

[1585] The device detects the user's emotional state in real time.

[1586] Input: User's facial expressions and voice data

[1587] Using the camera and microphone, the data is analyzed by an emotion engine (e.g., Google Cloud Vision API, Amazon Rekognition).

[1588] Output: Recognized emotion data

[1589] Step 11:

[1590] Advice adjustment

[1591] The server retrieves the recognized emotion data and adjusts the content of the strategic advice.

[1592] Input: Recognized sentiment data, customized strategy advice

[1593] If you are feeling stressed, change the advice to something that will relieve you.

[1594] Output: Tailored strategy advice

[1595] Step 12:

[1596] Providing advice

[1597] The server sends the adjusted strategy advice to the terminal.

[1598] Input: Tailored Strategy Advice

[1599] Use secure communication protocols (e.g. HTTPS).

[1600] Output: Strategy advice sent to the terminal

[1601] Step 13:

[1602] Providing advice

[1603] The device displays strategic advice and related action items to the user to assist with specific action planning.

[1604] Input: Strategy Advice

[1605] Display in ToDo list format, etc.

[1606] Output: A concrete action item presented to the user

[1607] Step 14:

[1608] Enter your feedback

[1609] The user inputs feedback (actual operational results and evaluation) on the advice into the terminal.

[1610] Input: Feedback

[1611] Record the effectiveness of the advice and areas for improvement.

[1612] Output: Input feedback data

[1613] Step 15:

[1614] Send Feedback

[1615] The device sends the feedback to the server, which formats it into the appropriate schema.

[1616] Input: Feedback data

[1617] The feedback is stored in a database.

[1618] Output: Feedback data sent to the server

[1619] Step 16:

[1620] Updating a Model

[1621] The server analyzes the feedback and updates the analysis model to improve the accuracy of the analysis from the next time onwards.

[1622] Input: Submitted feedback data

[1623] Machine learning frameworks such as TensorFlow and PyTorch are used for model training.

[1624] Output: Updated analytical model

[1625] (Application example 2)

[1626] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1627] Conventional security systems determine alert levels and countermeasures based on the target's behavior and belongings. However, because there is no mechanism for analyzing and reflecting the target's emotional state in real time, appropriate responses are often delayed, which can result in increased security risks. Furthermore, conventional systems have the problem of low response accuracy because the advice they determine is uniform and lacks personalization.

[1628] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for acquiring basic information about trading partner companies, means for collecting additional trading partner data from external data sources, means for cleaning and integrating the collected data, means for calculating trading partner credit risk scores, means for visually displaying the calculated credit risk scores, means for generating strategic advice based on market data and competitive analysis results, means for analyzing the user's real-time emotional state, and means for adjusting the strategic advice based on the analyzed emotional state. This makes it possible to analyze and reflect the user's emotional state in real time and provide personalized security countermeasures.

[1629] "Means for obtaining basic information about business partners" refers to methods for collecting basic information such as the name, address, industry, and sales of business partners.

[1630] "Means for collecting additional business partner data from external data sources" refers to methods for obtaining financial information, past transaction history, credit information, etc. of business partners from external data sources.

[1631] "Means for cleaning and integrating collected data" refers to methods for preparing collected data by filling in missing values, removing inaccurate data, and standardizing the format.

[1632] The "means for calculating a customer credit risk score" is a method for calculating a score for evaluating the credit risk of a customer based on the collected and cleaned customer data.

[1633] The "means for visually displaying the calculated credit risk score" is a method for visually displaying the counterparty credit risk score as a graph or dashboard.

[1634] The "means for generating strategic advice based on market data and competitor analysis results" is a method for generating effective strategic advice based on market data and competitor analysis results.

[1635] "Means for analyzing the user's real-time emotional state" refers to a method for analyzing the user's facial expressions and tone of voice through a camera or microphone to grasp the user's emotional state in real time.

[1636] The "means for adjusting strategic advice based on the analyzed emotional state" is a method for appropriately adjusting the strategic advice provided based on the results of the user's emotional analysis.

[1637] This invention relates to a system that combines "Security Navigator," a new AI data service for businesses, with an emotion engine to analyze real-time emotional processes and support strategic security responses. This system uses smart glasses to collect visual and audio information from the target, which is then sent to a server for analysis. The following describes in detail the mode for carrying out the invention.

[1638] Hardware Configuration

[1639] 1. Smart Glasses

[1640] Visual information collection: high-resolution cameras

[1641] Audio information collection: High-sensitivity microphone

[1642] Display: Built-in display for visual display

[1643] Processor: Initial processing and formatting of data

[1644] 2. Server

[1645] High-performance cloud servers (e.g. AWS, Google Cloud)

[1646] 3. Means of communication

[1647] Real-time communication: WebSocket, MQTT

[1648] Software Configuration

[1649] 1. Emotion Recognition Engine

[1650] Google Cloud Vision API

[1651] OpenAI sentiment analysis model

[1652] 2. Data Analysis Tools

[1653] Python

[1654] TensorFlow

[1655] PyTorch

[1656] Processing flow (system program)

[1657] The server implements the invention using the following means.

[1658] 1. Data Collection

[1659] Smart glasses collect visual and audio information from the wearer.

[1660] Video data is acquired through a camera, and audio data is acquired through a microphone.

[1661] This data is formatted by a processor within the smart glasses and sent to a server in real time.

[1662] 2. Data Analysis

[1663] The server analyzes the received video and audio data and recognizes the emotional state using an emotion engine.

[1664] The emotion recognition engine uses Google Cloud Vision API and OpenAI's emotion analysis model.

[1665] The analysis results are fed back visually and audibly in real time.

[1666] 3. Displaying the results

[1667] The analysis results, showing the emotional state of the target person, are displayed on the smart glasses' display.

[1668] Depending on the emotional state, a security level such as alert or normal is indicated.

[1669] 4. Generating strategic advice

[1670] The server generates optimal responses based on the emotional state.

[1671] If the subject appears suspicious, security guards will provide real-time advice on whether increased vigilance is needed or whether further investigation is required.

[1672] 5. Providing advice

[1673] Smart glasses display real-time strategic advice.

[1674] Countermeasures will also be provided through audio guidance, allowing the wearer to respond smoothly.

[1675] 6. Incorporating feedback

[1676] Staff feedback is entered into the smart glasses via voice input or a touch panel.

[1677] This feedback is sent to the server and used to update and improve the sentiment analysis model.

[1678] Specific examples

[1679] Example 1: Commercial Intruder Screening

[1680] 1. A security guard wearing smart glasses during an intruder inspection at a commercial facility.

[1681] 2. The smart glasses send visual and audio data to the server.

[1682] 3. The server recognizes through emotion analysis that the intruder is nervous and sets the alert level to high.

[1683] 4. The smart glasses display will show "High alert state" and a voice prompt will say "Prompt for fixation confirmation."

[1684] Prompt Sentence Examples

[1685] "Based on this video and audio data, please analyze the subject's emotional state and generate an appropriate response accordingly."

[1686] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1687] Step 1:

[1688] The smart glasses collect the wearer's visual and audio information. The video data is acquired through a camera, and the audio data is acquired through a microphone. The smart glasses receive the video and audio data from the camera and microphone as input, format these data, and send them to a server in real time through a processor within the smart glasses. The output is the formatted video and audio data.

[1689] Step 2:

[1690] The server analyzes the received video and audio data. As input, it receives the video and audio data sent from the smart glasses and performs data cleaning. Specifically, it removes unnecessary noise and extracts only the necessary information. As output, it obtains the cleaned video and audio data.

[1691] Step 3:

[1692] The server uses the cleaned data to analyze the emotional state using an emotion engine. It receives the cleaned video and audio data as input, and uses Google Cloud Vision API and OpenAI emotion analysis models to recognize the user's emotional state from their facial expressions and tone of voice. The output is an analysis result that represents the user's emotional state.

[1693] Step 4:

[1694] The server generates countermeasures based on the analyzed emotion results. As input, it receives the analysis results that represent the emotional state and generates optimal security countermeasures based on them. For example, if it detects a state of tension, it sets the alert level to high. The generated countermeasures are obtained as output.

[1695] Step 5:

[1696] The results and countermeasures are displayed and provided to the wearer in real time through the smart glasses' display and audio guide. Countermeasures sent from the server are received as input and displayed on the display. Specifically, instructions such as "High alert state" and "Encourage fixation confirmation" are displayed, and the same information is provided via audio guide. Countermeasures are delivered to the wearer visually and audibly as output.

[1697] Step 6:

[1698] The wearer performs the countermeasure and provides feedback on the results through the smart glasses. As input, the wearer's feedback data is received, which is collected through the touch panel or voice input. As output, the collected feedback data is sent to the server.

[1699] Step 7:

[1700] The server analyzes the received feedback data and updates the sentiment analysis model. It takes the feedback data as input and trains the model based on it, specifically using new data to improve the sentiment analysis algorithm. The output is an updated sentiment analysis model.

[1701] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1702] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1703] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1704] [Fourth embodiment]

[1705] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1706] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1707] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1708] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1709] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1710] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1711] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1712] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1713] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1714] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1715] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1716] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1717] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1718] This invention relates to "Business AI Navigator," a new AI data service for businesses. It is a system that streamlines credit approval procedures for clients and supports strategic decision-making. A specific embodiment of this system is shown below.

[1719] Specific program behavior

[1720] 1. Data Collection

[1721] User: Enter basic information about the business partner into the terminal, such as the company name, address, industry, and annual sales.

[1722] Terminal: Sends the information entered by the user to the server.

[1723] Server: Based on the received information, collects additional customer data (e.g., financial information and transaction history) from external data sources.

[1724] 2. Data Analysis

[1725] Server: Performs data cleaning based on the collected data, such as filling in missing values, removing inaccurate data, and standardizing formats.

[1726] Server: Consolidates the cleaned data and calculates the credit risk score for each counterparty, using machine learning algorithms and statistical models.

[1727] 3. Displaying the results

[1728] Server: Visualizes the calculated credit risk scores and generates graphs and dashboards.

[1729] Terminal: Presents the generated visualization data to the user for easy understanding.

[1730] 4. Generating strategic advice

[1731] Server: Generates strategic advice based on market data and competitive analysis, such as whether to enter new markets or improve contract terms with business partners.

[1732] Server: Customizes the generated strategic advice based on the user's industry profile.

[1733] 5. Providing advice

[1734] Server: Sends customized advice to the device.

[1735] On the device: Provide user-visible advice and list required action items.

[1736] 6. Incorporating feedback

[1737] User: Enters feedback and results on the advice provided into the device.

[1738] Terminal: Sends user feedback to the server.

[1739] Server: Updates the analysis model based on the feedback, improving the accuracy of the next analysis.

[1740] Specific examples

[1741] Example 1: Credit decision

[1742] User: A small business owner considering signing a new contract with a business partner. The user uses the device to enter basic information about the business partner (company name, address, etc.).

[1743] Terminal: Sends the entered information to the server.

[1744] Server: Collects additional financial data and transaction history from external data sources, performs data cleaning and analysis, calculates credit risk scores, and visualizes the results.

[1745] Terminal: Presents visualized risk assessment results to management and provides strategic advice on whether to proceed with a transaction and risk countermeasures.

[1746] Example 2: Market trend forecasting and strategic advice

[1747] User: A marketing manager from a large company looking to enter a new market. He enters the market conditions of interest into the terminal.

[1748] Terminal: Sends entered market conditions to the server.

[1749] Server: Predicts future market trends based on market data and competitive analysis, and generates strategic advice customized based on the user's industry profile.

[1750] Terminal: Presents strategic advice and relevant action items to the Marketing Manager.

[1751] In this way, Business AI Navigator is a system that supports corporate credit operations and strategic decision-making by providing efficient and accurate information.

[1752] The processing flow will be explained below.

[1753] Step 1:

[1754] The user enters basic information about the business partner (company name, address, industry, annual sales, etc.) into the terminal.

[1755] Step 2:

[1756] The terminal formats the basic information entered by the user and sends it to the server.

[1757] Step 3:

[1758] Based on the basic information received, the server collects additional customer data from external data sources, such as financial information, past transaction history, credit information, etc.

[1759] Step 4:

[1760] The server performs data cleaning on the collected raw data, such as filling in missing values, removing inaccurate data, and standardizing the format.

[1761] Step 5:

[1762] The server integrates the cleaned data and manages it as a single dataset.

[1763] Step 6:

[1764] The server uses machine learning algorithms and statistical models to calculate the credit risk score of the client based on the integrated data.

[1765] Step 7:

[1766] The server visualizes the calculated credit risk score and related analytical results by generating graphs and charts in a visually easy-to-understand format.

[1767] Step 8:

[1768] The server sends the visualized data to the terminal so that it can be presented to the user in an easy-to-understand format.

[1769] Step 9:

[1770] The terminal displays the visualized data sent from the server to the user, allowing the user to easily understand the data.

[1771] Step 10:

[1772] The server generates strategic advice based on market data and competitive analysis, such as whether to enter a new market or suggest improvements to contract terms.

[1773] Step 11:

[1774] The server customizes the generated strategic advice based on the user's industry profile.

[1775] Step 12:

[1776] The server sends customized strategy advice to the terminal.

[1777] Step 13:

[1778] The device displays strategic advice and related action items to the user, helping them create a concrete action plan.

[1779] Step 14:

[1780] The user inputs feedback on the provided advice into the terminal, including actual operational results and evaluations.

[1781] Step 15:

[1782] The terminal transmits the feedback from the user to the server.

[1783] Step 16:

[1784] The server updates the analysis model based on the collected feedback to improve the accuracy of future analyses. Model training is performed to reflect new data.

[1785] Example 1

[1786] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1787] Conventional credit decision-making systems had problems with complex credit risk assessment of counterparties and time-consuming data integration and cleaning. They also lacked a mechanism for effectively utilizing market data and competitive analysis results necessary for strategic decision-making. Furthermore, there was an insufficient means for appropriately collecting user feedback and incorporating it into the analysis model, which could lead to a decline in analysis accuracy. There was a need to solve these problems and provide information efficiently and accurately.

[1788] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1789] In this invention, the server includes means for acquiring basic information about counterparty companies, means for collecting additional counterparty data from external data sources, means for cleaning and integrating the collected data, means for calculating counterparty credit risk scores, means for visually displaying the calculated credit risk scores, means for generating strategic advice based on market data and competitive analysis results, means for customizing the generated advice and presenting it on a terminal, means for collecting user feedback to update the analytical model, and means for securely transmitting the feedback, thereby enabling more efficient and accurate credit risk assessment and strategic decision support.

[1790] "Basic information about a business partner" refers to basic identifying information about the business partner, such as company name, location, industry, and annual sales.

[1791] "External Data Sources" refers to external databases or information services that provide additional information about business partners.

[1792] "Data cleaning" refers to the process of correcting, completing, and standardizing missing values, inaccurate data, and improper formats from collected data.

[1793] "Synthesis" refers to the process of bringing together data collected from different sources into a single data set.

[1794] "Credit risk score" refers to a numerical evaluation value of a trading partner's credit risk, and is used to evaluate the trading partner's creditworthiness.

[1795] "Visually displaying" refers to displaying credit risk scores and other analytical results in a visual format, such as a graph or dashboard.

[1796] "Market data" refers to information regarding market movements, trends, the competitive environment, etc.

[1797] "Competitive analysis results" refers to the analysis results regarding competitors' strategies, performance, market share, etc.

[1798] "Strategic advice" refers to providing advice on trading strategies with business partners and entry into new markets based on market data and competitive analysis results.

[1799] An "industry profile" refers to information that compiles standard characteristics, trends, trade practices, etc. related to a particular industry.

[1800] "Gathering feedback" refers to collecting information such as opinions, ratings, and results from users.

[1801] "Updating the analytical model" refers to readjusting machine learning algorithms and statistical models based on newly collected feedback to improve prediction and analytical accuracy.

[1802] "Securely" refers to measures to ensure safe communications and data processing and prevent information leaks and unauthorized access.

[1803] This invention relates to "Business AI Navigator," a new AI data service for businesses, which is a system that streamlines credit approval operations for clients and supports strategic decision-making. This system is implemented using the following hardware and software.

[1804] Hardware and software used

[1805] Server: Uses cloud services (e.g. Amazon Web Services, Microsoft Azure) for data processing and storage.

[1806] Device: Desktop PC, laptop, tablet, etc. The user interface is a web browser (e.g., Google Chrome, Firefox).

[1807] Software: Data visualization tools (e.g., Tableau, Power BI) and machine learning algorithms (e.g., Scikit-learn, TensorFlow) will be used.

[1808] System Operation Overview

[1809] 1. Data Collection

[1810] User: Enter basic information about a business partner into the terminal. For example, enter the company name "ABC Co., Ltd.", address "Chiyoda-ku, Tokyo," industry "manufacturing," and annual sales of "10 billion yen."

[1811] Terminal: Sends the entered information to the server. The terminal uses a secure communication protocol (HTTPS).

[1812] Server: Collect additional counterparty data (financial information, transaction history, etc.) from external data sources, using APIs.

[1813] 2. Data Analysis

[1814] Server: Cleans the collected data, imputes missing values, removes inaccurate data, and standardizes the format.

[1815] Server: Integrates clean data and calculates credit risk scores for clients. Random forest and logistic regression models are used. Machine learning algorithms include Scikit-learn and TensorFlow.

[1816] 3. Displaying the results

[1817] Server: Visualize the calculated credit risk scores and generate graphs and dashboards using Tableau or Power BI as data visualization tools.

[1818] Terminal: Presents the generated visualization data to the user in an easy-to-understand format.

[1819] 4. Generating strategic advice

[1820] Server: Generates strategic advice based on market data and competitive analysis results, such as proposing whether to enter a new market or improve contract terms with business partners.

[1821] Server: Customizes generated strategic advice based on the user's industry profile.

[1822] 5. Providing advice

[1823] Server: Sends customized advice to the device.

[1824] Terminal: Provides advice to the user and lists required action items.

[1825] 6. Incorporating feedback

[1826] User: Enters feedback and results on the advice provided into the device.

[1827] Terminal: Sends user feedback to the server.

[1828] Server: Updates the analysis model based on the feedback, improving the accuracy of the next analysis.

[1829] Specific examples

[1830] Example 1: Credit decision

[1831] User: A small business owner considering signing a new contract with a business partner. He enters basic information about the business partner into the terminal.

[1832] Terminal: Sends the entered information to the server.

[1833] Server: Collects additional financial data and transaction history from external data sources, performs data cleaning and analysis, calculates credit risk scores, and visualizes the results.

[1834] Terminal: Presents visualized risk assessment results and provides strategic advice on whether to trade and how to mitigate risks.

[1835] Example 2: Market trend forecasting and strategic advice

[1836] User: A marketing manager from a large company looking to enter a new market. He enters the market conditions of interest into the terminal.

[1837] Terminal: Sends entered market conditions to the server.

[1838] Server: Predicts future market trends based on market data and competitive analysis, and generates strategic advice customized based on the user's industry profile.

[1839] Terminal: Presents strategic advice and relevant action items to the Marketing Manager.

[1840] Prompt Sentence Examples

[1841] "Based on the company information below, please assess the credit risk of your business partner and provide strategic advice."

[1842] 1. Company Name: ABC Co., Ltd.

[1843] 2. Address: Chiyoda-ku, Tokyo

[1844] 3. Industry: Manufacturing

[1845] 4. Annual sales: 10 billion yen

[1846] In this way, the Business AI Navigator supports corporate credit operations and strategic decision-making by providing efficient and accurate information. The above is an embodiment of the present invention.

[1847] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1848] Step 1:

[1849] Data collection

[1850] User: Enters basic information about a business partner into the terminal. The entered information is company name "ABC Co., Ltd.", address "Chiyoda-ku, Tokyo," industry "manufacturing," and annual sales of "10 billion yen."

[1851] Terminal: The input information is sent to the server using a secure communication protocol (HTTPS). The input data is transferred to the server in JSON format.

[1852] Server: Collects additional customer data (financial information, transaction history, etc.) from external data sources via API. The collected data is stored in a database in an organized format by field.

[1853] Step 2:

[1854] Data Cleaning

[1855] Server: Cleans the collected data by filling in missing values, removing inaccurate data, and standardizing date and currency formats. For example, if there is missing annual sales data, it fills in the approximate value using historical data or other relevant data. It also corrects inaccurate entries (e.g., company names with trailing spaces).

[1856] Input: The raw data collected.

[1857] Output: Cleaned consolidated data.

[1858] Step 3:

[1859] Data Integration

[1860] Server: Consolidates the cleaned data into a single dataset. This process involves combining information from multiple data sources using a common key (e.g., company ID). For example, combining financial data and transaction history based on company ID.

[1861] Input: Multiple cleaned datasets.

[1862] Output: A consolidated dataset.

[1863] Step 4:

[1864] Credit risk score calculation

[1865] Server: Calculates a credit risk score based on the integrated data. This uses a machine learning algorithm (e.g., random forest, logistic regression). The integrated dataset is input into the machine learning model, and a credit risk score is output in numerical form.

[1866] Input: Unified dataset.

[1867] Output: Credit risk score.

[1868] Step 5:

[1869] Visualizing the results

[1870] Server: Generate visualization data using the calculated credit risk scores. Use a data visualization tool (e.g., Tableau, Power BI) to convert the data into a format that is easy to display as graphs or dashboards.

[1871] Terminal: Receives the visualization data sent from the server and displays it to the user through a user interface, presenting the results in the form of easy-to-understand graphs and charts.

[1872] Input: Credit Risk Score.

[1873] Output: Visualized data (graphs, dashboards).

[1874] Step 6:

[1875] Generating Strategic Advice

[1876] Server: Generates strategic advice based on market data and competitive analysis results. Using AI models, it analyzes, for example, whether to enter a new market or how to improve transaction terms, and makes specific strategic proposals.

[1877] Server: Customizes the generated advice based on the user's industry profile. Provides strategic advice suited to the user's industry characteristics.

[1878] Inputs: Market data, competitive analysis results, and your industry profile.

[1879] Output: Strategy advice.

[1880] Step 7:

[1881] Providing advice

[1882] Server: Sends customized advice to the device.

[1883] On the device: Providing advice to users and listing necessary action items, such as specific bulleted steps for entering a new market.

[1884] Enter: strategic advice.

[1885] Output: Displayed advice and action items.

[1886] Step 8:

[1887] Gathering and implementing feedback

[1888] User: Enters feedback and results on the advice provided into the device.

[1889] Terminal: Sends user feedback to the server, which then forwards the feedback to the server in an appropriate format.

[1890] Server: Updates the analysis model based on the feedback to improve the accuracy of future analyses. Retrains the machine learning model based on new feedback data.

[1891] Input: User feedback.

[1892] Output: Updated analytical model.

[1893] (Application example 1)

[1894] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1895] While existing AI data services for businesses were capable of collecting basic information on business partners and calculating credit risk, they had issues with not being able to update analytical models that take into account individualized content recommendations or user feedback. Furthermore, they were unable to provide personalized content recommendations based on users' viewing history and interests, making it difficult to improve the user experience.

[1896] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1897] In this invention, the server includes means for acquiring basic information about trading companies, means for collecting additional trading data from external sources, means for cleaning and integrating the collected data, means for assessing trading credit risk, means for visually displaying the calculated credit risk, means for generating strategic advice based on market information and competitive analysis results, means for making personalized content recommendations based on the user's viewing history and genres of interest, means for presenting the generated recommendations to the user, and means for collecting feedback on the provided recommendations and updating the analytical model. This makes it possible to improve the analytical model based on the personalized content recommendations and feedback, in addition to assessing the credit risk of trading partners.

[1898] A "trading enterprise" is a legal entity or organization that transacts business with other organizations or individuals in the course of commercial activities.

[1899] "External sources" refers to additional data that can be accessed both inside and outside the company, such as from public or private organizations or industry reports.

[1900] "Trade credit risk" is an assessment criterion that indicates the credit risk regarding the financial soundness of a trading partner and the stability of transactions in commercial activities.

[1901] "Cleaning" refers to the process of removing inaccurate data, filling in missing values, and standardizing the format prior to data analysis.

[1902] "Synthesis" refers to the process of bringing together data sets collected from different sources and converting them into a format that can be used for analysis.

[1903] "Strategic advice" refers to advice and suggestions to support business decision-making based on market information and competitive analysis results.

[1904] "Viewing history" refers to information such as the content a user has viewed in the past, the viewing time, and ratings, and is used to provide personalized services.

[1905] A "genre" is a classification of content, referring to a category that indicates a particular theme or style of film, music, book, etc.

[1906] "Feedback" refers to the ratings, opinions, and experiences you provide regarding the services and recommendations you receive.

[1907] An "analytical model" refers to a computational model that uses mathematical and statistical techniques to find patterns and make predictions or classifications based on collected data.

[1908] "Content recommendation" refers to a service that individually suggests appropriate content based on a user's interests and history information.

[1909] This system is realized through data interaction between the server, the terminal, and the user.

[1910] Data collection

[1911] 1. User: Enters information about viewing history and genres of interest into the device. For example, the user enters their recent movie viewing history and movie genres of interest (action, drama, etc.).

[1912] 2. Terminal: Sends the entered information to the server.

[1913] 3. Server: Collects additional data from external sources (e.g., viewing history and reputation data of other users).

[1914] Data analysis

[1915] 1. Server: Cleans the collected data, removing inaccurate data, filling in missing values, and standardizing the data format.

[1916] 2. Server: Train an analytics model (e.g., RandomForestClassifier) ​​based on the cleaned data. The analytics model is used to recommend personalized content based on viewing history and genre information.

[1917] Displaying the results

[1918] 1. Server: Generates graphs and dashboards to visually present the calculated recommended content.

[1919] 2. Terminal: Displays the generated visualization data and content recommendation results to the user.

[1920] Strategic Advice Generation

[1921] 1. Server: Generates personalized strategic advice based on market intelligence and competitive analysis, such as recommendations for new movies or the next series to watch.

[1922] 2. Server: Customizes strategic advice based on the user's profile.

[1923] Providing advice

[1924] 1. Server: Sends customized strategy advice to devices.

[1925] 2. On the device: Provide user-visible advice and a list of what to watch next.

[1926] Reflecting feedback

[1927] 1. User: Enters feedback into the device regarding the advice and recommended content provided.

[1928] 2. Terminal: Sends feedback information to the server.

[1929] 3. Server: Updates the analysis model based on user feedback to improve the accuracy of future content recommendations.

[1930] Examples and prompts

[1931] Example: A scenario in which a user receives recommendations for the next movie to watch based on the genre and ratings of the movies they have watched. For example, recommendations for highly rated new action movies.

[1932] Example prompt:

[1933] "Recommend a personalized list of movies based on the genres and ratings of movies the user has watched. Also provide details about the latest top-rated action movies."

[1934] This system allows users to receive personalized content quickly and accurately, improving the viewing experience, while also providing optimal recommendations by updating the analysis model based on user feedback.

[1935] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1936] Step 1:

[1937] User: The user inputs information about their viewing history and genres of interest into the device. For example, they input their recent movie viewing history and movie genres of interest (action, drama, etc.). This is the input information.

[1938] Step 2:

[1939] Device: The device sends the information entered by the user to the server, such as data about viewing history and genres of interest.

[1940] Step 3:

[1941] Server: The server collects additional data from external sources (viewing history and reputation data from other users), thereby supplementing the input information to create a comprehensive dataset.

[1942] Step 4:

[1943] Server: Cleans the collected data. Ensures consistency of the dataset by removing inaccurate data and filling in missing information. The cleaned data becomes the output information.

[1944] Step 5:

[1945] Server: Trains a generative AI model (e.g., RandomForestClassifier) ​​based on the cleaned data. This model is used to recommend personalized content based on viewing history and genre information. The trained model is the output.

[1946] Step 6:

[1947] Server: Using the trained model, the server predicts the best content recommendations for the user. This prediction is performed using input data such as viewing history and genre information. The prediction results are the output information.

[1948] Step 7:

[1949] Server: Generates graphs and dashboards to visually present the calculated recommended content. The visualized data is the output information.

[1950] Step 8:

[1951] Terminal: The terminal displays the generated visualization data and content recommendation results to the user, allowing the user to check the recommended content.

[1952] Step 9:

[1953] User: Enters feedback on the provided recommended content into the device. The feedback data becomes input information.

[1954] Step 10:

[1955] Terminal: The terminal sends the feedback from the user to the server, which passes this feedback information to the server.

[1956] Step 11:

[1957] Server: The server updates the analytical model based on user feedback, improving the accuracy of future content recommendations. The updated model becomes the output information.

[1958] This process allows users to receive personalized content quickly and accurately, improving their viewing experience. Furthermore, by updating the analysis model based on feedback, the system can consistently provide optimal recommendations.

[1959] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1960] This invention relates to a system that combines an emotion engine with "Business AI Navigator," a new AI data service for businesses. This system not only streamlines credit approval procedures for clients and supports strategic decision-making, but also provides more personalized services by recognizing users' emotions. A specific embodiment of this system is shown below.

[1961] Specific program behavior

[1962] 1. Data Collection

[1963] The user enters basic information about the business partner into the terminal, such as the company name, address, industry, and annual sales.

[1964] The terminal formats the basic information entered by the user and sends it to the server.

[1965] Based on the information received, the server collects additional customer data from external data sources, such as financial information, past transaction history, credit information, etc.

[1966] 2. Data Analysis

[1967] The server performs data cleaning on the collected raw data, such as filling in missing values, removing inaccurate data, and standardizing the format.

[1968] The server integrates the cleaned data and manages it as a single dataset.

[1969] The server uses machine learning algorithms and statistical models to calculate the credit risk score of the client based on the integrated data.

[1970] 3. Displaying the results

[1971] The server visualizes the calculated credit risk score and related analytical results, generating graphs and dashboards.

[1972] The terminal presents the generated visualization data to the user for easy understanding.

[1973] 4. Generating strategic advice

[1974] The server generates strategic advice based on market data and competitive analysis results, such as whether to enter new markets or propose improvements to contract terms.

[1975] The server customizes the generated strategic advice based on the user's industry profile.

[1976] 5. Emotion recognition

[1977] The device detects the user's emotional state in real time. For example, the emotion engine analyzes subtle changes in facial expressions and voice through the camera and microphone.

[1978] The server retrieves the recognized emotion data and adjusts the content of the strategic advice based on the emotion.

[1979] 6. Providing advice

[1980] The server sends customized strategy advice to the terminal.

[1981] The device displays strategic advice and related action items to the user, helping them create a concrete action plan.

[1982] 7. Incorporating feedback

[1983] The user inputs feedback on the provided advice into the terminal, including actual operational results and evaluations.

[1984] The terminal transmits the feedback from the user to the server.

[1985] The server updates the analysis model based on the collected feedback to improve the accuracy of future analyses. Model training is performed to reflect new data.

[1986] Specific examples

[1987] Example 1: Credit decision

[1988] User: A small business owner considering signing a new contract with a business partner. The user uses the device to enter basic information about the business partner (company name, address, etc.).

[1989] Terminal: Sends the entered information to the server.

[1990] Server: Collects additional financial data and transaction history from external data sources, performs data cleaning and analysis, calculates credit risk scores, and visualizes the results.

[1991] Terminal: Presents visualized risk assessment results to management and provides strategic advice on whether to proceed with a transaction and risk countermeasures.

[1992] Device: The emotion engine analyzes the manager's facial expressions and tone of voice, and the advice is adjusted appropriately based on the manager's emotions.

[1993] Example 2: Market trend forecasting and strategic advice

[1994] User: A marketing manager from a large company looking to enter a new market. He enters the market conditions of interest into the terminal.

[1995] Terminal: Sends entered market conditions to the server.

[1996] Server: Predicts future market trends based on market data and competitive analysis, and generates strategic advice customized based on the user's industry profile.

[1997] Terminal: The emotion engine analyzes the marketing manager's emotions in real time and reflects them in strategic advice.

[1998] Terminal: Presents strategic advice and relevant action items to the Marketing Manager.

[1999] In this way, the Business AI Navigator is a system that supports corporate decision-making by providing more personalized information and advice by taking into account the user's real-time emotions.

[2000] The processing flow will be explained below.

[2001] Step 1:

[2002] The user enters basic information about the business partner (company name, address, industry, annual sales, etc.) into the terminal.

[2003] Step 2:

[2004] The terminal formats the basic information entered by the user and sends it to the server.

[2005] Step 3:

[2006] Based on the basic information received by the server, additional customer data (such as financial information, past transaction history, credit information, etc.) is collected from external data sources.

[2007] Step 4:

[2008] The server performs data cleaning on the collected raw data, such as filling in missing values, removing inaccurate data, and standardizing the format.

[2009] Step 5:

[2010] The server integrates the cleaned data and manages it as a single dataset.

[2011] Step 6:

[2012] The server uses machine learning algorithms and statistical models to calculate the credit risk score of the client based on the integrated data.

[2013] Step 7:

[2014] The server visualizes the calculated credit risk score and related analytical results, generating graphs and dashboards.

[2015] Step 8:

[2016] The server sends the visualized data to the terminal so that the user can view it in an easily understandable format.

[2017] Step 9:

[2018] The terminal displays the visualized data sent from the server to the user, making it easy to understand.

[2019] Step 10:

[2020] The server generates strategic advice based on market data and competitive analysis results, such as whether to enter a new market or propose improvements to contract terms with business partners.

[2021] Step 11:

[2022] The server customizes the generated strategic advice based on the user's industry profile.

[2023] Step 12:

[2024] The device detects the user's emotional state in real time. For example, the emotion engine analyzes subtle changes in facial expressions and voice through the camera and microphone.

[2025] Step 13:

[2026] The server retrieves the recognized emotion data and adjusts the content of the strategic advice based on the emotion.

[2027] Step 14:

[2028] The server sends customized strategy advice to the terminal.

[2029] Step 15:

[2030] The device displays strategic advice and related action items to the user, helping them create a concrete action plan.

[2031] Step 16:

[2032] The user inputs feedback on the provided advice into the terminal, including actual operational results and evaluations.

[2033] Step 17:

[2034] The terminal transmits the feedback from the user to the server.

[2035] Step 18:

[2036] The server updates the analysis model based on the collected feedback to improve the accuracy of future analyses. Model training is performed to reflect new data.

[2037] Example 2

[2038] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[2039] When companies make credit assessments or strategic decisions on their business partners, they need to collect and analyze massive amounts of data. However, existing systems require time and effort to clean and integrate data and calculate credit risk scores, making efficient decision-making difficult. Furthermore, they do not provide personalized advice that takes into account the user's emotional state, making it difficult to select the optimal course of action. To address these issues, a system is needed that can consistently perform everything from collecting business partner data to calculating credit risk scores, generating strategic advice, and even adjusting that advice based on the user's emotional state.

[2040] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[2041] In this invention, the server includes means for acquiring basic information about trading partners, means for collecting additional data from external data sources, means for cleaning and integrating the collected data, means for calculating a credit risk score, means for visually displaying the calculated credit risk score, means for generating strategic advice based on market data and competitive analysis results, means for detecting a user's emotional state, and means for adjusting the strategic advice based on the detected emotional state, thereby enabling efficient cleaning and integration of data, calculation of a reliable credit risk score, and provision of personalized strategic advice that takes user emotions into consideration.

[2042] "Basic information" refers to basic information such as the business partner's company name, address, industry, and annual sales.

[2043] "External Data Source" refers to an external source from which the server can gather additional data about a business partner, such as a financial database or credit bureau.

[2044] "Data cleaning" refers to the process of filling in missing values ​​from collected data, removing inaccurate data, and standardizing the format.

[2045] "Data integration" refers to the process of integrating cleaned data and managing it as a single data set.

[2046] "Credit Risk Score" means a quantitative value calculated to assess the credit risk of a counterparty.

[2047] "Visually displaying" refers to visually representing the calculated credit risk score and related analytical results in the form of graphs, dashboards, etc.

[2048] "Generating strategic advice" refers to creating strategic advice such as whether to enter a new market or proposing improvements to contract terms based on market data and competitive analysis results.

[2049] "Emotional state detection" refers to analyzing the user's facial expressions and voice through a camera or microphone using an emotion engine to grasp the user's real-time emotional state.

[2050] "Adjusting strategic advice" refers to dynamically changing the content of strategic advice based on the detected emotional state of the user.

[2051] This invention relates to a new data analysis system for businesses, and presents an embodiment of a system called "Business AI Navigator" that streamlines credit risk assessment and strategic advice provision for business partners. This system has the ability to recognize the emotional state of the user and provide more personalized services.

[2052] Hardware and software used

[2053] Terminal: Input and display device (PC, tablet, smartphone, etc.)

[2054] Server: a high-performance computer for data processing and analysis

[2055] Database: Data storage for storing business partner information and data from external data sources

[2056] External data sources: financial databases and credit bureaus

[2057] Machine learning algorithms: Models for calculating credit risk scores (e.g., random forests, logistic regression)

[2058] Data visualization tools: D3.js, Tableau, etc.

[2059] Emotion recognition engine: Google Cloud Vision API, Amazon Rekognition

[2060] Processing steps

[2061] Step 1. Data collection

[2062] The user inputs basic information about the business partner (company name, address, industry, annual turnover, etc.) into the terminal. As an example, let's say the owner of a small or medium-sized business inputs information about a business partner with whom he or she wants to start a new transaction.

[2063] The device formats the basic information entered and sends it to the server in a format such as JSON or XML, using HTTP or HTTPS as the transmission protocol.

[2064] Based on the basic information received by the server, additional customer data is collected from external data sources (e.g., financial information databases, credit bureaus).

[2065] Step 2. Data analysis

[2066] The raw data collected by the server is subjected to data cleaning, which involves filling in missing financial information, removing inaccurate data, and standardizing the format.

[2067] The server consolidates the cleaned data and manages it as a single dataset, for example by eliminating duplicate data from different data sources.

[2068] The server uses the integrated data to calculate the client's credit risk score using machine learning algorithms (e.g., random forest, logistic regression model).

[2069] Step 3. View the results

[2070] The server visualizes the calculated credit risk scores and related analytical results, generating graphs and dashboards using tools such as D3.js and Tableau.

[2071] The device presents the generated visualizations to the user in an easy-to-understand manner, such as line graphs or heat maps showing rising or falling risk scores.

[2072] Step 4. Generate strategic advice

[2073] The server generates strategic advice based on market data and competitive analysis, including recommendations on whether to enter new markets and improving contract terms.

[2074] The server customizes the generated strategic advice based on the user's industry profile, including industry-specific competitive information for users in the automotive industry.

[2075] Step 5. Emotion Recognition

[2076] The device detects the user's emotional state in real time using a camera and microphone, and facial and voice data is analyzed by an emotion engine (e.g., Google Cloud Vision API, Amazon Rekognition).

[2077] The server acquires the emotion data and adjusts the strategic advice based on the user's emotions. If the user is feeling stressed, the advice will be changed to help alleviate that stress.

[2078] Step 6. Providing advice

[2079] The server sends customized strategy advice to the device, using a secure communication protocol (e.g., HTTPS).

[2080] The device will display strategic advice and related action items to the user to help them plan specific actions, for example in the form of a to-do list.

[2081] Step 7. Incorporate feedback

[2082] The user inputs feedback (actual operational results and evaluation) on the advice into the terminal. For example, the user inputs comments such as "The advice was effective" or "Improvement is needed."

[2083] The device sends the feedback to the server, which formats it in the appropriate schema and stores it in a database.

[2084] The server analyzes the feedback and updates the analysis model to improve accuracy in future runs. Model training uses machine learning frameworks such as TensorFlow and PyTorch.

[2085] In this way, the Business AI Navigator is a system that supports corporate decision-making by taking into account users' real-time emotions and providing personalized information and advice.

[2086] Prompt Sentence Examples

[2087] "Please provide a strategy for entering a new market. The market conditions are as follows: the region is North America, the target customer base is small and medium-sized businesses, and the competitors are Company A and Company B."

[2088] The flow of the identification process in the second embodiment will be described with reference to FIG.

[2089] Step 1:

[2090] Enter basic information about your business partner

[2091] The user enters basic information about the business partner (company name, address, industry, annual sales, etc.) into the terminal.

[2092] Input: Basic information such as company name, address, industry, annual sales, etc.

[2093] The terminal formats this information and sends it to the server.

[2094] Output: Basic formatted information sent to the server.

[2095] Step 2:

[2096] External Data Collection

[2097] Based on the basic information received by the server, additional customer data is collected from external data sources (e.g., financial databases, credit bureaus).

[2098] Input: Basic formatted information

[2099] The server uses APIs to communicate with external data sources to obtain additional data (financial information, past transaction history, credit information, etc.).

[2100] Output: The additional customer data collected is stored on the server.

[2101] Step 3:

[2102] Data Cleaning

[2103] The server performs data cleaning on the collected raw data.

[2104] Input: Additional Account Data

[2105] Data cleaning includes imputing missing values, removing inaccurate data, and standardizing formats.

[2106] Output: Cleaned data is obtained.

[2107] Step 4:

[2108] Data Integration

[2109] The server integrates the cleaned data and manages it as a single dataset.

[2110] Input: Cleaned data

[2111] Eliminate duplicate data and standardize data formats.

[2112] Output: A consolidated dataset is created.

[2113] Step 5:

[2114] Credit risk score calculation

[2115] The server uses the integrated data to calculate the client's credit risk score using machine learning algorithms (e.g., random forest, logistic regression model).

[2116] Input: Integrated dataset

[2117] An algorithm analyzes the data and calculates a credit risk score.

[2118] Output: Calculated credit risk score

[2119] Step 6:

[2120] Visualizing the results

[2121] The server visualizes the calculated credit risk score and related analytical results, generating graphs and dashboards.

[2122] Input: Credit risk scores and associated analysis results

[2123] Use tools such as D3.js or Tableau to visualize the results.

[2124] Output: Visualized data

[2125] Step 7:

[2126] Presenting Visualized Data

[2127] The terminal presents the generated visualization data to the user and displays it in an easy-to-understand manner.

[2128] Input: Visualized data

[2129] Displays line graphs and heat maps showing increases or decreases in risk scores.

[2130] Output: Visualized data presented to the user

[2131] Step 8:

[2132] Strategic Advice Generation

[2133] The server generates strategic advice based on market data and competitive analysis results.

[2134] Input: Market data, competitive analysis results

[2135] This includes whether to enter new markets and proposals for improving contract terms.

[2136] Output: Strategic advice

[2137] Step 9:

[2138] Customization

[2139] The server customizes the generated strategic advice based on the user's industry profile.

[2140] Input: Strategic advice, user industry profile

[2141] For users in the automotive industry, industry-specific competitive information is added.

[2142] Output: Customized strategy advice

[2143] Step 10:

[2144] emotion recognition

[2145] The device detects the user's emotional state in real time.

[2146] Input: User's facial expressions and voice data

[2147] Using the camera and microphone, the data is analyzed by an emotion engine (e.g., Google Cloud Vision API, Amazon Rekognition).

[2148] Output: Recognized emotion data

[2149] Step 11:

[2150] Advice adjustment

[2151] The server retrieves the recognized emotion data and adjusts the content of the strategic advice.

[2152] Input: Recognized sentiment data, customized strategy advice

[2153] If you are feeling stressed, change the advice to something that will relieve you.

[2154] Output: Tailored strategy advice

[2155] Step 12:

[2156] Providing advice

[2157] The server sends the adjusted strategy advice to the terminal.

[2158] Input: Tailored Strategy Advice

[2159] Use secure communication protocols (e.g. HTTPS).

[2160] Output: Strategy advice sent to the terminal

[2161] Step 13:

[2162] Providing advice

[2163] The device displays strategic advice and related action items to the user to assist with specific action planning.

[2164] Input: Strategy Advice

[2165] Display in ToDo list format, etc.

[2166] Output: A concrete action item presented to the user

[2167] Step 14:

[2168] Enter your feedback

[2169] The user inputs feedback (actual operational results and evaluation) on the advice into the terminal.

[2170] Input: Feedback

[2171] Record the effectiveness of the advice and areas for improvement.

[2172] Output: Input feedback data

[2173] Step 15:

[2174] Send Feedback

[2175] The device sends the feedback to the server, which formats it into the appropriate schema.

[2176] Input: Feedback data

[2177] The feedback is stored in a database.

[2178] Output: Feedback data sent to the server

[2179] Step 16:

[2180] Updating a Model

[2181] The server analyzes the feedback and updates the analysis model to improve the accuracy of the analysis from the next time onwards.

[2182] Input: Submitted feedback data

[2183] Machine learning frameworks such as TensorFlow and PyTorch are used for model training.

[2184] Output: Updated analytical model

[2185] (Application example 2)

[2186] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[2187] Conventional security systems determine alert levels and countermeasures based on the target's behavior and belongings. However, because there is no mechanism for analyzing and reflecting the target's emotional state in real time, appropriate responses are often delayed, which can result in increased security risks. Furthermore, conventional systems have the problem of low response accuracy because the advice they determine is uniform and lacks personalization.

[2188] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for acquiring basic information about trading partner companies, means for collecting additional trading partner data from external data sources, means for cleaning and integrating the collected data, means for calculating trading partner credit risk scores, means for visually displaying the calculated credit risk scores, means for generating strategic advice based on market data and competitive analysis results, means for analyzing the user's real-time emotional state, and means for adjusting the strategic advice based on the analyzed emotional state. This makes it possible to analyze and reflect the user's emotional state in real time and provide personalized security countermeasures.

[2189] "Means for obtaining basic information about business partners" refers to methods for collecting basic information such as the name, address, industry, and sales of business partners.

[2190] "Means for collecting additional business partner data from external data sources" refers to methods for obtaining financial information, past transaction history, credit information, etc. of business partners from external data sources.

[2191] "Means for cleaning and integrating collected data" refers to methods for preparing collected data by filling in missing values, removing inaccurate data, and standardizing the format.

[2192] The "means for calculating a customer credit risk score" is a method for calculating a score for evaluating the credit risk of a customer based on the collected and cleaned customer data.

[2193] The "means for visually displaying the calculated credit risk score" is a method for visually displaying the counterparty credit risk score as a graph or dashboard.

[2194] The "means for generating strategic advice based on market data and competitor analysis results" is a method for generating effective strategic advice based on market data and competitor analysis results.

[2195] "Means for analyzing the user's real-time emotional state" refers to a method for analyzing the user's facial expressions and tone of voice through a camera or microphone to grasp the user's emotional state in real time.

[2196] The "means for adjusting strategic advice based on the analyzed emotional state" is a method for appropriately adjusting the strategic advice provided based on the results of the user's emotional analysis.

[2197] This invention relates to a system that combines "Security Navigator," a new AI data service for businesses, with an emotion engine to analyze real-time emotional processes and support strategic security responses. This system uses smart glasses to collect visual and audio information from the target, which is then sent to a server for analysis. The following describes in detail the mode for carrying out the invention.

[2198] Hardware Configuration

[2199] 1. Smart Glasses

[2200] Visual information collection: high-resolution cameras

[2201] Audio information collection: High-sensitivity microphone

[2202] Display: Built-in display for visual display

[2203] Processor: Initial processing and formatting of data

[2204] 2. Server

[2205] High-performance cloud servers (e.g. AWS, Google Cloud)

[2206] 3. Means of communication

[2207] Real-time communication: WebSocket, MQTT

[2208] Software Configuration

[2209] 1. Emotion Recognition Engine

[2210] Google Cloud Vision API

[2211] OpenAI sentiment analysis model

[2212] 2. Data Analysis Tools

[2213] Python

[2214] TensorFlow

[2215] PyTorch

[2216] Processing flow (system program)

[2217] The server implements the invention using the following means.

[2218] 1. Data Collection

[2219] Smart glasses collect visual and audio information from the wearer.

[2220] Video data is acquired through a camera, and audio data is acquired through a microphone.

[2221] This data is formatted by a processor within the smart glasses and sent to a server in real time.

[2222] 2. Data Analysis

[2223] The server analyzes the received video and audio data and recognizes the emotional state using an emotion engine.

[2224] The emotion recognition engine uses Google Cloud Vision API and OpenAI's emotion analysis model.

[2225] The analysis results are fed back visually and audibly in real time.

[2226] 3. Displaying the results

[2227] The analysis results, showing the emotional state of the target person, are displayed on the smart glasses' display.

[2228] Depending on the emotional state, a security level such as alert or normal is indicated.

[2229] 4. Generating strategic advice

[2230] The server generates optimal responses based on the emotional state.

[2231] If the subject appears suspicious, security guards will provide real-time advice on whether increased vigilance is needed or whether further investigation is required.

[2232] 5. Providing advice

[2233] Smart glasses display real-time strategic advice.

[2234] Countermeasures will also be provided through audio guidance, allowing the wearer to respond smoothly.

[2235] 6. Incorporating feedback

[2236] Staff feedback is entered into the smart glasses via voice input or a touch panel.

[2237] This feedback is sent to the server and used to update and improve the sentiment analysis model.

[2238] Specific examples

[2239] Example 1: Commercial Intruder Screening

[2240] 1. A security guard wearing smart glasses during an intruder inspection at a commercial facility.

[2241] 2. The smart glasses send visual and audio data to the server.

[2242] 3. The server recognizes through emotion analysis that the intruder is nervous and sets the alert level to high.

[2243] 4. The smart glasses display will show "High alert state" and a voice prompt will say "Prompt for fixation confirmation."

[2244] Prompt Sentence Examples

[2245] "Based on this video and audio data, please analyze the subject's emotional state and generate an appropriate response accordingly."

[2246] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[2247] Step 1:

[2248] The smart glasses collect the wearer's visual and audio information. The video data is acquired through a camera, and the audio data is acquired through a microphone. The smart glasses receive the video and audio data from the camera and microphone as input, format these data, and send them to a server in real time through a processor within the smart glasses. The output is the formatted video and audio data.

[2249] Step 2:

[2250] The server analyzes the received video and audio data. As input, it receives the video and audio data sent from the smart glasses and performs data cleaning. Specifically, it removes unnecessary noise and extracts only the necessary information. As output, it obtains the cleaned video and audio data.

[2251] Step 3:

[2252] The server uses the cleaned data to analyze the emotional state using an emotion engine. It receives the cleaned video and audio data as input, and uses Google Cloud Vision API and OpenAI emotion analysis models to recognize the user's emotional state from their facial expressions and tone of voice. The output is an analysis result that represents the user's emotional state.

[2253] Step 4:

[2254] The server generates countermeasures based on the analyzed emotion results. As input, it receives the analysis results that represent the emotional state and generates optimal security countermeasures based on them. For example, if it detects a state of tension, it sets the alert level to high. The generated countermeasures are obtained as output.

[2255] Step 5:

[2256] The results and countermeasures are displayed and provided to the wearer in real time through the smart glasses' display and audio guide. Countermeasures sent from the server are received as input and displayed on the display. Specifically, instructions such as "High alert state" and "Encourage fixation confirmation" are displayed, and the same information is provided via audio guide. Countermeasures are delivered to the wearer visually and audibly as output.

[2257] Step 6:

[2258] The wearer performs the countermeasure and provides feedback on the results through the smart glasses. As input, the wearer's feedback data is received, which is collected through the touch panel or voice input. As output, the collected feedback data is sent to the server.

[2259] Step 7:

[2260] The server analyzes the received feedback data and updates the sentiment analysis model. It takes the feedback data as input and trains the model based on it, specifically using new data to improve the sentiment analysis algorithm. The output is an updated sentiment analysis model.

[2261] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[2262] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[2263] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[2264] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[2265] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[2266] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[2267] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[2268] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[2269] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[2270] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[2271] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[2272] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[2273] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[2274] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[2275] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[2276] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[2277] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[2278] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[2279] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[2280] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[2281] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[2282] The following is further disclosed regarding the above embodiment.

[2283] (Claim 1)

[2284] A means of obtaining basic information about business partners,

[2285] a means of collecting additional account data from external data sources; and

[2286] a means for cleaning and synthesizing the collected data;

[2287] a means for calculating a counterparty credit risk score;

[2288] a means for visually displaying the calculated credit risk score;

[2289] A system that includes a means for generating strategic advice based on market data and competitive analysis results.

[2290] (Claim 2)

[2291] 10. The system of claim 1, further comprising: means for customizing strategic advice based on a user's industry profile.

[2292] (Claim 3)

[2293] 10. The system of claim 1, further comprising means for collecting and analyzing feedback from users and updating the model.

[2294] "Example 1"

[2295] (Claim 1)

[2296] A means of obtaining basic information about business partners,

[2297] a means of collecting additional account data from external data sources; and

[2298] a means for cleaning and synthesizing the collected data;

[2299] a means for calculating a counterparty credit risk score;

[2300] a means for visually displaying the calculated credit risk score;

[2301] A means of generating strategic advice based on market data and competitive analysis results;

[2302] a means for customizing the generated advice and presenting it on the device;

[2303] a means of collecting user feedback and updating the analytical model;

[2304] A secure way to submit feedback

[2305] A system including:

[2306] (Claim 2)

[2307] 10. The system of claim 1, further comprising: means for customizing strategic advice based on a user's industry profile.

[2308] (Claim 3)

[2309] 10. The system of claim 1, further comprising means for collecting and analyzing feedback from users and updating the model.

[2310] "Application Example 1"

[2311] (Claim 1)

[2312] A means of obtaining basic information about trading companies,

[2313] a means of collecting additional transaction data from external sources;

[2314] a means for cleaning and synthesizing the collected data;

[2315] a means of assessing trade credit risk;

[2316] a means for visually displaying the calculated credit risk;

[2317] A means of generating strategic advice based on market information and competitive analysis results;

[2318] a means for personalized content recommendations based on a user's viewing history and genre interests;

[2319] a means for presenting the generated recommendations to a user;

[2320] The system includes a means for collecting feedback on the recommendations provided and updating the analytical model.

[2321] (Claim 2)

[2322] 10. The system of claim 1, further comprising: means for customizing strategic advice based on a user's industry profile.

[2323] (Claim 3)

[2324] 10. The system of claim 1, further comprising means for collecting and analyzing feedback from users and updating the recommendation algorithm.

[2325] "Example 2: Combining Emotion Engines"

[2326] (Claim 1)

[2327] A means of obtaining basic information about business partners;

[2328] a means of collecting additional data from external data sources; and

[2329] a means for cleaning and synthesizing the collected data;

[2330] a means for calculating a credit risk score;

[2331] a means for visually displaying the calculated credit risk score;

[2332] A means of generating strategic advice based on market data and competitive analysis results;

[2333] means for detecting an emotional state of a user;

[2334] The system includes means for adjusting strategic advice based on the detected emotional state.

[2335] (Claim 2)

[2336] 10. The system of claim 1, further comprising: means for customizing strategic advice based on a user's industry profile.

[2337] (Claim 3)

[2338] 10. The system of claim 1, further comprising means for collecting and analyzing feedback from users and updating the analytical model.

[2339] "Application example 2 when combining emotion engines"

[2340] (Claim 1)

[2341] A means of obtaining basic information about business partners,

[2342] a means of collecting additional account data from external data sources; and

[2343] a means for cleaning and synthesizing the collected data;

[2344] a means for calculating a counterparty credit risk score;

[2345] a means for visually displaying the calculated credit risk score;

[2346] A means of generating strategic advice based on market data and competitive analysis results;

[2347] means for analyzing a user's real-time emotional state;

[2348] A means for tailoring strategic advice based on analyzed emotional states

[2349] A system including:

[2350] (Claim 2)

[2351] 10. The system of claim 1, further comprising: means for customizing strategic advice based on a user's industry profile.

[2352] (Claim 3)

[2353] 10. The system of claim 1, further comprising means for collecting and analyzing feedback from users and updating the model. [Explanation of symbols]

[2354] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of obtaining basic information about business partners, a means of collecting additional account data from external data sources; and a means for cleaning and synthesizing the collected data; a means for calculating a counterparty credit risk score; a means for visually displaying the calculated credit risk score; A system that includes a means for generating strategic advice based on market data and competitive analysis results.

2. The system of claim 1 , further comprising: means for customizing strategic advice based on a user's industry profile.

3. 10. The system of claim 1, further comprising means for collecting and analyzing feedback from users and updating the model.

Images