system

Abstract

Provide a system. 【Solution means】Means for collecting regional data, Means for storing the collected data in a database, Means for performing data cleansing on the stored data, Means for extracting feature quantities from the data, Means for training a generative artificial intelligence model using the extracted feature quantities, Means for analyzing new data using the trained model and predicting the market scale, Means for visualizing and displaying the prediction results, Means for proposing staffing arrangements based on the analysis results, A system including the above.

Description

Technical Field

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

Background Art

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

Summary of the Invention

Problems to be Solved by the Invention

[0004] In conventional market analysis methods, it has been difficult to accurately grasp the specific market scale for each region and the potential of untapped markets. Also, the process from data collection, cleaning, analysis, visualization, to the final proposal of personnel allocation was time-consuming and required a lot of resources. As a result, efficient personnel allocation was difficult, and the optimization of management resources was hindered. The present invention aims to solve these problems and provide a system that realizes efficient and highly accurate market analysis and optimization of personnel allocation.

Means for Solving the Problems

[0005] To solve this problem, the present invention provides the following means.

[0006] 1. Establish a means to collect regional data, and collect diverse data such as purchasing data, population information, and economic data related to specific regions.

[0007] 2. Provide a means to store the collected data in a database, enabling centralized management.

[0008] 3. Provide a means to perform data cleansing on the stored data to reduce unnecessary noise and impute missing values.

[0009] 4. Establish a means to extract features from the data and identify useful information such as demographic information, consumer behavior, and regional characteristics.

[0010] 5. A means is provided to train a generative artificial intelligence model using the extracted features, and the accuracy of the model is improved using historical data.

[0011] 6. Establish a means to analyze new data using a pre-trained model and predict market size and the potential of untapped markets.

[0012] 7. Provide a means to visualize and display the prediction results, and present them to users in an easy-to-understand format using graphs and maps.

[0013] 8. Provide a means to propose personnel allocation based on the analysis results, thereby offering users an efficient and effective personnel allocation plan.

[0014] These methods streamline the entire process from market analysis to personnel allocation planning, providing a system that enables highly accurate optimization of management resources.

[0015] 1. "Regional data" refers to purchasing data, demographic information, economic data, and other related data associated with a specific region.

[0016] 2. "Means of collection" refers to methods or tools for acquiring local data.

[0017] 3. The "database" refers to a structured digital storage system for storing and managing the collected data.

[0018] 4. The "data cleansing means" refers to a method or tool for reducing unnecessary noise from data and complementing missing values.

[0019] 5. The "feature extraction means" refers to a method or tool for identifying and extracting useful features from data.

[0020] 6. The "generative AI model" refers to a machine learning model trained to analyze data and generate specific predictions or insights.

[0021] 7. The "training means" refers to a method or tool for training a model using past data to improve the performance of the generative AI model.

[0022] 8. The "analysis means" refers to a method or tool for predicting the market size with a generative AI model using new data.

[0023] 9. The "visualization means" refers to a method or tool for visually representing the analysis results and displaying them in an easy-to-understand manner for the user.

[0024] 10. The "personnel allocation proposal means" refers to a method or tool for creating an optimal personnel allocation plan based on the analysis results and providing it to the user.

Brief Description of Drawings

[0025] [Figure 1] It is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] It is a conceptual diagram showing an example of the main functions of a data processing device and a smart device according to the first embodiment. [Figure 3]This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11] This is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] This is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] This is a sequence diagram showing the processing flow of the data processing system in Example 2, which incorporates an emotion engine. [Figure 14] This is a sequence diagram showing the processing flow of the data processing system in Application Example 2, which combines an emotion engine. [Modes for carrying out the invention]

[0026] Hereinafter, an example of an embodiment of the system relating to the technology of this disclosure will be described with reference to the attached drawings.

[0027] First, let's explain the terminology used in the following explanation.

[0028] In the following embodiments, the signed processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Furthermore, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include CPU (Central Processing Unit), GPU (Graphics Processing Unit), GPGPU (General-Purpose computing on Graphics Processing Units), and APU (Accelerated Processing Unit).

[0029] In the following embodiments, signed RAM (Random Access Memory) is a memory that temporarily stores information and is used as work memory by the processor.

[0030] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0031] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0032] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0033] [First Embodiment]

[0034] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0035] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0036] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0037] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0038] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0039] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0040] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0041] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0042] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0043] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0044] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0045] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0046] This invention relates to a system for collecting and analyzing regional data to provide effective market analysis and personnel allocation suggestions. The processing of this system's program is described below in natural language.

[0047] 1. Data Collection

[0048] User: Specify the region to be analyzed and the required data types (e.g., purchasing data, demographic information, economic data).

[0049] Terminal: Collects necessary data from the specified data source and sends it to the server.

[0050] Server: Stores collected data in a central database. This ensures that all data is managed centrally.

[0051] 2. Data preprocessing

[0052] Server: Perform data cleansing to reduce unwanted noise. Detect and remove inaccurate or abnormal data.

[0053] Server: Imputes missing data. For example, it may use imputation methods such as imputing with the mean or median, or using machine learning models.

[0054] 3. Feature Extraction

[0055] Server: Extracts features from pre-processed data. Specifically, it extracts demographic information (age, gender, income level, etc.), consumer behavior (purchase frequency, type of purchased goods, average purchase amount, etc.), and regional characteristics (infrastructure status, geographical characteristics, number of competing companies, etc.).

[0056] Server: Stores the extracted features as training data.

[0057] 4. Model training

[0058] Server: Trains a generative artificial intelligence model using the extracted features. The training process includes splitting the data (training set, validation set, test set).

[0059] Server: Evaluates the trained model and adjusts hyperparameters as needed.

[0060] 5. Analysis and Prediction

[0061] Server: Uses a pre-trained model to analyze new data. This analysis predicts the market size and potential of untapped markets in each region.

[0062] Server: Organizes the analysis results and saves them to the database.

[0063] 6. Visualization of Results

[0064] Server: Generates data to visualize analysis results in an easy-to-understand format (e.g., graphs, maps, dashboards).

[0065] Terminal: Displays visualized data to the user. The user understands the specific analysis results by viewing graphs and maps.

[0066] 7. Proposal for staffing arrangements

[0067] Server: Based on the analysis results, propose effective staffing arrangements. These proposals will take into account the market size and purchasing power of each region.

[0068] Terminal: Displays the proposed staffing plan to the user. The user can then create an optimal staffing plan based on this proposal.

[0069] Specific example

[0070] For example, if a user selects Tokyo and Osaka as target regions, the terminal collects purchasing data, population information, and economic data for Tokyo and Osaka. The server stores this data in a database and performs data cleansing and feature extraction. A generative artificial intelligence model is trained using the extracted features, and new data is analyzed by the model. As a result, the market size and potential of untapped markets in Tokyo and Osaka are predicted. The server visualizes the prediction results and displays them to the user through the terminal. Based on these results, the user can plan to deploy 12 sales staff in Tokyo and 8 sales staff in Osaka. In this way, efficient market analysis and personnel allocation are achieved.

[0071] Thus, the present invention provides a consistent process from the collection and analysis of regional data to the visualization of results and the proposal of personnel allocation, resulting in significant benefits and convenience.

[0072] The following describes the processing flow.

[0073] Step 1: Data Collection

[0074] User: Specify the region to be analyzed and the required data types (e.g., purchasing data, demographic information, economic data).

[0075] Terminal: Collects data from the specified data source (e.g., government statistics database, commercial database, web API, etc.).

[0076] Terminal: Sends collected data to the server.

[0077] Server: Stores received data in a database for later storage.

[0078] Step 2: Data Preprocessing

[0079] Server: Perform data cleansing. Detect and delete or correct inaccurate or abnormal data.

[0080] Server: Performs processing to impute missing data. For example, it can impute missing data with the mean or median, or use a machine learning model to calculate predicted values.

[0081] Server: Updates the database to store the data after preprocessing is complete.

[0082] Step 3: Feature Extraction

[0083] Server: Extracts useful features from pre-processed data. Specifically, it extracts demographic information (age, gender, income level, etc.), consumer behavior (purchase frequency, type of purchased goods, average purchase amount, etc.), and regional characteristics (infrastructure status, geographical characteristics, number of competing companies, etc.).

[0084] Server: Organizes the extracted features and saves them as training data.

[0085] Step 4: Model Training

[0086] Server: Trains a generative artificial intelligence model using the extracted features. The training process includes splitting the data (training set, validation set, test set).

[0087] Server: Evaluates the model's performance and adjusts hyperparameters as needed.

[0088] Server: Stores the trained model.

[0089] Step 5: Analysis and Prediction

[0090] Server: Performs analysis on new data using the trained model.

[0091] Server: Predicts market size and untapped market potential in each region.

[0092] Server: Organizes the prediction results and saves them to the database.

[0093] Step 6: Visualizing the Results

[0094] Server: Generates data to visualize prediction results in an easy-to-understand format (e.g., graphs, maps, dashboards).

[0095] Terminal: Displays visualized data to the user. The user understands the specific analysis results by viewing graphs and maps.

[0096] Step 7: Propose staffing arrangements

[0097] Server: Based on the analysis results, proposes effective staffing arrangements. These proposals take into account regional market size and purchasing power.

[0098] Terminal: Displays the proposed staffing plan to the user. The user then develops the optimal staffing plan based on this.

[0099] For example, if a user designates Tokyo and Osaka as target regions, the terminal collects purchasing data, demographic information, and economic data for these regions. The server performs data cleansing and feature extraction, and trains a generative artificial intelligence model. The server then analyzes the new data to predict the market size and untapped markets for Tokyo and Osaka. Finally, the server visualizes the prediction results and displays them to the user through the terminal. Based on the proposed staffing plan, the user effectively allocates 12 sales staff to Tokyo and 8 to Osaka.

[0100] (Example 1)

[0101] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0102] Traditional marketing analysis methods made it difficult to collect and analyze data that differed by region and to allocate personnel effectively. Furthermore, there was no system that consistently handled data cleansing, data reconciliation, and visualization of analysis results. As a result, companies were unable to conduct efficient market analysis and optimal personnel allocation, hindering business efficiency and accurate understanding of target markets.

[0103] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0104] In this invention, the server includes means for specifying the region and data type to be analyzed; means for collecting the specified data; means for storing the collected data in a database; means for performing data cleansing on the stored data to remove inaccurate data and outliers; means for imputing missing data; means for extracting features from the data; means for training a generative artificial intelligence model using the extracted features; means for analyzing new data and predicting market size using the trained artificial intelligence model; means for storing and visualizing the prediction results in a database; means for proposing staffing arrangements based on the analysis results; and means for displaying the proposed staffing arrangements. This makes it possible to consistently perform detailed market analysis and effective staffing arrangements for each region.

[0105] "Analysis target area" refers to the specific region that the system targets when performing marketing analysis.

[0106] "Data type" refers to the various types of data necessary for analysis in the collection of regional data. Examples include purchasing data, population information, and economic data.

[0107] "Data cleansing" refers to the process of detecting inaccurate information and outliers contained in collected data, and then removing or correcting them.

[0108] "Missing data" refers to information that is missing from a dataset.

[0109] "Features" refer to data attributes or variables that have significant meaning for data analysis.

[0110] A "generative artificial intelligence model" refers to a model that uses machine learning and deep learning techniques to make predictions and classifications on new data.

[0111] "Market size" refers to the expected sales or purchasing power in a particular region or market.

[0112] "Visualization" refers to the process of visually displaying data and analysis results using graphs, dashboards, maps, and other visual tools.

[0113] "Personnel allocation" refers to the process by which a company or organization distributes the necessary number of staff members to appropriate locations and roles in order to effectively carry out its operations.

[0114] This invention relates to a system for collecting and analyzing regional data to provide effective market analysis and personnel allocation suggestions.

[0115] System program configuration

[0116] This system begins with the user specifying the analysis area and data type. The terminal collects data and sends it to the server, which processes the collected data to propose market analysis and staffing strategies. This entire process involves multiple hardware and software components.

[0117] Hardware:

[0118] Device: A computer or smartphone that can access the internet or APIs for data collection.

[0119] Server: A high-performance computer used for data processing, storage, and analysis.

[0120] software:

[0121] Database management systems: Used to centrally manage collected data. Examples include MySQL® and PostgreSQL.

[0122] Data cleansing tools: These cleanse the collected data. For example, the pandas library in Python.

[0123] Generative AI models: Machine learning models that perform market analysis using features. Examples include Scikit-learn and TENSORFLOW®.

[0124] Visualization tools: Tools for visualizing and displaying analysis results. Examples include Tableau and D3.js.

[0125] Program processing flow

[0126] The system program performs the following steps:

[0127] 1. Data collection:

[0128] User: Specify the region to be analyzed and the required data types (purchase data, demographic information, economic data, etc.).

[0129] Terminal: Collects necessary data from specified data sources (open data portals or commercial databases) and sends it to the server.

[0130] Server: Stores data sent from terminals in a central database and manages it centrally.

[0131] 2. Data preprocessing:

[0132] Server: Performs data cleansing on the collected data, removing inaccurate data and outliers.

[0133] Server: Imputes missing data using mean values ​​and machine learning models.

[0134] 3. Feature extraction:

[0135] Server: Extracts features such as demographic information, consumer behavior, and regional characteristics from pre-processed data and saves them as training datasets.

[0136] 4. Model training:

[0137] Server: Trains a generative artificial intelligence model using the extracted feature data. If necessary, splits the data into training, validation, and test sets, and tunes hyperparameters.

[0138] 5. Analysis and Prediction:

[0139] Server: Uses a pre-trained model to analyze new data and predict market size and the potential of untapped markets.

[0140] Server: Organizes the prediction results and saves them to the database.

[0141] 6. Visualization of results:

[0142] Server: Generates visualization data such as graphs and maps based on the analysis results.

[0143] Terminal: Displays dashboards and graphs to the user based on visualization data sent from the server.

[0144] 7. Proposal for staffing arrangements:

[0145] Server: Based on the analysis results, propose effective staffing arrangements.

[0146] Terminal: Displays the proposed staffing plan to the user.

[0147] Explanation of specific examples

[0148] For example, if a user selects Tokyo and Osaka as target regions, the device collects purchasing data, demographic information, and economic data for Tokyo and Osaka through open data portals and commercial databases. The collected data is sent to a server and stored in a central database. Next, the server performs data cleansing, removing inaccurate data and filling in missing data. After that, the server extracts features such as purchase frequency and average purchase amount and stores them as a training dataset. Using the stored data, a generative AI model is trained to make predictions on new data. The analysis results are displayed visually, and the server proposes a staffing plan suitable for the user. Based on this information, the user can plan, for example, to deploy 12 sales staff in Tokyo and 8 in Osaka.

[0149] Example of a prompt

[0150] "Conduct a market analysis using purchasing data, demographic information, and economic data for Tokyo and Osaka, and propose effective staffing strategies."

[0151] Thus, the present invention provides a consistent process from the collection and analysis of regional data to the visualization of results and the proposal of personnel allocation, resulting in significant benefits and convenience.

[0152] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0153] Step 1:

[0154] User: Specify the region and data type to be analyzed. Specifically, the user uses a terminal to select "Tokyo purchasing data and population information" through the interface.

[0155] Input: Region to be analyzed and data type.

[0156] Output: Field information required for the data collection process.

[0157] Step 2:

[0158] Terminal: Collects data via the internet or APIs based on the user-specified analysis area and data type. For example, it accesses "open data portals" or "commercial databases" to obtain information.

[0159] Input: Region to be analyzed and data type.

[0160] Output: Collected raw data.

[0161] Specific operation: A request is sent to the "Open Data Portal API" to retrieve purchasing data for Tokyo.

[0162] Step 3:

[0163] Terminal: Sends collected data to the server.

[0164] Input: Collected raw data.

[0165] Output: Request containing collected data.

[0166] Specific operation: The collected data is sent to the server as an HTTP request via the API.

[0167] Step 4:

[0168] Server: Stores collected data in a central database. This ensures centralized data management.

[0169] Input: Collected data sent to the server.

[0170] Output: Data stored in the central database.

[0171] Specific operation: Use a database management system (e.g., MySQL) to insert data into a table.

[0172] Step 5:

[0173] Server: Perform data cleansing to remove inaccurate data and outliers. For example, remove outliers like "-1 yen" from purchase history data.

[0174] Input: Raw data stored in the central database.

[0175] Output: Cleansed data.

[0176] Specific operation: Use the Python pandas library to filter and remove inaccurate entries in the data.

[0177] Step 6:

[0178] Server: Imputes missing data using mean values ​​or machine learning models. For example, it can imputate missing age data using the average age of other data points.

[0179] Input: Cleansed data (including missing values).

[0180] Output: Interpolated data.

[0181] Specific operation: Missing values ​​are imputed using the K-nearest neighbors algorithm with the scikit-learn library in Python.

[0182] Step 7:

[0183] Server: Extracts features from pre-processed data. Specifically, it extracts demographic information such as age and income, and consumer behavior data such as purchase frequency and purchase amount.

[0184] Input: Completed data.

[0185] Output: Feature dataset.

[0186] Specific operation: Using the Python Feature-engine library, select important features and convert the data into a matrix format.

[0187] Step 8:

[0188] Server: The extracted features are used to split the data into training, validation, and test sets, and a generative artificial intelligence model is trained. Specifically, 70% is split into a training set, 15% into a validation set, and 15% into a test set, and a regression model is trained.

[0189] Input: Feature dataset.

[0190] Output: Trained model.

[0191] Specific operation: The data is split using the scikit-learn library, and a Linear Regression model is trained.

[0192] Step 9:

[0193] Server: Uses a pre-trained model to analyze new data and predict market size by region.

[0194] Input: A trained model and new data.

[0195] Output: Prediction result.

[0196] Specific actions: Input new data into the model and perform regression analysis to predict market size.

[0197] Step 10:

[0198] Server: Organizes and stores prediction results in a database.

[0199] Input: Prediction result.

[0200] Output: Prediction results stored in the database.

[0201] Specific operation: Use an SQL query to save the prediction results to the appropriate table in the database.

[0202] Step 11:

[0203] Server: Converts analysis results into visualization data such as graphs and maps.

[0204] Input: Prediction results stored in the database.

[0205] Output: Visualized data.

[0206] Specific operation: Use Matplotlib and D3.js to convert the prediction results into a visually displayable format.

[0207] Step 12:

[0208] Terminal: Displays dashboards and graphs to the user based on visualization data sent from the server.

[0209] Input: Visualization data from the server.

[0210] Output: Graphs and maps displayed in the user interface.

[0211] Specific operation: Use a web application framework (e.g., React) to dynamically display data in the user interface.

[0212] Step 13:

[0213] Server: Based on the analysis results, it proposes effective personnel allocation for each region.

[0214] Input: Analysis results.

[0215] Output: Staffing proposal.

[0216] Specific operation: Based on the prediction results, calculate the optimal staffing allocation for each region and generate a proposal.

[0217] Step 14:

[0218] Terminal: Displays the proposed staffing plan to the user. The user can make decisions based on the proposal.

[0219] Input: Personnel allocation proposal.

[0220] Output: Staffing suggestions displayed in the user interface.

[0221] Specific operation: Use a web application framework to display the contents of the proposal in a user interface.

[0222] (Application Example 1)

[0223] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0224] Efficient market analysis and staffing planning in physical stores are not easy, and responding quickly to changes in market trends is particularly challenging. Furthermore, data inaccuracies that occur during the process of collecting and analyzing regional data, as well as a lack of means to support visual decision-making, are also challenges.

[0225] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0226] In this invention, the server includes means for collecting regional data, means for storing the collected data in a database, means for performing data cleansing on the stored data, means for extracting features from the data, means for training a generative artificial intelligence model using the extracted features, means for analyzing new data using the trained model and predicting market size, means for visualizing and displaying the prediction results, means for proposing staffing arrangements based on the analysis results, means for notifying changes in market trends for a specific region in real time, and means for displaying the prediction results on an electronic map to support visual judgments for store placement and market analysis. This enables efficient market analysis and staffing planning, and allows for quick and accurate responses to changes in market trends.

[0227] "Regional data" refers to data that includes population information, economic data, purchasing data, and related infrastructure information for a specific geographical area.

[0228] A "database" is a system for centrally managing and storing collected data.

[0229] "Data cleansing" is a technique that improves the quality of collected data by removing noise and inaccurate data.

[0230] "Features" refer to statistically extracted information necessary for data analysis, and are primarily used as input for analytical models.

[0231] A "generative artificial intelligence model" refers to a machine learning algorithm that performs predictions and analyses based on collected and processed data.

[0232] "Prediction" refers to estimating future data and trends using a trained artificial intelligence model.

[0233] "Visualization" refers to displaying analysis results in a visually easy-to-understand format, such as graphs, maps, and dashboards.

[0234] "Personnel allocation" refers to planning the placement of personnel in the most suitable locations and positions.

[0235] "Real-time notifications" is a feature that instantly informs users of market trends and other data changes.

[0236] An "electronic map" is a digital map that visually displays geographical information and forecast results.

[0237] This invention relates to a system for collecting and analyzing regional data to perform market analysis and propose personnel allocation strategies. The system functions through the cooperation of a server, terminals, and users.

[0238] 1. Data Collection

[0239] The user specifies the region to be analyzed and the required data types (e.g., purchasing data, demographic information, economic data). The terminal collects the necessary data from the specified data sources and sends it to the server. The server stores the collected data in a central database and manages all data centrally.

[0240] 2. Data preprocessing

[0241] The server performs data cleansing to reduce unnecessary noise. Specifically, it detects and removes inaccurate and anomalous data. It also uses methods such as imputation with the mean or median, or imputation using machine learning models, to fill in missing data.

[0242] 3. Feature Extraction

[0243] The server extracts features from the pre-processed data. Specifically, it extracts demographic information (age, gender, income level, etc.) and consumer behavior (purchase frequency, type of purchased goods, average purchase amount, etc.). It also includes regional characteristics (infrastructure status, geographical features, number of competing companies, etc.). The extracted features are stored as training data.

[0244] 4. Model training

[0245] The server uses the extracted features to train a generative artificial intelligence model. The training process includes splitting the data (training set, validation set, test set). The trained model is evaluated, and hyperparameters are adjusted as needed.

[0246] 5. Analysis and Prediction

[0247] The server uses a pre-trained model to analyze new data. This analysis predicts the market size and potential of untapped markets in each region. The analysis results are stored in a database for reuse.

[0248] 6. Visualization of Results

[0249] The server generates data that visualizes the analysis results in an easy-to-understand format (e.g., graphs, maps, dashboards). The terminal displays this visualized data to the user. This allows the user to understand the specific analysis results while viewing graphs and maps.

[0250] 7. Proposal for staffing arrangements

[0251] Based on the analysis results, the server proposes an effective staffing plan. This proposal takes into account the market size and purchasing power of each region. The terminal displays the proposed staffing plan to the user, helping them to create the optimal staffing plan.

[0252] It also includes a feature that provides real-time notifications of changes in market trends in specific regions. This feature provides alerts to quickly respond to sudden market fluctuations. Furthermore, it can display forecast results on an electronic map, visually supporting store placement and market analysis. This allows users to make more intuitive decisions.

[0253] Specific example

[0254] For example, suppose a user collects purchasing data, population data, and competitor information based on location data in Tokyo to perform market analysis. In this case, the server collects and preprocesses this data, extracts features, and trains an AI model. This model predicts the market size of Tokyo, and the results are visually presented to the user in the form of graphs and maps. Then, the optimal staffing allocation for stores in Tokyo is suggested.

[0255] Examples of prompt statements are as follows:

[0256] "Based on location data for Tokyo, collect in-store purchasing data, population information, and competitor information to conduct market analysis. Then, create an application that proposes appropriate staffing arrangements based on the results."

[0257] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0258] Step 1:

[0259] Data collection

[0260] The user specifies the target region and the required data type (purchase data, population information, economic data, etc.). The terminal collects the necessary data from the specified data source (public API, website, statistical database, etc.) and sends the collected data to the server. The server stores the received data in a central database.

[0261] Input: User-specified region and data type

[0262] Output: Regional data stored on the server

[0263] Step 2:

[0264] Data preprocessing

[0265] The server performs data cleansing on the received data. Specifically, it normalizes unstructured data, reduces noisy data, and detects and removes inaccurate or outlier values. Next, to impute missing data, it uses the mean or median, or applies imputation methods using machine learning models.

[0266] Input: Raw data stored in the central database

[0267] Output: Cleansed and imputed data

[0268] Step 3:

[0269] Feature extraction

[0270] The server extracts features from the pre-processed data. Specifically, it extracts demographic information (age, gender, income level, etc.), consumer behavior (purchase frequency, type of purchased goods, average purchase amount, etc.), and regional characteristics (infrastructure status, geographical characteristics, number of competing stores, etc.), and stores them as training data.

[0271] Input: Pre-processed data

[0272] Output: Training data with extracted features

[0273] Step 4:

[0274] Model training

[0275] The server trains a generative artificial intelligence model using the extracted feature quantities. In this process, the training data is split into a training set, a validation set, and a test set, and appropriate hyperparameters are selected to optimize the model. The trained model is used for subsequent analysis and prediction.

[0276] Input: Training data with extracted feature quantities

[0277] Output: Trained generative artificial intelligence model

[0278] Step 5:

[0279] Analysis and prediction

[0280] The server analyzes new data using the trained generative artificial intelligence model. Through this analysis, the market scale of each region and the potential of untapped markets are predicted, and these prediction results are saved in a database.

[0281] Input: Trained generative artificial intelligence model, new data to be analyzed

[0282] Output: Prediction results saved in the database

[0283] Step 6:

[0284] Visualization of results

[0285] The server generates data for visualizing the prediction results in forms such as graphs, maps, dashboards, etc. The terminal displays these visualization data to the user, and the user can visually check the analysis results.

[0286] Input: Prediction results saved in the database

[0287] Output: Visualized analysis results

[0288] Step 7:

[0289] Proposal for staffing arrangements

[0290] The server proposes an effective staffing plan based on the analysis results. The proposal takes into account the market size and purchasing power of each region. The terminal displays the proposed staffing plan to the user, allowing the user to create the optimal staffing plan.

[0291] Input: Visualized analysis results

[0292] Output: Proposed staffing arrangements

[0293] Step 8:

[0294] Real-time notifications

[0295] The server monitors market trends in a specific region in real time and notifies the user when a change is detected. This allows the user to respond quickly and accurately to market changes.

[0296] Input: Real-time monitored market data

[0297] Output: Notification of market changes to users

[0298] Step 9:

[0299] Use of electronic maps

[0300] The server generates data to display prediction results on an electronic map, and the terminal displays this to the user. This allows the user to visually perform store placement and market analysis.

[0301] Input: Prediction result

[0302] Output: Prediction results displayed on the electronic map

[0303] Furthermore, an emotion engine for estimating the user's emotions may be combined. That is, the specific processing unit 290 may estimate the user's emotions using the emotion recognition model 59 and perform specific processing using the user's emotions.

[0304] This invention combines an emotion engine that recognizes the user's emotions with a system that collects and analyzes regional data and performs effective market analysis and personnel allocation proposals. The processing of the program of this system will be described in natural language below.

[0305] 1. Data Collection

[0306] User: Specify the target area for analysis and the necessary data types (e.g., purchase data, population information, economic data).

[0307] Terminal: Collect the necessary data from the specified data source and send it to the server.

[0308] Server: Save the received data in the central database. In this way, all data is centrally managed.

[0309] 2. Data Preprocessing

[0310] Server: Perform data cleansing to reduce unnecessary noise. Detect inaccurate or abnormal data and delete or correct it.

[0311] Server: Perform processing to complement missing data. For example, use methods of complementing with the average value or median value, or methods of complementing using a machine learning model.

[0312] Server: Update in the database to store the preprocessed data.

[0313] 3. Feature Extraction

[0314] Server: Extracts useful features from pre-processed data. Specifically, it extracts demographic information (age, gender, income level, etc.), consumer behavior (purchase frequency, type of purchased goods, average purchase amount, etc.), and regional characteristics (infrastructure status, geographical characteristics, number of competing companies, etc.).

[0315] Server: Organizes the extracted features and saves them as training data.

[0316] 4. Model training

[0317] Server: Trains a generative artificial intelligence model using the extracted features. The training process includes splitting the data (training set, validation set, test set).

[0318] Server: Evaluates the model's performance and adjusts hyperparameters as needed.

[0319] Server: Stores the trained model.

[0320] 5. Analysis and Prediction

[0321] Server: Performs analysis on new data using the trained model.

[0322] Server: Predicts market size and untapped market potential in each region.

[0323] Server: Organizes the prediction results and saves them to the database.

[0324] 6. Emotion recognition

[0325] Device: Recognizes the user's emotions using an emotion engine. For example, it analyzes emotions from facial expressions and voice using a camera and microphone.

[0326] Server: Stores recognized emotion data in a database and performs necessary processing to incorporate it into the analysis.

[0327] 7. Visualization of Results

[0328] Server: Combines prediction results and user sentiment data to generate data that can be visualized in an easy-to-understand format (e.g., graphs, maps, dashboards).

[0329] Terminal: Displays visualized data to the user. The user understands the specific analysis results by viewing graphs and maps.

[0330] 8. Emotion-based display adjustments

[0331] Server: Based on the emotion data recognized by the emotion engine, the server adjusts how the analysis results are displayed. For example, if the user is feeling stressed, the display is simplified.

[0332] 9. Proposal for staffing arrangements

[0333] Server: Based on the analysis results, propose effective staffing arrangements. These proposals will take into account regional market size, purchasing power, and user sentiment data.

[0334] Terminal: Displays the proposed staffing plan to the user. The user then develops the optimal staffing plan based on this.

[0335] Specific example

[0336] For example, if a user specifies Tokyo and Osaka as target regions, the device collects purchasing data, demographic information, and economic data for these regions. The server stores this data in a database, performs data cleansing and feature extraction, and trains a generative artificial intelligence model. The server then analyzes the new data to predict market size and untapped markets in Tokyo and Osaka. Along with the analysis results, the device recognizes the user's emotions using an emotion engine, and the server integrates this data and visualizes it appropriately. If the user is not feeling stressed, a detailed graphical interface is displayed; if they are feeling stressed, it switches to a simpler display. Finally, the server proposes a staffing plan and displays it to the user through the device. The user can then plan to effectively allocate 12 sales staff in Tokyo and 8 in Osaka.

[0337] Thus, the present invention provides a consistent series of processes, from the collection and analysis of regional data to sentiment recognition, visualization of results, and proposal of personnel allocation, and by taking user sentiment into consideration, it brings even greater convenience and accuracy.

[0338] The following describes the processing flow.

[0339] Step 1: Data Collection

[0340] User: Specify the region to be analyzed (e.g., Tokyo, Osaka) and the required data types (purchase data, demographic information, economic data).

[0341] Terminal: Collects necessary data from specified data sources (government statistics databases, commercial databases, web APIs, etc.) and sends it to the server.

[0342] Server: Receives data and stores it in a central database. This ensures that all data is managed centrally.

[0343] Step 2: Data Preprocessing

[0344] Server: Perform data cleansing to detect, delete, or correct inaccurate or abnormal data. For example, delete outliers in purchase data.

[0345] Server: Imputate missing data. Imputate missing data with the mean or median, or use a machine learning model to calculate predicted values.

[0346] Server: Saves the pre-processed data back to the database.

[0347] Step 3: Feature Extraction

[0348] Server: Extracts features from pre-processed data. Specifically, it extracts demographic information (age, gender, income level, etc.), consumer behavior (purchase frequency, type of purchased goods, average purchase amount, etc.), and regional characteristics (infrastructure status, geographical characteristics, number of competing companies, etc.).

[0349] Server: Organizes the extracted features and saves them as training data.

[0350] Step 4: Model Training

[0351] Server: Trains a generative artificial intelligence model using the extracted features. This process involves splitting the data into a training set, a validation set, and a test set.

[0352] Server: Evaluates the model's performance and adjusts hyperparameters as needed.

[0353] Server: Stores the trained model.

[0354] Step 5: Analysis and Prediction

[0355] Server: Uses a pre-trained model to analyze new data. The analysis predicts market size and untapped market potential in each region.

[0356] Server: Organizes the prediction results and saves them to the database.

[0357] Step 6: Emotion Recognition

[0358] Device: Recognizes the user's emotions using an emotion engine. Analyzes emotions from facial expressions and voice using the camera and microphone.

[0359] Server: Stores recognized emotion data in a database and performs necessary processing to incorporate it into the analysis.

[0360] Step 7: Visualizing the Results

[0361] Server: Combines prediction results and user sentiment data to generate data that can be visualized in an easy-to-understand format (e.g., graphs, maps, dashboards).

[0362] Terminal: Displays visualized data to the user. The user understands the specific analysis results by viewing graphs and maps.

[0363] Step 8: Emotion-based display adjustments

[0364] Server: Based on the emotion data recognized by the emotion engine, adjusts how the analysis results are displayed. For example, if the user is feeling stressed, simplify the display.

[0365] Step 9: Propose staffing arrangements

[0366] Server: Based on the analysis results, propose effective staffing arrangements. These proposals will take into account regional market size, purchasing power, and user sentiment data.

[0367] Terminal: Displays the proposed staffing plan to the user. The user then develops the optimal staffing plan based on this.

[0368] For example, if a user designates Tokyo and Osaka as target regions, the device collects purchasing data, demographic information, and economic data for these areas. The server stores this data in a database, performs data cleansing and feature extraction, and trains a generative artificial intelligence model. The server then analyzes the new data to predict market size and untapped markets in Tokyo and Osaka. Along with the analysis results, the device recognizes the user's emotions using an emotion engine, and the server integrates this data and visualizes it appropriately. If the user is not feeling stressed, a detailed graphical interface is displayed; if they are stressed, it switches to a simpler display. Finally, the server proposes a staffing plan and displays it to the user through the device. The user can then plan to effectively allocate 12 sales staff to Tokyo and 8 to Osaka.

[0369] (Example 2)

[0370] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0371] In modern society, effectively collecting and analyzing regional data and conducting market analysis is crucial for enhancing a company's competitiveness. However, current systems fragment the entire process from data collection to analysis and result display, making it difficult to perform efficiently. Furthermore, there is a lack of systems that provide advanced features such as displaying analysis results while considering user emotions and suggesting personnel allocation. This can lead to user stress and a risk of decreased business efficiency due to inability to implement optimal personnel allocation.

[0372] In Example 2, the identification processing by the identification processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for collecting regional data, means for storing the collected data in a database, means for performing data cleansing on the stored data, means for imputing missing data, means for extracting features from the data, means for training a generative artificial intelligence model using the extracted features, means for analyzing new data using the trained model and predicting market size, means for recognizing user emotions, means for reflecting the recognized emotion data in the analysis, means for visualizing and displaying the prediction results, means for adjusting the display method based on the emotion data, and means for proposing personnel allocation based on the analysis results. This enables the consistent execution of a series of processes, flexible display of analysis results in accordance with user emotions, and effective proposal of personnel allocation.

[0373] "Regional data" refers to various types of information related to a specific region, and specifically includes purchasing data, population information, and economic data.

[0374] A "database" is a system for centrally managing and storing collected data; examples include MySQL and PostgreSQL.

[0375] "Data cleansing" is the process of detecting, correcting, and removing inaccurate data and noise in order to improve the quality of the data.

[0376] "Missing data" refers to values ​​that are missing from a dataset, and a means of filling in these missing values ​​is required.

[0377] "Features" are elements of data extracted for use in analysis and model training, and specific examples include demographic information and consumer behavior.

[0378] A "generative artificial intelligence model" is a machine learning model that is trained on collected and pre-processed data to perform predictions and analyses.

[0379] "Emotion recognition" is a technology that detects a user's emotional state, using an emotion engine to analyze emotions from facial expressions and voice.

[0380] "Visualization" refers to displaying analysis results in a format that is easy for users to understand, and graphs and maps are commonly used for this purpose.

[0381] "Staffing" refers to a plan to effectively allocate the optimal number of staff based on analysis results.

[0382] The "display method" refers to the interface used to present analysis results to the user, and it is adjusted according to the user's emotions.

[0383] This invention is a system that collects and analyzes regional data to perform market analysis and propose personnel allocation, and in particular incorporates an emotion engine that recognizes user emotions. The following hardware and software are used in the implementation of this system.

[0384] The server is primarily responsible for data storage, preprocessing, feature extraction, model training, analysis, and visualization of results. The following software should be installed on the server:

[0385] Database management systems (e.g., MySQL, PostgreSQL)

[0386] Data analysis libraries (e.g., Pandas, NumPy)

[0387] Machine learning libraries (e.g., scikit-learn, TensorFlow, PyTorch)

[0388] Visualization libraries (e.g., Matplotlib, Plotly)

[0389] The device is responsible for data collection, emotion recognition, and partial data processing. Specifically, it is connected to a camera and microphone to recognize the user's emotions using an emotion engine. Affectiva's SDK is a suitable emotion engine to use.

[0390] The following is an example of the specific processing performed by this system.

[0391] Specific example

[0392] For example, suppose a user specifies Tokyo and Osaka as target regions. In this case,

[0393] 1. The device collects purchasing data, demographic information, and economic data related to these regions from APIs and websites. The device then sends the collected data to the server.

[0394] 2. The server saves the received data to the database. MySQL is used for database management.

[0395] 3. The server performs data cleansing, removing NaN values ​​and duplicate data. The Pandas library is used for this process.

[0396] 4. The server imputes missing data. For example, it can use scikit-learn's Inputter to impute missing values ​​with the mean.

[0397] 5. The server extracts useful features from the pre-processed data. These features include demographic information (age, gender, income level) and consumer behavior (purchase frequency, type of purchased goods, average purchase amount).

[0398] 6. Save the features as training data and train a generative artificial intelligence model. TensorFlow will be used to train the model.

[0399] 7. The server analyzes new data and predicts market size and untapped market potential in Tokyo and Osaka.

[0400] 8. The device uses an emotion engine to recognize the user's emotions. For example, it uses a camera to capture facial expressions and analyzes the emotional state.

[0401] 9. The server saves the prediction results and recognized emotion data to a database and incorporates them into the analysis results.

[0402] 10. The server visualizes the prediction results in the form of graphs or maps. For example, it can generate interactive graphs using the Plotly library.

[0403] 11. The device displays visualized data to the user. If the user is experiencing stress, measures are taken to simplify the display.

[0404] 12. Based on the analysis results, the server proposes an effective staffing plan. For example, it might plan to allocate 12 sales staff to Tokyo and 8 to Osaka.

[0405] 13. The terminal displays the proposed staffing plan to the user. The user can then use this to determine the optimal staffing arrangement.

[0406] This system uses a generative artificial intelligence model to perform market analysis while simultaneously incorporating user sentiment data to propose more appropriate information presentations and personnel allocation plans to users.

[0407] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0408] Step 1:

[0409] On the system's administration screen, users specify the target region for analysis (e.g., Tokyo, Osaka) and the required data types (purchase data, demographic information, economic data). Based on this, a request for data collection is created.

[0410] Input: Target region and data type

[0411] Output: Data collection request

[0412] Step 2:

[0413] The device collects the necessary data from a specified data source (e.g., an API, website, database, etc.). Specifically, it uses a Python script to issue an HTTP request to the API and sends the retrieved data to the server in JSON format.

[0414] Input: Data collection request

[0415] Output: Collected data (JSON format)

[0416] Step 3:

[0417] The server stores the received data in a central database. A DBMS such as MySQL is used for this purpose. Specifically, data is registered in the database by issuing SQL INSERT statements.

[0418] Input: Collected data (JSON format)

[0419] Output: Data stored in the database

[0420] Step 4:

[0421] The server performs data cleansing. For example, it processes the dataframe using the Pandas library to remove NaN values ​​and duplicate data. It detects and corrects inaccurate data and noise to generate a clean dataset.

[0422] Input: Data retrieved from the database

[0423] Output: Cleansed data

[0424] Step 5:

[0425] The server will impute the missing data. Specifically, it will use scikit-learn's Inputter to impute missing values ​​with the mean or median, or it will use a machine learning model to perform the imputation.

[0426] Input: Cleansed data

[0427] Output: Interpolated data

[0428] Step 6:

[0429] The server extracts useful features from pre-processed data. Specifically, it extracts features based on demographic information, consumer behavior, and regional characteristics using Pandas or NumPy.

[0430] Input: Completed data

[0431] Output: Feature dataset

[0432] Step 7:

[0433] The server uses the extracted features to train a generative artificial intelligence model. Specifically, it splits the data into a training set, a validation set, and a test set, and trains the model using TensorFlow.

[0434] Input: Feature dataset

[0435] Output: Trained model

[0436] Step 8:

[0437] The server uses a pre-trained model to perform analysis on new data. Specifically, it inputs new data into the model and predicts market size.

[0438] Input: New data, trained model

[0439] Output: Forecast results (market size, potential)

[0440] Step 9:

[0441] The device uses an emotion engine to recognize the user's emotions. For example, it uses a camera to capture facial expressions and performs emotion analysis.

[0442] Input: User's facial expression, emotion engine

[0443] Output: Recognized emotion data

[0444] Step 10:

[0445] The server stores the recognized emotion data in a database and performs the necessary processing to incorporate it into the analysis. Specifically, the emotion data is incorporated into the analysis algorithm.

[0446] Input: Prediction results, sentiment data

[0447] Output: Emotion-reflected analysis results

[0448] Step 11:

[0449] The server visualizes the sentiment-infused analysis results in formats such as graphs, maps, and dashboards. For example, it can generate interactive graphs using the Plotly library.

[0450] Input: Sentiment-reflected analysis results

[0451] Output: Visualization data (graphs, maps)

[0452] Step 12:

[0453] The device displays visualized data to the user. If the user is experiencing stress, measures such as simplifying the display will be taken.

[0454] Input: Visualization data, user emotional state

[0455] Output: Screen displayed to the user

[0456] Step 13:

[0457] The server analyzes the data and proposes an effective staffing plan. For example, it might plan to deploy 12 sales staff in Tokyo and 8 in Osaka.

[0458] Input: Visualization data, analysis results

[0459] Output: Staffing proposal

[0460] Step 14:

[0461] The terminal displays the proposed staffing plan to the user. The user can then use this to determine the optimal staffing arrangement.

[0462] Input: Personnel allocation proposal

[0463] Output: Final screen displayed to the user

[0464] (Application Example 2)

[0465] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0466] Conventional market analysis and staffing suggestion systems analyze data based solely on regional and purchasing data, failing to consider user emotions. This can lead to stress and confusion for users when receiving analysis results. Furthermore, there was no system that could acquire real-time emotional data and immediately suggest staffing and product placement based on it. To address these challenges, a system is needed that recognizes user emotions and adjusts the display of analysis results accordingly.

[0467] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for collecting regional data, means for storing the collected data in a database, means for performing data cleansing on the stored data, means for extracting features from the data, means for training a generative artificial intelligence model using the extracted features, means for analyzing new data using the trained model and predicting market size, means for visualizing and displaying the prediction results, means for proposing personnel allocation based on the analysis results, means for recognizing the user's emotions in real time, means for visualizing the analysis results including the recognized emotion data and adjusting the display method, and means for presenting the results to the user in real time using a device such as smart glasses. This makes it possible to present analysis results that take into account the user's emotions and to propose appropriate personnel allocation and product placement based on them.

[0468] "Regional data" is a general term for information including population data, economic data, and purchasing data for a specific geographical area.

[0469] A "database" is a digital system that centrally manages collected and stored data, enabling efficient searching and utilization.

[0470] "Data cleansing" is a processing method used to reduce inaccuracies and noise in data and prepare it for analysis.

[0471] "Features" are variables that represent important patterns or properties in data analysis and machine learning model training.

[0472] A "generative artificial intelligence model" is a type of machine learning model that generates useful information from data and makes predictions and decisions by learning patterns.

[0473] "Market size" refers to an indicator that shows the size of a market, such as the number of consumers or purchasing power in a particular region or market.

[0474] "Emotional data" refers to emotional information extracted from human facial expressions and voices collected using cameras and microphones.

[0475] "Smart glasses" are a type of wearable device, specifically glasses-shaped equipment that incorporates displays and sensors to show information in real time.

[0476] A "user" is a person or organization that uses this system and is the recipient of the analysis results and suggestions.

[0477] "Personnel allocation" refers to the plan or method of appropriately allocating staff or workers for the purpose of efficient business operations and service provision.

[0478] This invention relates to a system for collecting, analyzing, recognizing emotions in regional data, visualizing the results, and proposing personnel allocation. A key feature of this system is that it uses devices such as smart glasses to recognize the user's emotions in real time and reflects this in the analysis results and personnel allocation proposals.

[0479] 1. Data Collection

[0480] The server collects regional data from specified data sources and stores it in a database. This data includes diverse information such as purchasing data, population information, and economic data. This data collection allows the system to understand the market characteristics of each region.

[0481] 2. Data preprocessing

[0482] The server performs data cleansing on the collected data to reduce unwanted noise. Furthermore, it detects and removes or corrects inaccurate or anomalous data. Missing data is imputed using the mean or median, or through machine learning models.

[0483] 3. Feature Extraction

[0484] The server extracts useful features from the pre-processed data. These include demographic information (age, gender, income level, etc.), consumer behavior (purchase frequency, type of purchased goods, average purchase amount, etc.), and regional characteristics (infrastructure status, geographical features, number of competing companies, etc.). This extracts important data points that are then used for analysis.

[0485] 4. Model training

[0486] The server trains a generative AI model using the extracted features. This training process involves splitting the data (training set, validation set, test set). Once the training is complete, the model's performance is evaluated, and hyperparameters are adjusted as needed.

[0487] 5. Analysis and Prediction

[0488] The server uses a pre-trained model to analyze new data and predict market size and untapped market potential in each region. The prediction results are stored in a database.

[0489] 6. Emotion recognition

[0490] The device (smart glasses) uses a camera and microphone to capture the user's facial image and voice data, which are then analyzed by an emotion recognition engine. For example, it analyzes a customer's face in real time and obtains their emotion data. This emotion data is sent to a server and further incorporated into the analysis results.

[0491] 7. Visualization of Results

[0492] The server combines prediction results with user sentiment data and generates data to be visualized in an easy-to-understand format (graphs, maps, dashboards, etc.). The device (smart glasses) displays this visualized data to the user. The user can understand the analysis results while viewing graphs and maps.

[0493] 8. Emotion-based display adjustments

[0494] The server adjusts how the analysis results are displayed based on the emotional data recognized by the emotion recognition engine. For example, if the user is feeling stressed, the display may be simplified.

[0495] 9. Proposal for staffing arrangements

[0496] The server proposes an effective staffing plan based on the analysis results. This proposal takes into account regional market size, purchasing power, and even user sentiment data. The terminal (smart glasses) displays the proposed staffing plan to the user. The user can then develop an optimal staffing plan based on this.

[0497] Hardware and software to be used

[0498] Smart glasses: Recognizing customer emotions in real time (e.g., Google Glass®, Vuzix Blade)

[0499] Server: Cloud infrastructure for data collection and analysis (e.g., AWS®, Google Cloud)

[0500] Emotion recognition engine: (Examples: Microsoft® Azure® Face API, Google Cloud Vision API)

[0501] Database: (e.g., MySQL, MongoDB)

[0502] Machine learning libraries: (e.g., scikit-learn for Python, TensorFlow)

[0503] Example of a prompt:

[0504] "Using customer sentiment data and sales data, propose the optimal product placement strategy to increase store sales."

[0505] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0506] Step 1:

[0507] The server collects regional data from specified data sources and stores it in a database.

[0508] Input: Purchase data, demographic information, economic data

[0509] Data processing: Data collection and organization

[0510] Output: Data organized into a unified format is stored in the database.

[0511] Step 2:

[0512] The server performs data cleansing on the collected data to reduce unwanted noise.

[0513] Input: Original data stored in the database

[0514] Data processing: Noise reduction, detection and correction of inaccurate data.

[0515] Output: Cleansed data

[0516] Step 3:

[0517] The server will fill in any missing values ​​in the data.

[0518] Input: Cleansed data

[0519] Data processing: Imputation of missing values ​​(using mean, median, or machine learning models)

[0520] Output: Completed and cleansed data

[0521] Step 4:

[0522] The server extracts useful features from the pre-processed data.

[0523] Input: Completed and cleansed data

[0524] Data processing: Feature extraction (demographic information, consumer behavior, regional characteristics, etc.)

[0525] Output: Feature data

[0526] Step 5:

[0527] The server uses the extracted features to train a generative AI model.

[0528] Input: Feature data

[0529] Data processing: Splitting into training, validation, and test sets; model training and evaluation.

[0530] Output: Trained AI model

[0531] Step 6:

[0532] The server uses a pre-trained model to analyze new data and predict market size.

[0533] Input: Trained AI model, new data

[0534] Data processing: Analysis and prediction

[0535] Output: Forecast data (market size, potential of untapped markets)

[0536] Step 7:

[0537] The device (smart glasses) uses a camera and microphone to capture the user's facial image and voice data, which are then analyzed by an emotion recognition engine.

[0538] Input: Customer's facial image, voice data

[0539] Data processing: Analysis using an emotion recognition engine

[0540] Output: User sentiment data

[0541] Step 8:

[0542] The server combines prediction results with user sentiment data and visualizes them in an easy-to-understand format (graphs, maps, dashboards, etc.).

[0543] Input: Prediction result data, user sentiment data

[0544] Data processing: Generation of visualized data (graphs, maps, dashboards)

[0545] Output: Visualization data

[0546] Step 9:

[0547] The device (smart glasses) displays the visualized analysis results to the user and adjusts the display method as needed.

[0548] Input: Visualization data, user sentiment data

[0549] Data processing: Adjusting display methods based on emotions

[0550] Output: Adjusted display data

[0551] Step 10:

[0552] Based on the analysis results, the server proposes an effective staffing allocation.

[0553] Input: Analysis results data, user sentiment data

[0554] Data calculation: Optimization calculation of personnel allocation

[0555] Output: Personnel allocation proposal data

[0556] Step 11:

[0557] The device (smart glasses) displays the proposed staffing plan to the user.

[0558] Input: Personnel allocation proposal data

[0559] Data processing: Real-time display

[0560] Output: Staffing proposals to be reviewed by the user

[0561] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0562] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0563] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0564] [Second Embodiment]

[0565] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0566] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0567] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0568] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0569] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0570] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0571] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0572] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0573] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0574] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0575] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0576] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0577] This invention relates to a system for collecting and analyzing regional data to provide effective market analysis and personnel allocation suggestions. The processing of this system's program is described below in natural language.

[0578] 1. Data Collection

[0579] User: Specify the region to be analyzed and the required data types (e.g., purchasing data, demographic information, economic data).

[0580] Terminal: Collects necessary data from the specified data source and sends it to the server.

[0581] Server: Stores collected data in a central database. This ensures that all data is managed centrally.

[0582] 2. Data preprocessing

[0583] Server: Perform data cleansing to reduce unwanted noise. Detect and remove inaccurate or abnormal data.

[0584] Server: Imputes missing data. For example, it may use imputation methods such as imputing with the mean or median, or using machine learning models.

[0585] 3. Feature Extraction

[0586] Server: Extracts features from pre-processed data. Specifically, it extracts demographic information (age, gender, income level, etc.), consumer behavior (purchase frequency, type of purchased goods, average purchase amount, etc.), and regional characteristics (infrastructure status, geographical characteristics, number of competing companies, etc.).

[0587] Server: Stores the extracted features as training data.

[0588] 4. Model training

[0589] Server: Trains a generative artificial intelligence model using the extracted features. The training process includes splitting the data (training set, validation set, test set).

[0590] Server: Evaluates the trained model and adjusts hyperparameters as needed.

[0591] 5. Analysis and Prediction

[0592] Server: Uses a pre-trained model to analyze new data. This analysis predicts the market size and potential of untapped markets in each region.

[0593] Server: Organizes the analysis results and saves them to the database.

[0594] 6. Visualization of Results

[0595] Server: Generates data to visualize analysis results in an easy-to-understand format (e.g., graphs, maps, dashboards).

[0596] Terminal: Displays visualized data to the user. The user understands the specific analysis results by viewing graphs and maps.

[0597] 7. Proposal for staffing arrangements

[0598] Server: Based on the analysis results, propose effective staffing arrangements. These proposals will take into account the market size and purchasing power of each region.

[0599] Terminal: Displays the proposed staffing plan to the user. The user can then create an optimal staffing plan based on this proposal.

[0600] Specific example

[0601] For example, if a user selects Tokyo and Osaka as target regions, the terminal collects purchasing data, population information, and economic data for Tokyo and Osaka. The server stores this data in a database and performs data cleansing and feature extraction. A generative artificial intelligence model is trained using the extracted features, and new data is analyzed by the model. As a result, the market size and potential of untapped markets in Tokyo and Osaka are predicted. The server visualizes the prediction results and displays them to the user through the terminal. Based on these results, the user can plan to deploy 12 sales staff in Tokyo and 8 sales staff in Osaka. In this way, efficient market analysis and personnel allocation are achieved.

[0602] Thus, the present invention provides a consistent process from the collection and analysis of regional data to the visualization of results and the proposal of personnel allocation, resulting in significant benefits and convenience.

[0603] The following describes the processing flow.

[0604] Step 1: Data Collection

[0605] User: Specify the region to be analyzed and the required data types (e.g., purchasing data, demographic information, economic data).

[0606] Terminal: Collects data from the specified data source (e.g., government statistics database, commercial database, web API, etc.).

[0607] Terminal: Sends collected data to the server.

[0608] Server: Stores received data in a database for later storage.

[0609] Step 2: Data Preprocessing

[0610] Server: Perform data cleansing. Detect and delete or correct inaccurate or abnormal data.

[0611] Server: Performs processing to impute missing data. For example, it can impute missing data with the mean or median, or use a machine learning model to calculate predicted values.

[0612] Server: Updates the database to store the data after preprocessing is complete.

[0613] Step 3: Feature Extraction

[0614] Server: Extracts useful features from pre-processed data. Specifically, it extracts demographic information (age, gender, income level, etc.), consumer behavior (purchase frequency, type of purchased goods, average purchase amount, etc.), and regional characteristics (infrastructure status, geographical characteristics, number of competing companies, etc.).

[0615] Server: Organizes the extracted features and saves them as training data.

[0616] Step 4: Model Training

[0617] Server: Trains a generative artificial intelligence model using the extracted features. The training process includes splitting the data (training set, validation set, test set).

[0618] Server: Evaluates the model's performance and adjusts hyperparameters as needed.

[0619] Server: Stores the trained model.

[0620] Step 5: Analysis and Prediction

[0621] Server: Performs analysis on new data using the trained model.

[0622] Server: Predicts market size and untapped market potential in each region.

[0623] Server: Organizes the prediction results and saves them to the database.

[0624] Step 6: Visualizing the Results

[0625] Server: Generates data to visualize prediction results in an easy-to-understand format (e.g., graphs, maps, dashboards).

[0626] Terminal: Displays visualized data to the user. The user understands the specific analysis results by viewing graphs and maps.

[0627] Step 7: Propose staffing arrangements

[0628] Server: Based on the analysis results, proposes effective staffing arrangements. These proposals take into account regional market size and purchasing power.

[0629] Terminal: Displays the proposed staffing plan to the user. The user then develops the optimal staffing plan based on this.

[0630] For example, if a user designates Tokyo and Osaka as target regions, the terminal collects purchasing data, demographic information, and economic data for these regions. The server performs data cleansing and feature extraction, and trains a generative artificial intelligence model. The server then analyzes the new data to predict the market size and untapped markets for Tokyo and Osaka. Finally, the server visualizes the prediction results and displays them to the user through the terminal. Based on the proposed staffing plan, the user effectively allocates 12 sales staff to Tokyo and 8 to Osaka.

[0631] (Example 1)

[0632] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0633] Traditional marketing analysis methods made it difficult to collect and analyze data that differed by region and to allocate personnel effectively. Furthermore, there was no system that consistently handled data cleansing, data reconciliation, and visualization of analysis results. As a result, companies were unable to conduct efficient market analysis and optimal personnel allocation, hindering business efficiency and accurate understanding of target markets.

[0634] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0635] In this invention, the server includes means for specifying the region and data type to be analyzed; means for collecting the specified data; means for storing the collected data in a database; means for performing data cleansing on the stored data to remove inaccurate data and outliers; means for imputing missing data; means for extracting features from the data; means for training a generative artificial intelligence model using the extracted features; means for analyzing new data and predicting market size using the trained artificial intelligence model; means for storing and visualizing the prediction results in a database; means for proposing staffing arrangements based on the analysis results; and means for displaying the proposed staffing arrangements. This makes it possible to consistently perform detailed market analysis and effective staffing arrangements for each region.

[0636] "Analysis target area" refers to the specific region that the system targets when performing marketing analysis.

[0637] "Data type" refers to the various types of data necessary for analysis in the collection of regional data. Examples include purchasing data, population information, and economic data.

[0638] "Data cleansing" refers to the process of detecting inaccurate information and outliers contained in collected data, and then removing or correcting them.

[0639] "Missing data" refers to information that is missing from a dataset.

[0640] "Features" refer to data attributes or variables that have significant meaning for data analysis.

[0641] A "generative artificial intelligence model" refers to a model that uses machine learning and deep learning techniques to make predictions and classifications on new data.

[0642] "Market size" refers to the expected sales or purchasing power in a particular region or market.

[0643] "Visualization" refers to the process of visually displaying data and analysis results using graphs, dashboards, maps, and other visual tools.

[0644] "Personnel allocation" refers to the process by which a company or organization distributes the necessary number of staff members to appropriate locations and roles in order to effectively carry out its operations.

[0645] This invention relates to a system for collecting and analyzing regional data to provide effective market analysis and personnel allocation suggestions.

[0646] System program configuration

[0647] This system begins with the user specifying the analysis area and data type. The terminal collects data and sends it to the server, which processes the collected data to propose market analysis and staffing strategies. This entire process involves multiple hardware and software components.

[0648] Hardware:

[0649] Device: A computer or smartphone that can access the internet or APIs for data collection.

[0650] Server: A high-performance computer used for data processing, storage, and analysis.

[0651] software:

[0652] Database management systems: Used to centrally manage collected data. Examples include MySQL and PostgreSQL.

[0653] Data cleansing tools: These cleanse the collected data. For example, the pandas library in Python.

[0654] Generative AI models: Machine learning models that perform market analysis using features. Examples include Scikit-learn and TensorFlow.

[0655] Visualization tools: Tools for visualizing and displaying analysis results. Examples include Tableau and D3.js.

[0656] Program processing flow

[0657] The system program performs the following steps:

[0658] 1. Data collection:

[0659] User: Specify the region to be analyzed and the required data types (purchase data, demographic information, economic data, etc.).

[0660] Terminal: Collects necessary data from specified data sources (open data portals or commercial databases) and sends it to the server.

[0661] Server: Stores data sent from terminals in a central database and manages it centrally.

[0662] 2. Data preprocessing:

[0663] Server: Performs data cleansing on the collected data, removing inaccurate data and outliers.

[0664] Server: Imputes missing data using mean values ​​and machine learning models.

[0665] 3. Feature extraction:

[0666] Server: Extracts features such as demographic information, consumer behavior, and regional characteristics from pre-processed data and saves them as training datasets.

[0667] 4. Model training:

[0668] Server: Trains a generative artificial intelligence model using the extracted feature data. If necessary, splits the data into training, validation, and test sets, and tunes hyperparameters.

[0669] 5. Analysis and Prediction:

[0670] Server: Uses a pre-trained model to analyze new data and predict market size and the potential of untapped markets.

[0671] Server: Organizes the prediction results and saves them to the database.

[0672] 6. Visualization of results:

[0673] Server: Generates visualization data such as graphs and maps based on the analysis results.

[0674] Terminal: Displays dashboards and graphs to the user based on visualization data sent from the server.

[0675] 7. Proposal for staffing arrangements:

[0676] Server: Based on the analysis results, propose effective staffing arrangements.

[0677] Terminal: Displays the proposed staffing plan to the user.

[0678] Explanation of specific examples

[0679] For example, if a user selects Tokyo and Osaka as target regions, the device collects purchasing data, demographic information, and economic data for Tokyo and Osaka through open data portals and commercial databases. The collected data is sent to a server and stored in a central database. Next, the server performs data cleansing, removing inaccurate data and filling in missing data. After that, the server extracts features such as purchase frequency and average purchase amount and stores them as a training dataset. Using the stored data, a generative AI model is trained to make predictions on new data. The analysis results are displayed visually, and the server proposes a staffing plan suitable for the user. Based on this information, the user can plan, for example, to deploy 12 sales staff in Tokyo and 8 in Osaka.

[0680] Example of a prompt

[0681] "Conduct a market analysis using purchasing data, demographic information, and economic data for Tokyo and Osaka, and propose effective staffing strategies."

[0682] Thus, the present invention provides a consistent process from the collection and analysis of regional data to the visualization of results and the proposal of personnel allocation, resulting in significant benefits and convenience.

[0683] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0684] Step 1:

[0685] User: Specify the region and data type to be analyzed. Specifically, the user uses a terminal to select "Tokyo purchasing data and population information" through the interface.

[0686] Input: Region to be analyzed and data type.

[0687] Output: Field information required for the data collection process.

[0688] Step 2:

[0689] Terminal: Collects data via the internet or APIs based on the user-specified analysis area and data type. For example, it accesses "open data portals" or "commercial databases" to obtain information.

[0690] Input: Region to be analyzed and data type.

[0691] Output: Collected raw data.

[0692] Specific operation: A request is sent to the "Open Data Portal API" to retrieve purchasing data for Tokyo.

[0693] Step 3:

[0694] Terminal: Sends collected data to the server.

[0695] Input: Collected raw data.

[0696] Output: Request containing collected data.

[0697] Specific operation: The collected data is sent to the server as an HTTP request via the API.

[0698] Step 4:

[0699] Server: Stores collected data in a central database. This ensures centralized data management.

[0700] Input: Collected data sent to the server.

[0701] Output: Data stored in the central database.

[0702] Specific operation: Use a database management system (e.g., MySQL) to insert data into a table.

[0703] Step 5:

[0704] Server: Perform data cleansing to remove inaccurate data and outliers. For example, remove outliers like "-1 yen" from purchase history data.

[0705] Input: Raw data stored in the central database.

[0706] Output: Cleansed data.

[0707] Specific operation: Use the Python pandas library to filter and remove inaccurate entries in the data.

[0708] Step 6:

[0709] Server: Imputes missing data using mean values ​​or machine learning models. For example, it can imputate missing age data using the average age of other data points.

[0710] Input: Cleansed data (including missing values).

[0711] Output: Interpolated data.

[0712] Specific operation: Missing values ​​are imputed using the K-nearest neighbors algorithm with the scikit-learn library in Python.

[0713] Step 7:

[0714] Server: Extracts features from pre-processed data. Specifically, it extracts demographic information such as age and income, and consumer behavior data such as purchase frequency and purchase amount.

[0715] Input: Completed data.

[0716] Output: Feature dataset.

[0717] Specific operation: Using the Python Feature-engine library, select important features and convert the data into a matrix format.

[0718] Step 8:

[0719] Server: The extracted features are used to split the data into training, validation, and test sets, and a generative artificial intelligence model is trained. Specifically, 70% is split into a training set, 15% into a validation set, and 15% into a test set, and a regression model is trained.

[0720] Input: Feature dataset.

[0721] Output: Trained model.

[0722] Specific operation: The data is split using the scikit-learn library, and a Linear Regression model is trained.

[0723] Step 9:

[0724] Server: Uses a pre-trained model to analyze new data and predict market size by region.

[0725] Input: A trained model and new data.

[0726] Output: Prediction result.

[0727] Specific actions: Input new data into the model and perform regression analysis to predict market size.

[0728] Step 10:

[0729] Server: Organizes and stores prediction results in a database.

[0730] Input: Prediction result.

[0731] Output: Prediction results stored in the database.

[0732] Specific operation: Use an SQL query to save the prediction results to the appropriate table in the database.

[0733] Step 11:

[0734] Server: Converts analysis results into visualization data such as graphs and maps.

[0735] Input: Prediction results stored in the database.

[0736] Output: Visualized data.

[0737] Specific operation: Use Matplotlib and D3.js to convert the prediction results into a visually displayable format.

[0738] Step 12:

[0739] Terminal: Displays dashboards and graphs to the user based on visualization data sent from the server.

[0740] Input: Visualization data from the server.

[0741] Output: Graphs and maps displayed in the user interface.

[0742] Specific operation: Use a web application framework (e.g., React) to dynamically display data in the user interface.

[0743] Step 13:

[0744] Server: Based on the analysis results, it proposes effective personnel allocation for each region.

[0745] Input: Analysis results.

[0746] Output: Staffing proposal.

[0747] Specific operation: Based on the prediction results, calculate the optimal staffing allocation for each region and generate a proposal.

[0748] Step 14:

[0749] Terminal: Displays the proposed staffing plan to the user. The user can make decisions based on the proposal.

[0750] Input: Personnel allocation proposal.

[0751] Output: Staffing suggestions displayed in the user interface.

[0752] Specific operation: Use a web application framework to display the contents of the proposal in a user interface.

[0753] (Application Example 1)

[0754] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0755] Efficient market analysis and staffing planning in physical stores are not easy, and responding quickly to changes in market trends is particularly challenging. Furthermore, data inaccuracies that occur during the process of collecting and analyzing regional data, as well as a lack of means to support visual decision-making, are also challenges.

[0756] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0757] In this invention, the server includes means for collecting regional data, means for storing the collected data in a database, means for performing data cleansing on the stored data, means for extracting features from the data, means for training a generative artificial intelligence model using the extracted features, means for analyzing new data using the trained model and predicting market size, means for visualizing and displaying the prediction results, means for proposing staffing arrangements based on the analysis results, means for notifying changes in market trends for a specific region in real time, and means for displaying the prediction results on an electronic map to support visual judgments for store placement and market analysis. This enables efficient market analysis and staffing planning, and allows for quick and accurate responses to changes in market trends.

[0758] "Regional data" refers to data that includes population information, economic data, purchasing data, and related infrastructure information for a specific geographical area.

[0759] A "database" is a system for centrally managing and storing collected data.

[0760] "Data cleansing" is a technique that improves the quality of collected data by removing noise and inaccurate data.

[0761] "Features" refer to statistically extracted information necessary for data analysis, and are primarily used as input for analytical models.

[0762] A "generative artificial intelligence model" refers to a machine learning algorithm that performs predictions and analyses based on collected and processed data.

[0763] "Prediction" refers to estimating future data and trends using a trained artificial intelligence model.

[0764] "Visualization" refers to displaying analysis results in a visually easy-to-understand format, such as graphs, maps, and dashboards.

[0765] "Personnel allocation" refers to planning the placement of personnel in the most suitable locations and positions.

[0766] "Real-time notifications" is a feature that instantly informs users of market trends and other data changes.

[0767] An "electronic map" is a digital map that visually displays geographical information and forecast results.

[0768] This invention relates to a system for collecting and analyzing regional data to perform market analysis and propose personnel allocation strategies. The system functions through the cooperation of a server, terminals, and users.

[0769] 1. Data Collection

[0770] The user specifies the region to be analyzed and the required data types (e.g., purchasing data, demographic information, economic data). The terminal collects the necessary data from the specified data sources and sends it to the server. The server stores the collected data in a central database and manages all data centrally.

[0771] 2. Data preprocessing

[0772] The server performs data cleansing to reduce unnecessary noise. Specifically, it detects and removes inaccurate and anomalous data. It also uses methods such as imputation with the mean or median, or imputation using machine learning models, to fill in missing data.

[0773] 3. Feature Extraction

[0774] The server extracts features from the pre-processed data. Specifically, it extracts demographic information (age, gender, income level, etc.) and consumer behavior (purchase frequency, type of purchased goods, average purchase amount, etc.). It also includes regional characteristics (infrastructure status, geographical features, number of competing companies, etc.). The extracted features are stored as training data.

[0775] 4. Model training

[0776] The server uses the extracted features to train a generative artificial intelligence model. The training process includes splitting the data (training set, validation set, test set). The trained model is evaluated, and hyperparameters are adjusted as needed.

[0777] 5. Analysis and Prediction

[0778] The server uses a pre-trained model to analyze new data. This analysis predicts the market size and potential of untapped markets in each region. The analysis results are stored in a database for reuse.

[0779] 6. Visualization of Results

[0780] The server generates data that visualizes the analysis results in an easy-to-understand format (e.g., graphs, maps, dashboards). The terminal displays this visualized data to the user. This allows the user to understand the specific analysis results while viewing graphs and maps.

[0781] 7. Proposal for staffing arrangements

[0782] Based on the analysis results, the server proposes an effective staffing plan. This proposal takes into account the market size and purchasing power of each region. The terminal displays the proposed staffing plan to the user, helping them to create the optimal staffing plan.

[0783] It also includes a feature that provides real-time notifications of changes in market trends in specific regions. This feature provides alerts to quickly respond to sudden market fluctuations. Furthermore, it can display forecast results on an electronic map, visually supporting store placement and market analysis. This allows users to make more intuitive decisions.

[0784] Specific example

[0785] For example, suppose a user collects purchasing data, population data, and competitor information based on location data in Tokyo to perform market analysis. In this case, the server collects and preprocesses this data, extracts features, and trains an AI model. This model predicts the market size of Tokyo, and the results are visually presented to the user in the form of graphs and maps. Then, the optimal staffing allocation for stores in Tokyo is suggested.

[0786] Examples of prompt statements are as follows:

[0787] "Based on location data for Tokyo, collect in-store purchasing data, population information, and competitor information to conduct market analysis. Then, create an application that proposes appropriate staffing arrangements based on the results."

[0788] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0789] Step 1:

[0790] Data collection

[0791] The user specifies the target region and the required data type (purchase data, population information, economic data, etc.). The terminal collects the necessary data from the specified data source (public API, website, statistical database, etc.) and sends the collected data to the server. The server stores the received data in a central database.

[0792] Input: User-specified region and data type

[0793] Output: Regional data stored on the server

[0794] Step 2:

[0795] Data preprocessing

[0796] The server performs data cleansing on the received data. Specifically, it normalizes unstructured data, reduces noisy data, and detects and removes inaccurate or outlier values. Next, to impute missing data, it uses the mean or median, or applies imputation methods using machine learning models.

[0797] Input: Raw data stored in the central database

[0798] Output: Cleansed and imputed data

[0799] Step 3:

[0800] Feature extraction

[0801] The server extracts features from the pre-processed data. Specifically, it extracts demographic information (age, gender, income level, etc.), consumer behavior (purchase frequency, type of purchased goods, average purchase amount, etc.), and regional characteristics (infrastructure status, geographical characteristics, number of competing stores, etc.), and stores them as training data.

[0802] Input: Pre-processed data

[0803] Output: Training data with extracted features

[0804] Step 4:

[0805] Model training

[0806] The server uses the extracted features to train a generative artificial intelligence model. This process involves splitting the training data into training, validation, and test sets, selecting appropriate hyperparameters, and optimizing the model. The trained model is then used for subsequent analysis and prediction.

[0807] Input: Training data with extracted features

[0808] Output: Trained generative artificial intelligence model

[0809] Step 5:

[0810] Analysis and prediction

[0811] The server uses a pre-trained generative artificial intelligence model to analyze new data. This analysis predicts the market size and potential of untapped markets in each region, and these predictions are stored in a database.

[0812] Input: A trained generative AI model, and new data to be analyzed.

[0813] Output: Prediction results stored in the database

[0814] Step 6:

[0815] Visualization of results

[0816] The server generates data that visualizes the prediction results in formats such as graphs, maps, and dashboards. The terminal displays this visualized data to the user, allowing the user to visually confirm the analysis results.

[0817] Input: Prediction results stored in the database

[0818] Output: Visualized analysis results

[0819] Step 7:

[0820] Proposal for staffing arrangements

[0821] The server proposes an effective staffing plan based on the analysis results. The proposal takes into account the market size and purchasing power of each region. The terminal displays the proposed staffing plan to the user, allowing the user to create the optimal staffing plan.

[0822] Input: Visualized analysis results

[0823] Output: Proposed staffing arrangements

[0824] Step 8:

[0825] Real-time notifications

[0826] The server monitors market trends in a specific region in real time and notifies the user when a change is detected. This allows the user to respond quickly and accurately to market changes.

[0827] Input: Real-time monitored market data

[0828] Output: Notification of market changes to users

[0829] Step 9:

[0830] Use of electronic maps

[0831] The server generates data to display prediction results on an electronic map, and the terminal displays this to the user. This allows the user to visually perform store placement and market analysis.

[0832] Input: Prediction result

[0833] Output: Prediction results displayed on the electronic map

[0834] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0835] This invention combines a system that collects and analyzes regional data to perform effective market analysis and personnel allocation proposals with an emotion engine that recognizes user emotions. The processing of this system's program is described below in natural language.

[0836] 1. Data Collection

[0837] User: Specify the region to be analyzed and the required data types (e.g., purchasing data, demographic information, economic data).

[0838] Terminal: Collects necessary data from the specified data source and sends it to the server.

[0839] Server: Receives data and stores it in a central database. This ensures that all data is managed centrally.

[0840] 2. Data preprocessing

[0841] Server: Perform data cleansing to reduce unnecessary noise. Detect and delete or correct inaccurate or abnormal data.

[0842] Server: Performs processing to impute missing data. For example, it may use imputation methods such as imputation using the mean or median, or imputation using machine learning models.

[0843] Server: Updates the database to store the data after preprocessing is complete.

[0844] 3. Feature Extraction

[0845] Server: Extracts useful features from pre-processed data. Specifically, it extracts demographic information (age, gender, income level, etc.), consumer behavior (purchase frequency, type of purchased goods, average purchase amount, etc.), and regional characteristics (infrastructure status, geographical characteristics, number of competing companies, etc.).

[0846] Server: Organizes the extracted features and saves them as training data.

[0847] 4. Model training

[0848] Server: Trains a generative artificial intelligence model using the extracted features. The training process includes splitting the data (training set, validation set, test set).

[0849] Server: Evaluates the model's performance and adjusts hyperparameters as needed.

[0850] Server: Stores the trained model.

[0851] 5. Analysis and Prediction

[0852] Server: Performs analysis on new data using the trained model.

[0853] Server: Predicts market size and untapped market potential in each region.

[0854] Server: Organizes the prediction results and saves them to the database.

[0855] 6. Emotion recognition

[0856] Device: Recognizes the user's emotions using an emotion engine. For example, it analyzes emotions from facial expressions and voice using a camera and microphone.

[0857] Server: Stores recognized emotion data in a database and performs necessary processing to incorporate it into the analysis.

[0858] 7. Visualization of Results

[0859] Server: Combines prediction results and user sentiment data to generate data that can be visualized in an easy-to-understand format (e.g., graphs, maps, dashboards).

[0860] Terminal: Displays visualized data to the user. The user understands the specific analysis results by viewing graphs and maps.

[0861] 8. Emotion-based display adjustments

[0862] Server: Based on the emotion data recognized by the emotion engine, the server adjusts how the analysis results are displayed. For example, if the user is feeling stressed, the display is simplified.

[0863] 9. Proposal for staffing arrangements

[0864] Server: Based on the analysis results, propose effective staffing arrangements. These proposals will take into account regional market size, purchasing power, and user sentiment data.

[0865] Terminal: Displays the proposed staffing plan to the user. The user then develops the optimal staffing plan based on this.

[0866] Specific example

[0867] For example, if a user specifies Tokyo and Osaka as target regions, the device collects purchasing data, demographic information, and economic data for these regions. The server stores this data in a database, performs data cleansing and feature extraction, and trains a generative artificial intelligence model. The server then analyzes the new data to predict market size and untapped markets in Tokyo and Osaka. Along with the analysis results, the device recognizes the user's emotions using an emotion engine, and the server integrates this data and visualizes it appropriately. If the user is not feeling stressed, a detailed graphical interface is displayed; if they are feeling stressed, it switches to a simpler display. Finally, the server proposes a staffing plan and displays it to the user through the device. The user can then plan to effectively allocate 12 sales staff in Tokyo and 8 in Osaka.

[0868] Thus, the present invention provides a consistent series of processes, from the collection and analysis of regional data to sentiment recognition, visualization of results, and proposal of personnel allocation, and by taking user sentiment into consideration, it brings even greater convenience and accuracy.

[0869] The following describes the processing flow.

[0870] Step 1: Data Collection

[0871] User: Specify the region to be analyzed (e.g., Tokyo, Osaka) and the required data types (purchase data, demographic information, economic data).

[0872] Terminal: Collects necessary data from specified data sources (government statistics databases, commercial databases, web APIs, etc.) and sends it to the server.

[0873] Server: Receives data and stores it in a central database. This ensures that all data is managed centrally.

[0874] Step 2: Data Preprocessing

[0875] Server: Perform data cleansing to detect, delete, or correct inaccurate or abnormal data. For example, delete outliers in purchase data.

[0876] Server: Imputate missing data. Imputate missing data with the mean or median, or use a machine learning model to calculate predicted values.

[0877] Server: Saves the pre-processed data back to the database.

[0878] Step 3: Feature Extraction

[0879] Server: Extracts features from pre-processed data. Specifically, it extracts demographic information (age, gender, income level, etc.), consumer behavior (purchase frequency, type of purchased goods, average purchase amount, etc.), and regional characteristics (infrastructure status, geographical characteristics, number of competing companies, etc.).

[0880] Server: Organizes the extracted features and saves them as training data.

[0881] Step 4: Model Training

[0882] Server: Trains a generative artificial intelligence model using the extracted features. This process involves splitting the data into a training set, a validation set, and a test set.

[0883] Server: Evaluates the model's performance and adjusts hyperparameters as needed.

[0884] Server: Stores the trained model.

[0885] Step 5: Analysis and Prediction

[0886] Server: Uses a pre-trained model to analyze new data. The analysis predicts market size and untapped market potential in each region.

[0887] Server: Organizes the prediction results and saves them to the database.

[0888] Step 6: Emotion Recognition

[0889] Device: Recognizes the user's emotions using an emotion engine. Analyzes emotions from facial expressions and voice using the camera and microphone.

[0890] Server: Stores recognized emotion data in a database and performs necessary processing to incorporate it into the analysis.

[0891] Step 7: Visualizing the Results

[0892] Server: Combines prediction results and user sentiment data to generate data that can be visualized in an easy-to-understand format (e.g., graphs, maps, dashboards).

[0893] Terminal: Displays visualized data to the user. The user understands the specific analysis results by viewing graphs and maps.

[0894] Step 8: Emotion-based display adjustments

[0895] Server: Based on the emotion data recognized by the emotion engine, adjusts how the analysis results are displayed. For example, if the user is feeling stressed, simplify the display.

[0896] Step 9: Propose staffing arrangements

[0897] Server: Based on the analysis results, propose effective staffing arrangements. These proposals will take into account regional market size, purchasing power, and user sentiment data.

[0898] Terminal: Displays the proposed staffing plan to the user. The user then develops the optimal staffing plan based on this.

[0899] For example, if a user designates Tokyo and Osaka as target regions, the device collects purchasing data, demographic information, and economic data for these areas. The server stores this data in a database, performs data cleansing and feature extraction, and trains a generative artificial intelligence model. The server then analyzes the new data to predict market size and untapped markets in Tokyo and Osaka. Along with the analysis results, the device recognizes the user's emotions using an emotion engine, and the server integrates this data and visualizes it appropriately. If the user is not feeling stressed, a detailed graphical interface is displayed; if they are stressed, it switches to a simpler display. Finally, the server proposes a staffing plan and displays it to the user through the device. The user can then plan to effectively allocate 12 sales staff to Tokyo and 8 to Osaka.

[0900] (Example 2)

[0901] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0902] In modern society, effectively collecting and analyzing regional data and conducting market analysis is crucial for enhancing a company's competitiveness. However, current systems fragment the entire process from data collection to analysis and result display, making it difficult to perform efficiently. Furthermore, there is a lack of systems that provide advanced features such as displaying analysis results while considering user emotions and suggesting personnel allocation. This can lead to user stress and a risk of decreased business efficiency due to inability to implement optimal personnel allocation.

[0903] In Example 2, the identification processing by the identification processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for collecting regional data, means for storing the collected data in a database, means for performing data cleansing on the stored data, means for imputing missing data, means for extracting features from the data, means for training a generative artificial intelligence model using the extracted features, means for analyzing new data using the trained model and predicting market size, means for recognizing user emotions, means for reflecting the recognized emotion data in the analysis, means for visualizing and displaying the prediction results, means for adjusting the display method based on the emotion data, and means for proposing personnel allocation based on the analysis results. This enables the consistent execution of a series of processes, flexible display of analysis results in accordance with user emotions, and effective proposal of personnel allocation.

[0904] "Regional data" refers to various types of information related to a specific region, and specifically includes purchasing data, population information, and economic data.

[0905] A "database" is a system for centrally managing and storing collected data; examples include MySQL and PostgreSQL.

[0906] "Data cleansing" is the process of detecting, correcting, and removing inaccurate data and noise in order to improve the quality of the data.

[0907] "Missing data" refers to values ​​that are missing from a dataset, and a means of filling in these missing values ​​is required.

[0908] "Features" are elements of data extracted for use in analysis and model training, and specific examples include demographic information and consumer behavior.

[0909] A "generative artificial intelligence model" is a machine learning model that is trained on collected and pre-processed data to perform predictions and analyses.

[0910] "Emotion recognition" is a technology that detects a user's emotional state, using an emotion engine to analyze emotions from facial expressions and voice.

[0911] "Visualization" refers to displaying analysis results in a format that is easy for users to understand, and graphs and maps are commonly used for this purpose.

[0912] "Staffing" refers to a plan to effectively allocate the optimal number of staff based on analysis results.

[0913] The "display method" refers to the interface used to present analysis results to the user, and it is adjusted according to the user's emotions.

[0914] This invention is a system that collects and analyzes regional data to perform market analysis and propose personnel allocation, and in particular incorporates an emotion engine that recognizes user emotions. The following hardware and software are used in the implementation of this system.

[0915] The server is primarily responsible for data storage, preprocessing, feature extraction, model training, analysis, and visualization of results. The following software should be installed on the server:

[0916] Database management systems (e.g., MySQL, PostgreSQL)

[0917] Data analysis libraries (e.g., Pandas, NumPy)

[0918] Machine learning libraries (e.g., scikit-learn, TensorFlow, PyTorch)

[0919] Visualization libraries (e.g., Matplotlib, Plotly)

[0920] The device is responsible for data collection, emotion recognition, and partial data processing. Specifically, it is connected to a camera and microphone to recognize the user's emotions using an emotion engine. Affectiva's SDK is a suitable emotion engine to use.

[0921] The following is an example of the specific processing performed by this system.

[0922] Specific example

[0923] For example, suppose a user specifies Tokyo and Osaka as target regions. In this case,

[0924] 1. The device collects purchasing data, demographic information, and economic data related to these regions from APIs and websites. The device then sends the collected data to the server.

[0925] 2. The server saves the received data to the database. MySQL is used for database management.

[0926] 3. The server performs data cleansing, removing NaN values ​​and duplicate data. The Pandas library is used for this process.

[0927] 4. The server imputes missing data. For example, it can use scikit-learn's Inputter to impute missing values ​​with the mean.

[0928] 5. The server extracts useful features from the pre-processed data. These features include demographic information (age, gender, income level) and consumer behavior (purchase frequency, type of purchased goods, average purchase amount).

[0929] 6. Save the features as training data and train a generative artificial intelligence model. TensorFlow will be used to train the model.

[0930] 7. The server analyzes new data and predicts market size and untapped market potential in Tokyo and Osaka.

[0931] 8. The device uses an emotion engine to recognize the user's emotions. For example, it uses a camera to capture facial expressions and analyzes the emotional state.

[0932] 9. The server saves the prediction results and recognized emotion data to a database and incorporates them into the analysis results.

[0933] 10. The server visualizes the prediction results in the form of graphs or maps. For example, it can generate interactive graphs using the Plotly library.

[0934] 11. The device displays visualized data to the user. If the user is experiencing stress, measures are taken to simplify the display.

[0935] 12. Based on the analysis results, the server proposes an effective staffing plan. For example, it might plan to allocate 12 sales staff to Tokyo and 8 to Osaka.

[0936] 13. The terminal displays the proposed staffing plan to the user. The user can then use this to determine the optimal staffing arrangement.

[0937] This system uses a generative artificial intelligence model to perform market analysis while simultaneously incorporating user sentiment data to propose more appropriate information presentations and personnel allocation plans to users.

[0938] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0939] Step 1:

[0940] On the system's administration screen, users specify the target region for analysis (e.g., Tokyo, Osaka) and the required data types (purchase data, demographic information, economic data). Based on this, a request for data collection is created.

[0941] Input: Target region and data type

[0942] Output: Data collection request

[0943] Step 2:

[0944] The device collects the necessary data from a specified data source (e.g., an API, website, database, etc.). Specifically, it uses a Python script to issue an HTTP request to the API and sends the retrieved data to the server in JSON format.

[0945] Input: Data collection request

[0946] Output: Collected data (JSON format)

[0947] Step 3:

[0948] The server stores the received data in a central database. A DBMS such as MySQL is used for this purpose. Specifically, data is registered in the database by issuing SQL INSERT statements.

[0949] Input: Collected data (JSON format)

[0950] Output: Data stored in the database

[0951] Step 4:

[0952] The server performs data cleansing. For example, it processes the dataframe using the Pandas library to remove NaN values ​​and duplicate data. It detects and corrects inaccurate data and noise to generate a clean dataset.

[0953] Input: Data retrieved from the database

[0954] Output: Cleansed data

[0955] Step 5:

[0956] The server will impute the missing data. Specifically, it will use scikit-learn's Inputter to impute missing values ​​with the mean or median, or it will use a machine learning model to perform the imputation.

[0957] Input: Cleansed data

[0958] Output: Interpolated data

[0959] Step 6:

[0960] The server extracts useful features from pre-processed data. Specifically, it extracts features based on demographic information, consumer behavior, and regional characteristics using Pandas or NumPy.

[0961] Input: Completed data

[0962] Output: Feature dataset

[0963] Step 7:

[0964] The server uses the extracted features to train a generative artificial intelligence model. Specifically, it splits the data into a training set, a validation set, and a test set, and trains the model using TensorFlow.

[0965] Input: Feature dataset

[0966] Output: Trained model

[0967] Step 8:

[0968] The server uses a pre-trained model to perform analysis on new data. Specifically, it inputs new data into the model and predicts market size.

[0969] Input: New data, trained model

[0970] Output: Forecast results (market size, potential)

[0971] Step 9:

[0972] The device uses an emotion engine to recognize the user's emotions. For example, it uses a camera to capture facial expressions and performs emotion analysis.

[0973] Input: User's facial expression, emotion engine

[0974] Output: Recognized emotion data

[0975] Step 10:

[0976] The server stores the recognized emotion data in a database and performs the necessary processing to incorporate it into the analysis. Specifically, the emotion data is incorporated into the analysis algorithm.

[0977] Input: Prediction results, sentiment data

[0978] Output: Emotion-reflected analysis results

[0979] Step 11:

[0980] The server visualizes the sentiment-infused analysis results in formats such as graphs, maps, and dashboards. For example, it can generate interactive graphs using the Plotly library.

[0981] Input: Sentiment-reflected analysis results

[0982] Output: Visualization data (graphs, maps)

[0983] Step 12:

[0984] The device displays visualized data to the user. If the user is experiencing stress, measures such as simplifying the display will be taken.

[0985] Input: Visualization data, user emotional state

[0986] Output: Screen displayed to the user

[0987] Step 13:

[0988] The server analyzes the data and proposes an effective staffing plan. For example, it might plan to deploy 12 sales staff in Tokyo and 8 in Osaka.

[0989] Input: Visualization data, analysis results

[0990] Output: Staffing proposal

[0991] Step 14:

[0992] The terminal displays the proposed staffing plan to the user. The user can then use this to determine the optimal staffing arrangement.

[0993] Input: Personnel allocation proposal

[0994] Output: Final screen displayed to the user

[0995] (Application Example 2)

[0996] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0997] Conventional market analysis and staffing suggestion systems analyze data based solely on regional and purchasing data, failing to consider user emotions. This can lead to stress and confusion for users when receiving analysis results. Furthermore, there was no system that could acquire real-time emotional data and immediately suggest staffing and product placement based on it. To address these challenges, a system is needed that recognizes user emotions and adjusts the display of analysis results accordingly.

[0998] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for collecting regional data, means for storing the collected data in a database, means for performing data cleansing on the stored data, means for extracting features from the data, means for training a generative artificial intelligence model using the extracted features, means for analyzing new data using the trained model and predicting market size, means for visualizing and displaying the prediction results, means for proposing personnel allocation based on the analysis results, means for recognizing the user's emotions in real time, means for visualizing the analysis results including the recognized emotion data and adjusting the display method, and means for presenting the results to the user in real time using a device such as smart glasses. This makes it possible to present analysis results that take into account the user's emotions and to propose appropriate personnel allocation and product placement based on them.

[0999] "Regional data" is a general term for information including population data, economic data, and purchasing data for a specific geographical area.

[1000] A "database" is a digital system that centrally manages collected and stored data, enabling efficient searching and utilization.

[1001] "Data cleansing" is a processing method used to reduce inaccuracies and noise in data and prepare it for analysis.

[1002] "Features" are variables that represent important patterns or properties in data analysis and machine learning model training.

[1003] A "generative artificial intelligence model" is a type of machine learning model that generates useful information from data and makes predictions and decisions by learning patterns.

[1004] "Market size" refers to an indicator that shows the size of a market, such as the number of consumers or purchasing power in a particular region or market.

[1005] "Emotional data" refers to emotional information extracted from human facial expressions and voices collected using cameras and microphones.

[1006] "Smart glasses" are a type of wearable device, specifically glasses-shaped equipment that incorporates displays and sensors to show information in real time.

[1007] A "user" is a person or organization that uses this system and is the recipient of the analysis results and suggestions.

[1008] "Personnel allocation" refers to the plan or method of appropriately allocating staff or workers for the purpose of efficient business operations and service provision.

[1009] This invention relates to a system for collecting, analyzing, recognizing emotions in regional data, visualizing the results, and proposing personnel allocation. A key feature of this system is that it uses devices such as smart glasses to recognize the user's emotions in real time and reflects this in the analysis results and personnel allocation proposals.

[1010] 1. Data Collection

[1011] The server collects regional data from specified data sources and stores it in a database. This data includes diverse information such as purchasing data, population information, and economic data. This data collection allows the system to understand the market characteristics of each region.

[1012] 2. Data preprocessing

[1013] The server performs data cleansing on the collected data to reduce unwanted noise. Furthermore, it detects and removes or corrects inaccurate or anomalous data. Missing data is imputed using the mean or median, or through machine learning models.

[1014] 3. Feature Extraction

[1015] The server extracts useful features from the pre-processed data. These include demographic information (age, gender, income level, etc.), consumer behavior (purchase frequency, type of purchased goods, average purchase amount, etc.), and regional characteristics (infrastructure status, geographical features, number of competing companies, etc.). This extracts important data points that are then used for analysis.

[1016] 4. Model training

[1017] The server trains a generative AI model using the extracted features. This training process involves splitting the data (training set, validation set, test set). Once the training is complete, the model's performance is evaluated, and hyperparameters are adjusted as needed.

[1018] 5. Analysis and Prediction

[1019] The server uses a pre-trained model to analyze new data and predict market size and untapped market potential in each region. The prediction results are stored in a database.

[1020] 6. Emotion recognition

[1021] The device (smart glasses) uses a camera and microphone to capture the user's facial image and voice data, which are then analyzed by an emotion recognition engine. For example, it analyzes a customer's face in real time and obtains their emotion data. This emotion data is sent to a server and further incorporated into the analysis results.

[1022] 7. Visualization of Results

[1023] The server combines prediction results with user sentiment data and generates data to be visualized in an easy-to-understand format (graphs, maps, dashboards, etc.). The device (smart glasses) displays this visualized data to the user. The user can understand the analysis results while viewing graphs and maps.

[1024] 8. Emotion-based display adjustments

[1025] The server adjusts how the analysis results are displayed based on the emotional data recognized by the emotion recognition engine. For example, if the user is feeling stressed, the display may be simplified.

[1026] 9. Proposal for staffing arrangements

[1027] The server proposes an effective staffing plan based on the analysis results. This proposal takes into account regional market size, purchasing power, and even user sentiment data. The terminal (smart glasses) displays the proposed staffing plan to the user. The user can then develop an optimal staffing plan based on this.

[1028] Hardware and software to be used

[1029] Smart glasses: Recognizing customer emotions in real time (e.g., Google Glass, Vuzix Blade)

[1030] Server: Cloud infrastructure for data collection and analysis (e.g., AWS, Google Cloud)

[1031] Emotion recognition engine: (Examples: Microsoft Azure Face API, Google Cloud Vision API)

[1032] Database: (e.g., MySQL, MongoDB)

[1033] Machine learning libraries: (e.g., scikit-learn for Python, TensorFlow)

[1034] Example of a prompt:

[1035] "Using customer sentiment data and sales data, propose the optimal product placement strategy to increase store sales."

[1036] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1037] Step 1:

[1038] The server collects regional data from specified data sources and stores it in a database.

[1039] Input: Purchase data, demographic information, economic data

[1040] Data processing: Data collection and organization

[1041] Output: Data organized into a unified format is stored in the database.

[1042] Step 2:

[1043] The server performs data cleansing on the collected data to reduce unwanted noise.

[1044] Input: Original data stored in the database

[1045] Data processing: Noise reduction, detection and correction of inaccurate data.

[1046] Output: Cleansed data

[1047] Step 3:

[1048] The server will fill in any missing values ​​in the data.

[1049] Input: Cleansed data

[1050] Data processing: Imputation of missing values ​​(using mean, median, or machine learning models)

[1051] Output: Completed and cleansed data

[1052] Step 4:

[1053] The server extracts useful features from the pre-processed data.

[1054] Input: Completed and cleansed data

[1055] Data processing: Feature extraction (demographic information, consumer behavior, regional characteristics, etc.)

[1056] Output: Feature data

[1057] Step 5:

[1058] The server uses the extracted features to train a generative AI model.

[1059] Input: Feature data

[1060] Data processing: Splitting into training, validation, and test sets; model training and evaluation.

[1061] Output: Trained AI model

[1062] Step 6:

[1063] The server uses a pre-trained model to analyze new data and predict market size.

[1064] Input: Trained AI model, new data

[1065] Data processing: Analysis and prediction

[1066] Output: Forecast data (market size, potential of untapped markets)

[1067] Step 7:

[1068] The device (smart glasses) uses a camera and microphone to capture the user's facial image and voice data, which are then analyzed by an emotion recognition engine.

[1069] Input: Customer's facial image, voice data

[1070] Data processing: Analysis using an emotion recognition engine

[1071] Output: User sentiment data

[1072] Step 8:

[1073] The server combines prediction results with user sentiment data and visualizes them in an easy-to-understand format (graphs, maps, dashboards, etc.).

[1074] Input: Prediction result data, user sentiment data

[1075] Data processing: Generation of visualized data (graphs, maps, dashboards)

[1076] Output: Visualization data

[1077] Step 9:

[1078] The device (smart glasses) displays the visualized analysis results to the user and adjusts the display method as needed.

[1079] Input: Visualization data, user sentiment data

[1080] Data processing: Adjusting display methods based on emotions

[1081] Output: Adjusted display data

[1082] Step 10:

[1083] Based on the analysis results, the server proposes an effective staffing allocation.

[1084] Input: Analysis results data, user sentiment data

[1085] Data calculation: Optimization calculation of personnel allocation

[1086] Output: Personnel allocation proposal data

[1087] Step 11:

[1088] The device (smart glasses) displays the proposed staffing plan to the user.

[1089] Input: Personnel allocation proposal data

[1090] Data processing: Real-time display

[1091] Output: Staffing proposals to be reviewed by the user

[1092] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1093] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1094] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[1095] [Third Embodiment]

[1096] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[1097] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[1098] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1099] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[1100] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[1101] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[1102] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[1103] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[1104] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1105] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1106] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[1107] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[1108] This invention relates to a system for collecting and analyzing regional data to provide effective market analysis and personnel allocation suggestions. The processing of this system's program is described below in natural language.

[1109] 1. Data Collection

[1110] User: Specify the region to be analyzed and the required data types (e.g., purchasing data, demographic information, economic data).

[1111] Terminal: Collects necessary data from the specified data source and sends it to the server.

[1112] Server: Stores collected data in a central database. This ensures that all data is managed centrally.

[1113] 2. Data preprocessing

[1114] Server: Perform data cleansing to reduce unwanted noise. Detect and remove inaccurate or abnormal data.

[1115] Server: Imputes missing data. For example, it may use imputation methods such as imputing with the mean or median, or using machine learning models.

[1116] 3. Feature Extraction

[1117] Server: Extracts features from pre-processed data. Specifically, it extracts demographic information (age, gender, income level, etc.), consumer behavior (purchase frequency, type of purchased goods, average purchase amount, etc.), and regional characteristics (infrastructure status, geographical characteristics, number of competing companies, etc.).

[1118] Server: Stores the extracted features as training data.

[1119] 4. Model training

[1120] Server: Trains a generative artificial intelligence model using the extracted features. The training process includes splitting the data (training set, validation set, test set).

[1121] Server: Evaluates the trained model and adjusts hyperparameters as needed.

[1122] 5. Analysis and Prediction

[1123] Server: Uses a pre-trained model to analyze new data. This analysis predicts the market size and potential of untapped markets in each region.

[1124] Server: Organizes the analysis results and saves them to the database.

[1125] 6. Visualization of Results

[1126] Server: Generates data to visualize analysis results in an easy-to-understand format (e.g., graphs, maps, dashboards).

[1127] Terminal: Displays visualized data to the user. The user understands the specific analysis results by viewing graphs and maps.

[1128] 7. Proposal for staffing arrangements

[1129] Server: Based on the analysis results, propose effective staffing arrangements. These proposals will take into account the market size and purchasing power of each region.

[1130] Terminal: Displays the proposed staffing plan to the user. The user can then create an optimal staffing plan based on this proposal.

[1131] Specific example

[1132] For example, if a user selects Tokyo and Osaka as target regions, the terminal collects purchasing data, population information, and economic data for Tokyo and Osaka. The server stores this data in a database and performs data cleansing and feature extraction. A generative artificial intelligence model is trained using the extracted features, and new data is analyzed by the model. As a result, the market size and potential of untapped markets in Tokyo and Osaka are predicted. The server visualizes the prediction results and displays them to the user through the terminal. Based on these results, the user can plan to deploy 12 sales staff in Tokyo and 8 sales staff in Osaka. In this way, efficient market analysis and personnel allocation are achieved.

[1133] Thus, the present invention provides a consistent process from the collection and analysis of regional data to the visualization of results and the proposal of personnel allocation, resulting in significant benefits and convenience.

[1134] The following describes the processing flow.

[1135] Step 1: Data Collection

[1136] User: Specify the region to be analyzed and the required data types (e.g., purchasing data, demographic information, economic data).

[1137] Terminal: Collects data from the specified data source (e.g., government statistics database, commercial database, web API, etc.).

[1138] Terminal: Sends collected data to the server.

[1139] Server: Stores received data in a database for later storage.

[1140] Step 2: Data Preprocessing

[1141] Server: Perform data cleansing. Detect and delete or correct inaccurate or abnormal data.

[1142] Server: Performs processing to impute missing data. For example, it can impute missing data with the mean or median, or use a machine learning model to calculate predicted values.

[1143] Server: Updates the database to store the data after preprocessing is complete.

[1144] Step 3: Feature Extraction

[1145] Server: Extracts useful features from pre-processed data. Specifically, it extracts demographic information (age, gender, income level, etc.), consumer behavior (purchase frequency, type of purchased goods, average purchase amount, etc.), and regional characteristics (infrastructure status, geographical characteristics, number of competing companies, etc.).

[1146] Server: Organizes the extracted features and saves them as training data.

[1147] Step 4: Model Training

[1148] Server: Trains a generative artificial intelligence model using the extracted features. The training process includes splitting the data (training set, validation set, test set).

[1149] Server: Evaluates the model's performance and adjusts hyperparameters as needed.

[1150] Server: Stores the trained model.

[1151] Step 5: Analysis and Prediction

[1152] Server: Performs analysis on new data using the trained model.

[1153] Server: Predicts market size and untapped market potential in each region.

[1154] Server: Organizes the prediction results and saves them to the database.

[1155] Step 6: Visualizing the Results

[1156] Server: Generates data to visualize prediction results in an easy-to-understand format (e.g., graphs, maps, dashboards).

[1157] Terminal: Displays visualized data to the user. The user understands the specific analysis results by viewing graphs and maps.

[1158] Step 7: Propose staffing arrangements

[1159] Server: Based on the analysis results, proposes effective staffing arrangements. These proposals take into account regional market size and purchasing power.

[1160] Terminal: Displays the proposed staffing plan to the user. The user then develops the optimal staffing plan based on this.

[1161] For example, if a user designates Tokyo and Osaka as target regions, the terminal collects purchasing data, demographic information, and economic data for these regions. The server performs data cleansing and feature extraction, and trains a generative artificial intelligence model. The server then analyzes the new data to predict the market size and untapped markets for Tokyo and Osaka. Finally, the server visualizes the prediction results and displays them to the user through the terminal. Based on the proposed staffing plan, the user effectively allocates 12 sales staff to Tokyo and 8 to Osaka.

[1162] (Example 1)

[1163] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1164] Traditional marketing analysis methods made it difficult to collect and analyze data that differed by region and to allocate personnel effectively. Furthermore, there was no system that consistently handled data cleansing, data reconciliation, and visualization of analysis results. As a result, companies were unable to conduct efficient market analysis and optimal personnel allocation, hindering business efficiency and accurate understanding of target markets.

[1165] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1166] In this invention, the server includes means for specifying the region and data type to be analyzed; means for collecting the specified data; means for storing the collected data in a database; means for performing data cleansing on the stored data to remove inaccurate data and outliers; means for imputing missing data; means for extracting features from the data; means for training a generative artificial intelligence model using the extracted features; means for analyzing new data and predicting market size using the trained artificial intelligence model; means for storing and visualizing the prediction results in a database; means for proposing staffing arrangements based on the analysis results; and means for displaying the proposed staffing arrangements. This makes it possible to consistently perform detailed market analysis and effective staffing arrangements for each region.

[1167] "Analysis target area" refers to the specific region that the system targets when performing marketing analysis.

[1168] "Data type" refers to the various types of data necessary for analysis in the collection of regional data. Examples include purchasing data, population information, and economic data.

[1169] "Data cleansing" refers to the process of detecting inaccurate information and outliers contained in collected data, and then removing or correcting them.

[1170] "Missing data" refers to information that is missing from a dataset.

[1171] "Features" refer to data attributes or variables that have significant meaning for data analysis.

[1172] A "generative artificial intelligence model" refers to a model that uses machine learning and deep learning techniques to make predictions and classifications on new data.

[1173] "Market size" refers to the expected sales or purchasing power in a particular region or market.

[1174] "Visualization" refers to the process of visually displaying data and analysis results using graphs, dashboards, maps, and other visual tools.

[1175] "Personnel allocation" refers to the process by which a company or organization distributes the necessary number of staff members to appropriate locations and roles in order to effectively carry out its operations.

[1176] This invention relates to a system for collecting and analyzing regional data to provide effective market analysis and personnel allocation suggestions.

[1177] System program configuration

[1178] This system begins with the user specifying the analysis area and data type. The terminal collects data and sends it to the server, which processes the collected data to propose market analysis and staffing strategies. This entire process involves multiple hardware and software components.

[1179] Hardware:

[1180] Device: A computer or smartphone that can access the internet or APIs for data collection.

[1181] Server: A high-performance computer used for data processing, storage, and analysis.

[1182] software:

[1183] Database management systems: Used to centrally manage collected data. Examples include MySQL and PostgreSQL.

[1184] Data cleansing tools: These cleanse the collected data. For example, the pandas library in Python.

[1185] Generative AI models: Machine learning models that perform market analysis using features. Examples include Scikit-learn and TensorFlow.

[1186] Visualization tools: Tools for visualizing and displaying analysis results. Examples include Tableau and D3.js.

[1187] Program processing flow

[1188] The system program performs the following steps:

[1189] 1. Data collection:

[1190] User: Specify the region to be analyzed and the required data types (purchase data, demographic information, economic data, etc.).

[1191] Terminal: Collects necessary data from specified data sources (open data portals or commercial databases) and sends it to the server.

[1192] Server: Stores data sent from terminals in a central database and manages it centrally.

[1193] 2. Data preprocessing:

[1194] Server: Performs data cleansing on the collected data, removing inaccurate data and outliers.

[1195] Server: Imputes missing data using mean values ​​and machine learning models.

[1196] 3. Feature extraction:

[1197] Server: Extracts features such as demographic information, consumer behavior, and regional characteristics from pre-processed data and saves them as training datasets.

[1198] 4. Model training:

[1199] Server: Trains a generative artificial intelligence model using the extracted feature data. If necessary, splits the data into training, validation, and test sets, and tunes hyperparameters.

[1200] 5. Analysis and Prediction:

[1201] Server: Uses a pre-trained model to analyze new data and predict market size and the potential of untapped markets.

[1202] Server: Organizes the prediction results and saves them to the database.

[1203] 6. Visualization of results:

[1204] Server: Generates visualization data such as graphs and maps based on the analysis results.

[1205] Terminal: Displays dashboards and graphs to the user based on visualization data sent from the server.

[1206] 7. Proposal for staffing arrangements:

[1207] Server: Based on the analysis results, propose effective staffing arrangements.

[1208] Terminal: Displays the proposed staffing plan to the user.

[1209] Explanation of specific examples

[1210] For example, if a user selects Tokyo and Osaka as target regions, the device collects purchasing data, demographic information, and economic data for Tokyo and Osaka through open data portals and commercial databases. The collected data is sent to a server and stored in a central database. Next, the server performs data cleansing, removing inaccurate data and filling in missing data. After that, the server extracts features such as purchase frequency and average purchase amount and stores them as a training dataset. Using the stored data, a generative AI model is trained to make predictions on new data. The analysis results are displayed visually, and the server proposes a staffing plan suitable for the user. Based on this information, the user can plan, for example, to deploy 12 sales staff in Tokyo and 8 in Osaka.

[1211] Example of a prompt

[1212] "Conduct a market analysis using purchasing data, demographic information, and economic data for Tokyo and Osaka, and propose effective staffing strategies."

[1213] Thus, the present invention provides a consistent process from the collection and analysis of regional data to the visualization of results and the proposal of personnel allocation, resulting in significant benefits and convenience.

[1214] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1215] Step 1:

[1216] User: Specify the region and data type to be analyzed. Specifically, the user uses a terminal to select "Tokyo purchasing data and population information" through the interface.

[1217] Input: Region to be analyzed and data type.

[1218] Output: Field information required for the data collection process.

[1219] Step 2:

[1220] Terminal: Collects data via the internet or APIs based on the user-specified analysis area and data type. For example, it accesses "open data portals" or "commercial databases" to obtain information.

[1221] Input: Region to be analyzed and data type.

[1222] Output: Collected raw data.

[1223] Specific operation: A request is sent to the "Open Data Portal API" to retrieve purchasing data for Tokyo.

[1224] Step 3:

[1225] Terminal: Sends collected data to the server.

[1226] Input: Collected raw data.

[1227] Output: Request containing collected data.

[1228] Specific operation: The collected data is sent to the server as an HTTP request via the API.

[1229] Step 4:

[1230] Server: Stores collected data in a central database. This ensures centralized data management.

[1231] Input: Collected data sent to the server.

[1232] Output: Data stored in the central database.

[1233] Specific operation: Use a database management system (e.g., MySQL) to insert data into a table.

[1234] Step 5:

[1235] Server: Perform data cleansing to remove inaccurate data and outliers. For example, remove outliers like "-1 yen" from purchase history data.

[1236] Input: Raw data stored in the central database.

[1237] Output: Cleansed data.

[1238] Specific operation: Use the Python pandas library to filter and remove inaccurate entries in the data.

[1239] Step 6:

[1240] Server: Imputes missing data using mean values ​​or machine learning models. For example, it can imputate missing age data using the average age of other data points.

[1241] Input: Cleansed data (including missing values).

[1242] Output: Interpolated data.

[1243] Specific operation: Missing values ​​are imputed using the K-nearest neighbors algorithm with the scikit-learn library in Python.

[1244] Step 7:

[1245] Server: Extracts features from pre-processed data. Specifically, it extracts demographic information such as age and income, and consumer behavior data such as purchase frequency and purchase amount.

[1246] Input: Completed data.

[1247] Output: Feature dataset.

[1248] Specific operation: Using the Python Feature-engine library, select important features and convert the data into a matrix format.

[1249] Step 8:

[1250] Server: The extracted features are used to split the data into training, validation, and test sets, and a generative artificial intelligence model is trained. Specifically, 70% is split into a training set, 15% into a validation set, and 15% into a test set, and a regression model is trained.

[1251] Input: Feature dataset.

[1252] Output: Trained model.

[1253] Specific operation: The data is split using the scikit-learn library, and a Linear Regression model is trained.

[1254] Step 9:

[1255] Server: Uses a pre-trained model to analyze new data and predict market size by region.

[1256] Input: A trained model and new data.

[1257] Output: Prediction result.

[1258] Specific actions: Input new data into the model and perform regression analysis to predict market size.

[1259] Step 10:

[1260] Server: Organizes and stores prediction results in a database.

[1261] Input: Prediction result.

[1262] Output: Prediction results stored in the database.

[1263] Specific operation: Use an SQL query to save the prediction results to the appropriate table in the database.

[1264] Step 11:

[1265] Server: Converts analysis results into visualization data such as graphs and maps.

[1266] Input: Prediction results stored in the database.

[1267] Output: Visualized data.

[1268] Specific operation: Use Matplotlib and D3.js to convert the prediction results into a visually displayable format.

[1269] Step 12:

[1270] Terminal: Displays dashboards and graphs to the user based on visualization data sent from the server.

[1271] Input: Visualization data from the server.

[1272] Output: Graphs and maps displayed in the user interface.

[1273] Specific operation: Use a web application framework (e.g., React) to dynamically display data in the user interface.

[1274] Step 13:

[1275] Server: Based on the analysis results, it proposes effective personnel allocation for each region.

[1276] Input: Analysis results.

[1277] Output: Staffing proposal.

[1278] Specific operation: Based on the prediction results, calculate the optimal staffing allocation for each region and generate a proposal.

[1279] Step 14:

[1280] Terminal: Displays the proposed staffing plan to the user. The user can make decisions based on the proposal.

[1281] Input: Personnel allocation proposal.

[1282] Output: Staffing suggestions displayed in the user interface.

[1283] Specific operation: Use a web application framework to display the contents of the proposal in a user interface.

[1284] (Application Example 1)

[1285] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1286] Efficient market analysis and staffing planning in physical stores are not easy, and responding quickly to changes in market trends is particularly challenging. Furthermore, data inaccuracies that occur during the process of collecting and analyzing regional data, as well as a lack of means to support visual decision-making, are also challenges.

[1287] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1288] In this invention, the server includes means for collecting regional data, means for storing the collected data in a database, means for performing data cleansing on the stored data, means for extracting features from the data, means for training a generative artificial intelligence model using the extracted features, means for analyzing new data using the trained model and predicting market size, means for visualizing and displaying the prediction results, means for proposing staffing arrangements based on the analysis results, means for notifying changes in market trends for a specific region in real time, and means for displaying the prediction results on an electronic map to support visual judgments for store placement and market analysis. This enables efficient market analysis and staffing planning, and allows for quick and accurate responses to changes in market trends.

[1289] "Regional data" refers to data that includes population information, economic data, purchasing data, and related infrastructure information for a specific geographical area.

[1290] A "database" is a system for centrally managing and storing collected data.

[1291] "Data cleansing" is a technique that improves the quality of collected data by removing noise and inaccurate data.

[1292] "Features" refer to statistically extracted information necessary for data analysis, and are primarily used as input for analytical models.

[1293] A "generative artificial intelligence model" refers to a machine learning algorithm that performs predictions and analyses based on collected and processed data.

[1294] "Prediction" refers to estimating future data and trends using a trained artificial intelligence model.

[1295] "Visualization" refers to displaying analysis results in a visually easy-to-understand format, such as graphs, maps, and dashboards.

[1296] "Personnel allocation" refers to planning the placement of personnel in the most suitable locations and positions.

[1297] "Real-time notifications" is a feature that instantly informs users of market trends and other data changes.

[1298] An "electronic map" is a digital map that visually displays geographical information and forecast results.

[1299] This invention relates to a system for collecting and analyzing regional data to perform market analysis and propose personnel allocation strategies. The system functions through the cooperation of a server, terminals, and users.

[1300] 1. Data Collection

[1301] The user specifies the region to be analyzed and the required data types (e.g., purchasing data, demographic information, economic data). The terminal collects the necessary data from the specified data sources and sends it to the server. The server stores the collected data in a central database and manages all data centrally.

[1302] 2. Data preprocessing

[1303] The server performs data cleansing to reduce unnecessary noise. Specifically, it detects and removes inaccurate and anomalous data. It also uses methods such as imputation with the mean or median, or imputation using machine learning models, to fill in missing data.

[1304] 3. Feature Extraction

[1305] The server extracts features from the pre-processed data. Specifically, it extracts demographic information (age, gender, income level, etc.) and consumer behavior (purchase frequency, type of purchased goods, average purchase amount, etc.). It also includes regional characteristics (infrastructure status, geographical features, number of competing companies, etc.). The extracted features are stored as training data.

[1306] 4. Model training

[1307] The server uses the extracted features to train a generative artificial intelligence model. The training process includes splitting the data (training set, validation set, test set). The trained model is evaluated, and hyperparameters are adjusted as needed.

[1308] 5. Analysis and Prediction

[1309] The server uses a pre-trained model to analyze new data. This analysis predicts the market size and potential of untapped markets in each region. The analysis results are stored in a database for reuse.

[1310] 6. Visualization of Results

[1311] The server generates data that visualizes the analysis results in an easy-to-understand format (e.g., graphs, maps, dashboards). The terminal displays this visualized data to the user. This allows the user to understand the specific analysis results while viewing graphs and maps.

[1312] 7. Proposal for staffing arrangements

[1313] Based on the analysis results, the server proposes an effective staffing plan. This proposal takes into account the market size and purchasing power of each region. The terminal displays the proposed staffing plan to the user, helping them to create the optimal staffing plan.

[1314] It also includes a feature that provides real-time notifications of changes in market trends in specific regions. This feature provides alerts to quickly respond to sudden market fluctuations. Furthermore, it can display forecast results on an electronic map, visually supporting store placement and market analysis. This allows users to make more intuitive decisions.

[1315] Specific example

[1316] For example, suppose a user collects purchasing data, population data, and competitor information based on location data in Tokyo to perform market analysis. In this case, the server collects and preprocesses this data, extracts features, and trains an AI model. This model predicts the market size of Tokyo, and the results are visually presented to the user in the form of graphs and maps. Then, the optimal staffing allocation for stores in Tokyo is suggested.

[1317] Examples of prompt statements are as follows:

[1318] "Based on location data for Tokyo, collect in-store purchasing data, population information, and competitor information to conduct market analysis. Then, create an application that proposes appropriate staffing arrangements based on the results."

[1319] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1320] Step 1:

[1321] Data collection

[1322] The user specifies the target region and the required data type (purchase data, population information, economic data, etc.). The terminal collects the necessary data from the specified data source (public API, website, statistical database, etc.) and sends the collected data to the server. The server stores the received data in a central database.

[1323] Input: User-specified region and data type

[1324] Output: Regional data stored on the server

[1325] Step 2:

[1326] Data preprocessing

[1327] The server performs data cleansing on the received data. Specifically, it normalizes unstructured data, reduces noisy data, and detects and removes inaccurate or outlier values. Next, to impute missing data, it uses the mean or median, or applies imputation methods using machine learning models.

[1328] Input: Raw data stored in the central database

[1329] Output: Cleansed and imputed data

[1330] Step 3:

[1331] Feature extraction

[1332] The server extracts features from the pre-processed data. Specifically, it extracts demographic information (age, gender, income level, etc.), consumer behavior (purchase frequency, type of purchased goods, average purchase amount, etc.), and regional characteristics (infrastructure status, geographical characteristics, number of competing stores, etc.), and stores them as training data.

[1333] Input: Pre-processed data

[1334] Output: Training data with extracted features

[1335] Step 4:

[1336] Model training

[1337] The server uses the extracted features to train a generative artificial intelligence model. This process involves splitting the training data into training, validation, and test sets, selecting appropriate hyperparameters, and optimizing the model. The trained model is then used for subsequent analysis and prediction.

[1338] Input: Training data with extracted features

[1339] Output: Trained generative artificial intelligence model

[1340] Step 5:

[1341] Analysis and prediction

[1342] The server uses a pre-trained generative artificial intelligence model to analyze new data. This analysis predicts the market size and potential of untapped markets in each region, and these predictions are stored in a database.

[1343] Input: A trained generative AI model, and new data to be analyzed.

[1344] Output: Prediction results stored in the database

[1345] Step 6:

[1346] Visualization of results

[1347] The server generates data that visualizes the prediction results in formats such as graphs, maps, and dashboards. The terminal displays this visualized data to the user, allowing the user to visually confirm the analysis results.

[1348] Input: Prediction results stored in the database

[1349] Output: Visualized analysis results

[1350] Step 7:

[1351] Proposal for staffing arrangements

[1352] The server proposes an effective staffing plan based on the analysis results. The proposal takes into account the market size and purchasing power of each region. The terminal displays the proposed staffing plan to the user, allowing the user to create the optimal staffing plan.

[1353] Input: Visualized analysis results

[1354] Output: Proposed staffing arrangements

[1355] Step 8:

[1356] Real-time notifications

[1357] The server monitors market trends in a specific region in real time and notifies the user when a change is detected. This allows the user to respond quickly and accurately to market changes.

[1358] Input: Real-time monitored market data

[1359] Output: Notification of market changes to users

[1360] Step 9:

[1361] Use of electronic maps

[1362] The server generates data to display prediction results on an electronic map, and the terminal displays this to the user. This allows the user to visually perform store placement and market analysis.

[1363] Input: Prediction result

[1364] Output: Prediction results displayed on the electronic map

[1365] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1366] This invention combines a system that collects and analyzes regional data to perform effective market analysis and personnel allocation proposals with an emotion engine that recognizes user emotions. The processing of this system's program is described below in natural language.

[1367] 1. Data Collection

[1368] User: Specify the region to be analyzed and the required data types (e.g., purchasing data, demographic information, economic data).

[1369] Terminal: Collects necessary data from the specified data source and sends it to the server.

[1370] Server: Receives data and stores it in a central database. This ensures that all data is managed centrally.

[1371] 2. Data preprocessing

[1372] Server: Perform data cleansing to reduce unnecessary noise. Detect and delete or correct inaccurate or abnormal data.

[1373] Server: Performs processing to impute missing data. For example, it may use imputation methods such as imputation using the mean or median, or imputation using machine learning models.

[1374] Server: Updates the database to store the data after preprocessing is complete.

[1375] 3. Feature Extraction

[1376] Server: Extracts useful features from pre-processed data. Specifically, it extracts demographic information (age, gender, income level, etc.), consumer behavior (purchase frequency, type of purchased goods, average purchase amount, etc.), and regional characteristics (infrastructure status, geographical characteristics, number of competing companies, etc.).

[1377] Server: Organizes the extracted features and saves them as training data.

[1378] 4. Model training

[1379] Server: Trains a generative artificial intelligence model using the extracted features. The training process includes splitting the data (training set, validation set, test set).

[1380] Server: Evaluates the model's performance and adjusts hyperparameters as needed.

[1381] Server: Stores the trained model.

[1382] 5. Analysis and Prediction

[1383] Server: Performs analysis on new data using the trained model.

[1384] Server: Predicts market size and untapped market potential in each region.

[1385] Server: Organizes the prediction results and saves them to the database.

[1386] 6. Emotion recognition

[1387] Device: Recognizes the user's emotions using an emotion engine. For example, it analyzes emotions from facial expressions and voice using a camera and microphone.

[1388] Server: Stores recognized emotion data in a database and performs necessary processing to incorporate it into the analysis.

[1389] 7. Visualization of Results

[1390] Server: Combines prediction results and user sentiment data to generate data that can be visualized in an easy-to-understand format (e.g., graphs, maps, dashboards).

[1391] Terminal: Displays visualized data to the user. The user understands the specific analysis results by viewing graphs and maps.

[1392] 8. Emotion-based display adjustments

[1393] Server: Based on the emotion data recognized by the emotion engine, the server adjusts how the analysis results are displayed. For example, if the user is feeling stressed, the display is simplified.

[1394] 9. Proposal for staffing arrangements

[1395] Server: Based on the analysis results, propose effective staffing arrangements. These proposals will take into account regional market size, purchasing power, and user sentiment data.

[1396] Terminal: Displays the proposed staffing plan to the user. The user then develops the optimal staffing plan based on this.

[1397] Specific example

[1398] For example, if a user specifies Tokyo and Osaka as target regions, the device collects purchasing data, demographic information, and economic data for these regions. The server stores this data in a database, performs data cleansing and feature extraction, and trains a generative artificial intelligence model. The server then analyzes the new data to predict market size and untapped markets in Tokyo and Osaka. Along with the analysis results, the device recognizes the user's emotions using an emotion engine, and the server integrates this data and visualizes it appropriately. If the user is not feeling stressed, a detailed graphical interface is displayed; if they are feeling stressed, it switches to a simpler display. Finally, the server proposes a staffing plan and displays it to the user through the device. The user can then plan to effectively allocate 12 sales staff in Tokyo and 8 in Osaka.

[1399] Thus, the present invention provides a consistent series of processes, from the collection and analysis of regional data to sentiment recognition, visualization of results, and proposal of personnel allocation, and by taking user sentiment into consideration, it brings even greater convenience and accuracy.

[1400] The following describes the processing flow.

[1401] Step 1: Data Collection

[1402] User: Specify the region to be analyzed (e.g., Tokyo, Osaka) and the required data types (purchase data, demographic information, economic data).

[1403] Terminal: Collects necessary data from specified data sources (government statistics databases, commercial databases, web APIs, etc.) and sends it to the server.

[1404] Server: Receives data and stores it in a central database. This ensures that all data is managed centrally.

[1405] Step 2: Data Preprocessing

[1406] Server: Perform data cleansing to detect, delete, or correct inaccurate or abnormal data. For example, delete outliers in purchase data.

[1407] Server: Imputate missing data. Imputate missing data with the mean or median, or use a machine learning model to calculate predicted values.

[1408] Server: Saves the pre-processed data back to the database.

[1409] Step 3: Feature Extraction

[1410] Server: Extracts features from pre-processed data. Specifically, it extracts demographic information (age, gender, income level, etc.), consumer behavior (purchase frequency, type of purchased goods, average purchase amount, etc.), and regional characteristics (infrastructure status, geographical characteristics, number of competing companies, etc.).

[1411] Server: Organizes the extracted features and saves them as training data.

[1412] Step 4: Model Training

[1413] Server: Trains a generative artificial intelligence model using the extracted features. This process involves splitting the data into a training set, a validation set, and a test set.

[1414] Server: Evaluates the model's performance and adjusts hyperparameters as needed.

[1415] Server: Stores the trained model.

[1416] Step 5: Analysis and Prediction

[1417] Server: Uses a pre-trained model to analyze new data. The analysis predicts market size and untapped market potential in each region.

[1418] Server: Organizes the prediction results and saves them to the database.

[1419] Step 6: Emotion Recognition

[1420] Device: Recognizes the user's emotions using an emotion engine. Analyzes emotions from facial expressions and voice using the camera and microphone.

[1421] Server: Stores recognized emotion data in a database and performs necessary processing to incorporate it into the analysis.

[1422] Step 7: Visualizing the Results

[1423] Server: Combines prediction results and user sentiment data to generate data that can be visualized in an easy-to-understand format (e.g., graphs, maps, dashboards).

[1424] Terminal: Displays visualized data to the user. The user understands the specific analysis results by viewing graphs and maps.

[1425] Step 8: Emotion-based display adjustments

[1426] Server: Based on the emotion data recognized by the emotion engine, adjusts how the analysis results are displayed. For example, if the user is feeling stressed, simplify the display.

[1427] Step 9: Propose staffing arrangements

[1428] Server: Based on the analysis results, propose effective staffing arrangements. These proposals will take into account regional market size, purchasing power, and user sentiment data.

[1429] Terminal: Displays the proposed staffing plan to the user. The user then develops the optimal staffing plan based on this.

[1430] For example, if a user designates Tokyo and Osaka as target regions, the device collects purchasing data, demographic information, and economic data for these areas. The server stores this data in a database, performs data cleansing and feature extraction, and trains a generative artificial intelligence model. The server then analyzes the new data to predict market size and untapped markets in Tokyo and Osaka. Along with the analysis results, the device recognizes the user's emotions using an emotion engine, and the server integrates this data and visualizes it appropriately. If the user is not feeling stressed, a detailed graphical interface is displayed; if they are stressed, it switches to a simpler display. Finally, the server proposes a staffing plan and displays it to the user through the device. The user can then plan to effectively allocate 12 sales staff to Tokyo and 8 to Osaka.

[1431] (Example 2)

[1432] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1433] In modern society, effectively collecting and analyzing regional data and conducting market analysis is crucial for enhancing a company's competitiveness. However, current systems fragment the entire process from data collection to analysis and result display, making it difficult to perform efficiently. Furthermore, there is a lack of systems that provide advanced features such as displaying analysis results while considering user emotions and suggesting personnel allocation. This can lead to user stress and a risk of decreased business efficiency due to inability to implement optimal personnel allocation.

[1434] In Example 2, the identification processing by the identification processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for collecting regional data, means for storing the collected data in a database, means for performing data cleansing on the stored data, means for imputing missing data, means for extracting features from the data, means for training a generative artificial intelligence model using the extracted features, means for analyzing new data using the trained model and predicting market size, means for recognizing user emotions, means for reflecting the recognized emotion data in the analysis, means for visualizing and displaying the prediction results, means for adjusting the display method based on the emotion data, and means for proposing personnel allocation based on the analysis results. This enables the consistent execution of a series of processes, flexible display of analysis results in accordance with user emotions, and effective proposal of personnel allocation.

[1435] "Regional data" refers to various types of information related to a specific region, and specifically includes purchasing data, population information, and economic data.

[1436] A "database" is a system for centrally managing and storing collected data; examples include MySQL and PostgreSQL.

[1437] "Data cleansing" is the process of detecting, correcting, and removing inaccurate data and noise in order to improve the quality of the data.

[1438] "Missing data" refers to values ​​that are missing from a dataset, and a means of filling in these missing values ​​is required.

[1439] "Features" are elements of data extracted for use in analysis and model training, and specific examples include demographic information and consumer behavior.

[1440] A "generative artificial intelligence model" is a machine learning model that is trained on collected and pre-processed data to perform predictions and analyses.

[1441] "Emotion recognition" is a technology that detects a user's emotional state, using an emotion engine to analyze emotions from facial expressions and voice.

[1442] "Visualization" refers to displaying analysis results in a format that is easy for users to understand, and graphs and maps are commonly used for this purpose.

[1443] "Staffing" refers to a plan to effectively allocate the optimal number of staff based on analysis results.

[1444] The "display method" refers to the interface used to present analysis results to the user, and it is adjusted according to the user's emotions.

[1445] This invention is a system that collects and analyzes regional data to perform market analysis and propose personnel allocation, and in particular incorporates an emotion engine that recognizes user emotions. The following hardware and software are used in the implementation of this system.

[1446] The server is primarily responsible for data storage, preprocessing, feature extraction, model training, analysis, and visualization of results. The following software should be installed on the server:

[1447] Database management systems (e.g., MySQL, PostgreSQL)

[1448] Data analysis libraries (e.g., Pandas, NumPy)

[1449] Machine learning libraries (e.g., scikit-learn, TensorFlow, PyTorch)

[1450] Visualization libraries (e.g., Matplotlib, Plotly)

[1451] The device is responsible for data collection, emotion recognition, and partial data processing. Specifically, it is connected to a camera and microphone to recognize the user's emotions using an emotion engine. Affectiva's SDK is a suitable emotion engine to use.

[1452] The following is an example of the specific processing performed by this system.

[1453] Specific example

[1454] For example, suppose a user specifies Tokyo and Osaka as target regions. In this case,

[1455] 1. The device collects purchasing data, demographic information, and economic data related to these regions from APIs and websites. The device then sends the collected data to the server.

[1456] 2. The server saves the received data to the database. MySQL is used for database management.

[1457] 3. The server performs data cleansing, removing NaN values ​​and duplicate data. The Pandas library is used for this process.

[1458] 4. The server imputes missing data. For example, it can use scikit-learn's Inputter to impute missing values ​​with the mean.

[1459] 5. The server extracts useful features from the pre-processed data. These features include demographic information (age, gender, income level) and consumer behavior (purchase frequency, type of purchased goods, average purchase amount).

[1460] 6. Save the features as training data and train a generative artificial intelligence model. TensorFlow will be used to train the model.

[1461] 7. The server analyzes new data and predicts market size and untapped market potential in Tokyo and Osaka.

[1462] 8. The device uses an emotion engine to recognize the user's emotions. For example, it uses a camera to capture facial expressions and analyzes the emotional state.

[1463] 9. The server saves the prediction results and recognized emotion data to a database and incorporates them into the analysis results.

[1464] 10. The server visualizes the prediction results in the form of graphs or maps. For example, it can generate interactive graphs using the Plotly library.

[1465] 11. The device displays visualized data to the user. If the user is experiencing stress, measures are taken to simplify the display.

[1466] 12. Based on the analysis results, the server proposes an effective staffing plan. For example, it might plan to allocate 12 sales staff to Tokyo and 8 to Osaka.

[1467] 13. The terminal displays the proposed staffing plan to the user. The user can then use this to determine the optimal staffing arrangement.

[1468] This system uses a generative artificial intelligence model to perform market analysis while simultaneously incorporating user sentiment data to propose more appropriate information presentations and personnel allocation plans to users.

[1469] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1470] Step 1:

[1471] On the system's administration screen, users specify the target region for analysis (e.g., Tokyo, Osaka) and the required data types (purchase data, demographic information, economic data). Based on this, a request for data collection is created.

[1472] Input: Target region and data type

[1473] Output: Data collection request

[1474] Step 2:

[1475] The device collects the necessary data from a specified data source (e.g., an API, website, database, etc.). Specifically, it uses a Python script to issue an HTTP request to the API and sends the retrieved data to the server in JSON format.

[1476] Input: Data collection request

[1477] Output: Collected data (JSON format)

[1478] Step 3:

[1479] The server stores the received data in a central database. A DBMS such as MySQL is used for this purpose. Specifically, data is registered in the database by issuing SQL INSERT statements.

[1480] Input: Collected data (JSON format)

[1481] Output: Data stored in the database

[1482] Step 4:

[1483] The server performs data cleansing. For example, it processes the dataframe using the Pandas library to remove NaN values ​​and duplicate data. It detects and corrects inaccurate data and noise to generate a clean dataset.

[1484] Input: Data retrieved from the database

[1485] Output: Cleansed data

[1486] Step 5:

[1487] The server will impute the missing data. Specifically, it will use scikit-learn's Inputter to impute missing values ​​with the mean or median, or it will use a machine learning model to perform the imputation.

[1488] Input: Cleansed data

[1489] Output: Interpolated data

[1490] Step 6:

[1491] The server extracts useful features from pre-processed data. Specifically, it extracts features based on demographic information, consumer behavior, and regional characteristics using Pandas or NumPy.

[1492] Input: Completed data

[1493] Output: Feature dataset

[1494] Step 7:

[1495] The server uses the extracted features to train a generative artificial intelligence model. Specifically, it splits the data into a training set, a validation set, and a test set, and trains the model using TensorFlow.

[1496] Input: Feature dataset

[1497] Output: Trained model

[1498] Step 8:

[1499] The server uses a pre-trained model to perform analysis on new data. Specifically, it inputs new data into the model and predicts market size.

[1500] Input: New data, trained model

[1501] Output: Forecast results (market size, potential)

[1502] Step 9:

[1503] The device uses an emotion engine to recognize the user's emotions. For example, it uses a camera to capture facial expressions and performs emotion analysis.

[1504] Input: User's facial expression, emotion engine

[1505] Output: Recognized emotion data

[1506] Step 10:

[1507] The server stores the recognized emotion data in a database and performs the necessary processing to incorporate it into the analysis. Specifically, the emotion data is incorporated into the analysis algorithm.

[1508] Input: Prediction results, sentiment data

[1509] Output: Emotion-reflected analysis results

[1510] Step 11:

[1511] The server visualizes the sentiment-infused analysis results in formats such as graphs, maps, and dashboards. For example, it can generate interactive graphs using the Plotly library.

[1512] Input: Sentiment-reflected analysis results

[1513] Output: Visualization data (graphs, maps)

[1514] Step 12:

[1515] The device displays visualized data to the user. If the user is experiencing stress, measures such as simplifying the display will be taken.

[1516] Input: Visualization data, user emotional state

[1517] Output: Screen displayed to the user

[1518] Step 13:

[1519] The server analyzes the data and proposes an effective staffing plan. For example, it might plan to deploy 12 sales staff in Tokyo and 8 in Osaka.

[1520] Input: Visualization data, analysis results

[1521] Output: Staffing proposal

[1522] Step 14:

[1523] The terminal displays the proposed staffing plan to the user. The user can then use this to determine the optimal staffing arrangement.

[1524] Input: Personnel allocation proposal

[1525] Output: Final screen displayed to the user

[1526] (Application Example 2)

[1527] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1528] Conventional market analysis and staffing suggestion systems analyze data based solely on regional and purchasing data, failing to consider user emotions. This can lead to stress and confusion for users when receiving analysis results. Furthermore, there was no system that could acquire real-time emotional data and immediately suggest staffing and product placement based on it. To address these challenges, a system is needed that recognizes user emotions and adjusts the display of analysis results accordingly.

[1529] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for collecting regional data, means for storing the collected data in a database, means for performing data cleansing on the stored data, means for extracting features from the data, means for training a generative artificial intelligence model using the extracted features, means for analyzing new data using the trained model and predicting market size, means for visualizing and displaying the prediction results, means for proposing personnel allocation based on the analysis results, means for recognizing the user's emotions in real time, means for visualizing the analysis results including the recognized emotion data and adjusting the display method, and means for presenting the results to the user in real time using a device such as smart glasses. This makes it possible to present analysis results that take into account the user's emotions and to propose appropriate personnel allocation and product placement based on them.

[1530] "Regional data" is a general term for information including population data, economic data, and purchasing data for a specific geographical area.

[1531] A "database" is a digital system that centrally manages collected and stored data, enabling efficient searching and utilization.

[1532] "Data cleansing" is a processing method used to reduce inaccuracies and noise in data and prepare it for analysis.

[1533] "Features" are variables that represent important patterns or properties in data analysis and machine learning model training.

[1534] A "generative artificial intelligence model" is a type of machine learning model that generates useful information from data and makes predictions and decisions by learning patterns.

[1535] "Market size" refers to an indicator that shows the size of a market, such as the number of consumers or purchasing power in a particular region or market.

[1536] "Emotional data" refers to emotional information extracted from human facial expressions and voices collected using cameras and microphones.

[1537] "Smart glasses" are a type of wearable device, specifically glasses-shaped equipment that incorporates displays and sensors to show information in real time.

[1538] A "user" is a person or organization that uses this system and is the recipient of the analysis results and suggestions.

[1539] "Personnel allocation" refers to the plan or method of appropriately allocating staff or workers for the purpose of efficient business operations and service provision.

[1540] This invention relates to a system for collecting, analyzing, recognizing emotions in regional data, visualizing the results, and proposing personnel allocation. A key feature of this system is that it uses devices such as smart glasses to recognize the user's emotions in real time and reflects this in the analysis results and personnel allocation proposals.

[1541] 1. Data Collection

[1542] The server collects regional data from specified data sources and stores it in a database. This data includes diverse information such as purchasing data, population information, and economic data. This data collection allows the system to understand the market characteristics of each region.

[1543] 2. Data preprocessing

[1544] The server performs data cleansing on the collected data to reduce unwanted noise. Furthermore, it detects and removes or corrects inaccurate or anomalous data. Missing data is imputed using the mean or median, or through machine learning models.

[1545] 3. Feature Extraction

[1546] The server extracts useful features from the pre-processed data. These include demographic information (age, gender, income level, etc.), consumer behavior (purchase frequency, type of purchased goods, average purchase amount, etc.), and regional characteristics (infrastructure status, geographical features, number of competing companies, etc.). This extracts important data points that are then used for analysis.

[1547] 4. Model training

[1548] The server trains a generative AI model using the extracted features. This training process involves splitting the data (training set, validation set, test set). Once the training is complete, the model's performance is evaluated, and hyperparameters are adjusted as needed.

[1549] 5. Analysis and Prediction

[1550] The server uses a pre-trained model to analyze new data and predict market size and untapped market potential in each region. The prediction results are stored in a database.

[1551] 6. Emotion recognition

[1552] The device (smart glasses) uses a camera and microphone to capture the user's facial image and voice data, which are then analyzed by an emotion recognition engine. For example, it analyzes a customer's face in real time and obtains their emotion data. This emotion data is sent to a server and further incorporated into the analysis results.

[1553] 7. Visualization of Results

[1554] The server combines prediction results with user sentiment data and generates data to be visualized in an easy-to-understand format (graphs, maps, dashboards, etc.). The device (smart glasses) displays this visualized data to the user. The user can understand the analysis results while viewing graphs and maps.

[1555] 8. Emotion-based display adjustments

[1556] The server adjusts how the analysis results are displayed based on the emotional data recognized by the emotion recognition engine. For example, if the user is feeling stressed, the display may be simplified.

[1557] 9. Proposal for staffing arrangements

[1558] The server proposes an effective staffing plan based on the analysis results. This proposal takes into account regional market size, purchasing power, and even user sentiment data. The terminal (smart glasses) displays the proposed staffing plan to the user. The user can then develop an optimal staffing plan based on this.

[1559] Hardware and software to be used

[1560] Smart glasses: Recognizing customer emotions in real time (e.g., Google Glass, Vuzix Blade)

[1561] Server: Cloud infrastructure for data collection and analysis (e.g., AWS, Google Cloud)

[1562] Emotion recognition engine: (Examples: Microsoft Azure Face API, Google Cloud Vision API)

[1563] Database: (e.g., MySQL, MongoDB)

[1564] Machine learning libraries: (e.g., scikit-learn for Python, TensorFlow)

[1565] Example of a prompt:

[1566] "Using customer sentiment data and sales data, propose the optimal product placement strategy to increase store sales."

[1567] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1568] Step 1:

[1569] The server collects regional data from specified data sources and stores it in a database.

[1570] Input: Purchase data, demographic information, economic data

[1571] Data processing: Data collection and organization

[1572] Output: Data organized into a unified format is stored in the database.

[1573] Step 2:

[1574] The server performs data cleansing on the collected data to reduce unwanted noise.

[1575] Input: Original data stored in the database

[1576] Data processing: Noise reduction, detection and correction of inaccurate data.

[1577] Output: Cleansed data

[1578] Step 3:

[1579] The server will fill in any missing values ​​in the data.

[1580] Input: Cleansed data

[1581] Data processing: Imputation of missing values ​​(using mean, median, or machine learning models)

[1582] Output: Completed and cleansed data

[1583] Step 4:

[1584] The server extracts useful features from the pre-processed data.

[1585] Input: Completed and cleansed data

[1586] Data processing: Feature extraction (demographic information, consumer behavior, regional characteristics, etc.)

[1587] Output: Feature data

[1588] Step 5:

[1589] The server uses the extracted features to train a generative AI model.

[1590] Input: Feature data

[1591] Data processing: Splitting into training, validation, and test sets; model training and evaluation.

[1592] Output: Trained AI model

[1593] Step 6:

[1594] The server uses a pre-trained model to analyze new data and predict market size.

[1595] Input: Trained AI model, new data

[1596] Data processing: Analysis and prediction

[1597] Output: Forecast data (market size, potential of untapped markets)

[1598] Step 7:

[1599] The device (smart glasses) uses a camera and microphone to capture the user's facial image and voice data, which are then analyzed by an emotion recognition engine.

[1600] Input: Customer's facial image, voice data

[1601] Data processing: Analysis using an emotion recognition engine

[1602] Output: User sentiment data

[1603] Step 8:

[1604] The server combines prediction results with user sentiment data and visualizes them in an easy-to-understand format (graphs, maps, dashboards, etc.).

[1605] Input: Prediction result data, user sentiment data

[1606] Data processing: Generation of visualized data (graphs, maps, dashboards)

[1607] Output: Visualization data

[1608] Step 9:

[1609] The device (smart glasses) displays the visualized analysis results to the user and adjusts the display method as needed.

[1610] Input: Visualization data, user sentiment data

[1611] Data processing: Adjusting display methods based on emotions

[1612] Output: Adjusted display data

[1613] Step 10:

[1614] Based on the analysis results, the server proposes an effective staffing allocation.

[1615] Input: Analysis results data, user sentiment data

[1616] Data calculation: Optimization calculation of personnel allocation

[1617] Output: Personnel allocation proposal data

[1618] Step 11:

[1619] The device (smart glasses) displays the proposed staffing plan to the user.

[1620] Input: Personnel allocation proposal data

[1621] Data processing: Real-time display

[1622] Output: Staffing proposals to be reviewed by the user

[1623] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1624] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1625] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[1626] [Fourth Embodiment]

[1627] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1628] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1629] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1630] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[1631] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[1632] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[1633] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[1634] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[1635] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[1636] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1637] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1638] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[1639] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1640] This invention relates to a system for collecting and analyzing regional data to provide effective market analysis and personnel allocation suggestions. The processing of this system's program is described below in natural language.

[1641] 1. Data Collection

[1642] User: Specify the region to be analyzed and the required data types (e.g., purchasing data, demographic information, economic data).

[1643] Terminal: Collects necessary data from the specified data source and sends it to the server.

[1644] Server: Stores collected data in a central database. This ensures that all data is managed centrally.

[1645] 2. Data preprocessing

[1646] Server: Perform data cleansing to reduce unwanted noise. Detect and remove inaccurate or abnormal data.

[1647] Server: Imputes missing data. For example, it may use imputation methods such as imputing with the mean or median, or using machine learning models.

[1648] 3. Feature Extraction

[1649] Server: Extracts features from pre-processed data. Specifically, it extracts demographic information (age, gender, income level, etc.), consumer behavior (purchase frequency, type of purchased goods, average purchase amount, etc.), and regional characteristics (infrastructure status, geographical characteristics, number of competing companies, etc.).

[1650] Server: Stores the extracted features as training data.

[1651] 4. Model training

[1652] Server: Trains a generative artificial intelligence model using the extracted features. The training process includes splitting the data (training set, validation set, test set).

[1653] Server: Evaluates the trained model and adjusts hyperparameters as needed.

[1654] 5. Analysis and Prediction

[1655] Server: Uses a pre-trained model to analyze new data. This analysis predicts the market size and potential of untapped markets in each region.

[1656] Server: Organizes the analysis results and saves them to the database.

[1657] 6. Visualization of Results

[1658] Server: Generates data to visualize analysis results in an easy-to-understand format (e.g., graphs, maps, dashboards).

[1659] Terminal: Displays visualized data to the user. The user understands the specific analysis results by viewing graphs and maps.

[1660] 7. Proposal for staffing arrangements

[1661] Server: Based on the analysis results, propose effective staffing arrangements. These proposals will take into account the market size and purchasing power of each region.

[1662] Terminal: Displays the proposed staffing plan to the user. The user can then create an optimal staffing plan based on this proposal.

[1663] Specific example

[1664] For example, if a user selects Tokyo and Osaka as target regions, the terminal collects purchasing data, population information, and economic data for Tokyo and Osaka. The server stores this data in a database and performs data cleansing and feature extraction. A generative artificial intelligence model is trained using the extracted features, and new data is analyzed by the model. As a result, the market size and potential of untapped markets in Tokyo and Osaka are predicted. The server visualizes the prediction results and displays them to the user through the terminal. Based on these results, the user can plan to deploy 12 sales staff in Tokyo and 8 sales staff in Osaka. In this way, efficient market analysis and personnel allocation are achieved.

[1665] Thus, the present invention provides a consistent process from the collection and analysis of regional data to the visualization of results and the proposal of personnel allocation, resulting in significant benefits and convenience.

[1666] The following describes the processing flow.

[1667] Step 1: Data Collection

[1668] User: Specify the region to be analyzed and the required data types (e.g., purchasing data, demographic information, economic data).

[1669] Terminal: Collects data from the specified data source (e.g., government statistics database, commercial database, web API, etc.).

[1670] Terminal: Sends collected data to the server.

[1671] Server: Stores received data in a database for later storage.

[1672] Step 2: Data Preprocessing

[1673] Server: Perform data cleansing. Detect and delete or correct inaccurate or abnormal data.

[1674] Server: Performs processing to impute missing data. For example, it can impute missing data with the mean or median, or use a machine learning model to calculate predicted values.

[1675] Server: Updates the database to store the data after preprocessing is complete.

[1676] Step 3: Feature Extraction

[1677] Server: Extracts useful features from pre-processed data. Specifically, it extracts demographic information (age, gender, income level, etc.), consumer behavior (purchase frequency, type of purchased goods, average purchase amount, etc.), and regional characteristics (infrastructure status, geographical characteristics, number of competing companies, etc.).

[1678] Server: Organizes the extracted features and saves them as training data.

[1679] Step 4: Model Training

[1680] Server: Trains a generative artificial intelligence model using the extracted features. The training process includes splitting the data (training set, validation set, test set).

[1681] Server: Evaluates the model's performance and adjusts hyperparameters as needed.

[1682] Server: Stores the trained model.

[1683] Step 5: Analysis and Prediction

[1684] Server: Performs analysis on new data using the trained model.

[1685] Server: Predicts market size and untapped market potential in each region.

[1686] Server: Organizes the prediction results and saves them to the database.

[1687] Step 6: Visualizing the Results

[1688] Server: Generates data to visualize prediction results in an easy-to-understand format (e.g., graphs, maps, dashboards).

[1689] Terminal: Displays visualized data to the user. The user understands the specific analysis results by viewing graphs and maps.

[1690] Step 7: Propose staffing arrangements

[1691] Server: Based on the analysis results, proposes effective staffing arrangements. These proposals take into account regional market size and purchasing power.

[1692] Terminal: Displays the proposed staffing plan to the user. The user then develops the optimal staffing plan based on this.

[1693] For example, if a user designates Tokyo and Osaka as target regions, the terminal collects purchasing data, demographic information, and economic data for these regions. The server performs data cleansing and feature extraction, and trains a generative artificial intelligence model. The server then analyzes the new data to predict the market size and untapped markets for Tokyo and Osaka. Finally, the server visualizes the prediction results and displays them to the user through the terminal. Based on the proposed staffing plan, the user effectively allocates 12 sales staff to Tokyo and 8 to Osaka.

[1694] (Example 1)

[1695] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1696] Traditional marketing analysis methods made it difficult to collect and analyze data that differed by region and to allocate personnel effectively. Furthermore, there was no system that consistently handled data cleansing, data reconciliation, and visualization of analysis results. As a result, companies were unable to conduct efficient market analysis and optimal personnel allocation, hindering business efficiency and accurate understanding of target markets.

[1697] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1698] In this invention, the server includes means for specifying the region and data type to be analyzed; means for collecting the specified data; means for storing the collected data in a database; means for performing data cleansing on the stored data to remove inaccurate data and outliers; means for imputing missing data; means for extracting features from the data; means for training a generative artificial intelligence model using the extracted features; means for analyzing new data and predicting market size using the trained artificial intelligence model; means for storing and visualizing the prediction results in a database; means for proposing staffing arrangements based on the analysis results; and means for displaying the proposed staffing arrangements. This makes it possible to consistently perform detailed market analysis and effective staffing arrangements for each region.

[1699] "Analysis target area" refers to the specific region that the system targets when performing marketing analysis.

[1700] "Data type" refers to the various types of data necessary for analysis in the collection of regional data. Examples include purchasing data, population information, and economic data.

[1701] "Data cleansing" refers to the process of detecting inaccurate information and outliers contained in collected data, and then removing or correcting them.

[1702] "Missing data" refers to information that is missing from a dataset.

[1703] "Features" refer to data attributes or variables that have significant meaning for data analysis.

[1704] A "generative artificial intelligence model" refers to a model that uses machine learning and deep learning techniques to make predictions and classifications on new data.

[1705] "Market size" refers to the expected sales or purchasing power in a particular region or market.

[1706] "Visualization" refers to the process of visually displaying data and analysis results using graphs, dashboards, maps, and other visual tools.

[1707] "Personnel allocation" refers to the process by which a company or organization distributes the necessary number of staff members to appropriate locations and roles in order to effectively carry out its operations.

[1708] This invention relates to a system for collecting and analyzing regional data to provide effective market analysis and personnel allocation suggestions.

[1709] System program configuration

[1710] This system begins with the user specifying the analysis area and data type. The terminal collects data and sends it to the server, which processes the collected data to propose market analysis and staffing strategies. This entire process involves multiple hardware and software components.

[1711] Hardware:

[1712] Device: A computer or smartphone that can access the internet or APIs for data collection.

[1713] Server: A high-performance computer used for data processing, storage, and analysis.

[1714] software:

[1715] Database management systems: Used to centrally manage collected data. Examples include MySQL and PostgreSQL.

[1716] Data cleansing tools: These cleanse the collected data. For example, the pandas library in Python.

[1717] Generative AI models: Machine learning models that perform market analysis using features. Examples include Scikit-learn and TensorFlow.

[1718] Visualization tools: Tools for visualizing and displaying analysis results. Examples include Tableau and D3.js.

[1719] Program processing flow

[1720] The system program performs the following steps:

[1721] 1. Data collection:

[1722] User: Specify the region to be analyzed and the required data types (purchase data, demographic information, economic data, etc.).

[1723] Terminal: Collects necessary data from specified data sources (open data portals or commercial databases) and sends it to the server.

[1724] Server: Stores data sent from terminals in a central database and manages it centrally.

[1725] 2. Data preprocessing:

[1726] Server: Performs data cleansing on the collected data, removing inaccurate data and outliers.

[1727] Server: Imputes missing data using mean values ​​and machine learning models.

[1728] 3. Feature extraction:

[1729] Server: Extracts features such as demographic information, consumer behavior, and regional characteristics from pre-processed data and saves them as training datasets.

[1730] 4. Model training:

[1731] Server: Trains a generative artificial intelligence model using the extracted feature data. If necessary, splits the data into training, validation, and test sets, and tunes hyperparameters.

[1732] 5. Analysis and Prediction:

[1733] Server: Uses a pre-trained model to analyze new data and predict market size and the potential of untapped markets.

[1734] Server: Organizes the prediction results and saves them to the database.

[1735] 6. Visualization of results:

[1736] Server: Generates visualization data such as graphs and maps based on the analysis results.

[1737] Terminal: Displays dashboards and graphs to the user based on visualization data sent from the server.

[1738] 7. Proposal for staffing arrangements:

[1739] Server: Based on the analysis results, propose effective staffing arrangements.

[1740] Terminal: Displays the proposed staffing plan to the user.

[1741] Explanation of specific examples

[1742] For example, if a user selects Tokyo and Osaka as target regions, the device collects purchasing data, demographic information, and economic data for Tokyo and Osaka through open data portals and commercial databases. The collected data is sent to a server and stored in a central database. Next, the server performs data cleansing, removing inaccurate data and filling in missing data. After that, the server extracts features such as purchase frequency and average purchase amount and stores them as a training dataset. Using the stored data, a generative AI model is trained to make predictions on new data. The analysis results are displayed visually, and the server proposes a staffing plan suitable for the user. Based on this information, the user can plan, for example, to deploy 12 sales staff in Tokyo and 8 in Osaka.

[1743] Example of a prompt

[1744] "Conduct a market analysis using purchasing data, demographic information, and economic data for Tokyo and Osaka, and propose effective staffing strategies."

[1745] Thus, the present invention provides a consistent process from the collection and analysis of regional data to the visualization of results and the proposal of personnel allocation, resulting in significant benefits and convenience.

[1746] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1747] Step 1:

[1748] User: Specify the region and data type to be analyzed. Specifically, the user uses a terminal to select "Tokyo purchasing data and population information" through the interface.

[1749] Input: Region to be analyzed and data type.

[1750] Output: Field information required for the data collection process.

[1751] Step 2:

[1752] Terminal: Collects data via the internet or APIs based on the user-specified analysis area and data type. For example, it accesses "open data portals" or "commercial databases" to obtain information.

[1753] Input: Region to be analyzed and data type.

[1754] Output: Collected raw data.

[1755] Specific operation: A request is sent to the "Open Data Portal API" to retrieve purchasing data for Tokyo.

[1756] Step 3:

[1757] Terminal: Sends collected data to the server.

[1758] Input: Collected raw data.

[1759] Output: Request containing collected data.

[1760] Specific operation: The collected data is sent to the server as an HTTP request via the API.

[1761] Step 4:

[1762] Server: Stores collected data in a central database. This ensures centralized data management.

[1763] Input: Collected data sent to the server.

[1764] Output: Data stored in the central database.

[1765] Specific operation: Use a database management system (e.g., MySQL) to insert data into a table.

[1766] Step 5:

[1767] Server: Perform data cleansing to remove inaccurate data and outliers. For example, remove outliers like "-1 yen" from purchase history data.

[1768] Input: Raw data stored in the central database.

[1769] Output: Cleansed data.

[1770] Specific operation: Use the Python pandas library to filter and remove inaccurate entries in the data.

[1771] Step 6:

[1772] Server: Imputes missing data using mean values ​​or machine learning models. For example, it can imputate missing age data using the average age of other data points.

[1773] Input: Cleansed data (including missing values).

[1774] Output: Interpolated data.

[1775] Specific operation: Missing values ​​are imputed using the K-nearest neighbors algorithm with the scikit-learn library in Python.

[1776] Step 7:

[1777] Server: Extracts features from pre-processed data. Specifically, it extracts demographic information such as age and income, and consumer behavior data such as purchase frequency and purchase amount.

[1778] Input: Completed data.

[1779] Output: Feature dataset.

[1780] Specific operation: Using the Python Feature-engine library, select important features and convert the data into a matrix format.

[1781] Step 8:

[1782] Server: The extracted features are used to split the data into training, validation, and test sets, and a generative artificial intelligence model is trained. Specifically, 70% is split into a training set, 15% into a validation set, and 15% into a test set, and a regression model is trained.

[1783] Input: Feature dataset.

[1784] Output: Trained model.

[1785] Specific operation: The data is split using the scikit-learn library, and a Linear Regression model is trained.

[1786] Step 9:

[1787] Server: Uses a pre-trained model to analyze new data and predict market size by region.

[1788] Input: A trained model and new data.

[1789] Output: Prediction result.

[1790] Specific actions: Input new data into the model and perform regression analysis to predict market size.

[1791] Step 10:

[1792] Server: Organizes and stores prediction results in a database.

[1793] Input: Prediction result.

[1794] Output: Prediction results stored in the database.

[1795] Specific operation: Use an SQL query to save the prediction results to the appropriate table in the database.

[1796] Step 11:

[1797] Server: Converts analysis results into visualization data such as graphs and maps.

[1798] Input: Prediction results stored in the database.

[1799] Output: Visualized data.

[1800] Specific operation: Use Matplotlib and D3.js to convert the prediction results into a visually displayable format.

[1801] Step 12:

[1802] Terminal: Displays dashboards and graphs to the user based on visualization data sent from the server.

[1803] Input: Visualization data from the server.

[1804] Output: Graphs and maps displayed in the user interface.

[1805] Specific operation: Use a web application framework (e.g., React) to dynamically display data in the user interface.

[1806] Step 13:

[1807] Server: Based on the analysis results, it proposes effective personnel allocation for each region.

[1808] Input: Analysis results.

[1809] Output: Staffing proposal.

[1810] Specific operation: Based on the prediction results, calculate the optimal staffing allocation for each region and generate a proposal.

[1811] Step 14:

[1812] Terminal: Displays the proposed staffing plan to the user. The user can make decisions based on the proposal.

[1813] Input: Personnel allocation proposal.

[1814] Output: Staffing suggestions displayed in the user interface.

[1815] Specific operation: Use a web application framework to display the contents of the proposal in a user interface.

[1816] (Application Example 1)

[1817] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1818] Efficient market analysis and staffing planning in physical stores are not easy, and responding quickly to changes in market trends is particularly challenging. Furthermore, data inaccuracies that occur during the process of collecting and analyzing regional data, as well as a lack of means to support visual decision-making, are also challenges.

[1819] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1820] In this invention, the server includes means for collecting regional data, means for storing the collected data in a database, means for performing data cleansing on the stored data, means for extracting features from the data, means for training a generative artificial intelligence model using the extracted features, means for analyzing new data using the trained model and predicting market size, means for visualizing and displaying the prediction results, means for proposing staffing arrangements based on the analysis results, means for notifying changes in market trends for a specific region in real time, and means for displaying the prediction results on an electronic map to support visual judgments for store placement and market analysis. This enables efficient market analysis and staffing planning, and allows for quick and accurate responses to changes in market trends.

[1821] "Regional data" refers to data that includes population information, economic data, purchasing data, and related infrastructure information for a specific geographical area.

[1822] A "database" is a system for centrally managing and storing collected data.

[1823] "Data cleansing" is a technique that improves the quality of collected data by removing noise and inaccurate data.

[1824] "Features" refer to statistically extracted information necessary for data analysis, and are primarily used as input for analytical models.

[1825] A "generative artificial intelligence model" refers to a machine learning algorithm that performs predictions and analyses based on collected and processed data.

[1826] "Prediction" refers to estimating future data and trends using a trained artificial intelligence model.

[1827] "Visualization" refers to displaying analysis results in a visually easy-to-understand format, such as graphs, maps, and dashboards.

[1828] "Personnel allocation" refers to planning the placement of personnel in the most suitable locations and positions.

[1829] "Real-time notifications" is a feature that instantly informs users of market trends and other data changes.

[1830] An "electronic map" is a digital map that visually displays geographical information and forecast results.

[1831] This invention relates to a system for collecting and analyzing regional data to perform market analysis and propose personnel allocation strategies. The system functions through the cooperation of a server, terminals, and users.

[1832] 1. Data Collection

[1833] The user specifies the region to be analyzed and the required data types (e.g., purchasing data, demographic information, economic data). The terminal collects the necessary data from the specified data sources and sends it to the server. The server stores the collected data in a central database and manages all data centrally.

[1834] 2. Data preprocessing

[1835] The server performs data cleansing to reduce unnecessary noise. Specifically, it detects and removes inaccurate and anomalous data. It also uses methods such as imputation with the mean or median, or imputation using machine learning models, to fill in missing data.

[1836] 3. Feature Extraction

[1837] The server extracts features from the pre-processed data. Specifically, it extracts demographic information (age, gender, income level, etc.) and consumer behavior (purchase frequency, type of purchased goods, average purchase amount, etc.). It also includes regional characteristics (infrastructure status, geographical features, number of competing companies, etc.). The extracted features are stored as training data.

[1838] 4. Model training

[1839] The server uses the extracted features to train a generative artificial intelligence model. The training process includes splitting the data (training set, validation set, test set). The trained model is evaluated, and hyperparameters are adjusted as needed.

[1840] 5. Analysis and Prediction

[1841] The server uses a pre-trained model to analyze new data. This analysis predicts the market size and potential of untapped markets in each region. The analysis results are stored in a database for reuse.

[1842] 6. Visualization of Results

[1843] The server generates data that visualizes the analysis results in an easy-to-understand format (e.g., graphs, maps, dashboards). The terminal displays this visualized data to the user. This allows the user to understand the specific analysis results while viewing graphs and maps.

[1844] 7. Proposal for staffing arrangements

[1845] Based on the analysis results, the server proposes an effective staffing plan. This proposal takes into account the market size and purchasing power of each region. The terminal displays the proposed staffing plan to the user, helping them to create the optimal staffing plan.

[1846] It also includes a feature that provides real-time notifications of changes in market trends in specific regions. This feature provides alerts to quickly respond to sudden market fluctuations. Furthermore, it can display forecast results on an electronic map, visually supporting store placement and market analysis. This allows users to make more intuitive decisions.

[1847] Specific example

[1848] For example, suppose a user collects purchasing data, population data, and competitor information based on location data in Tokyo to perform market analysis. In this case, the server collects and preprocesses this data, extracts features, and trains an AI model. This model predicts the market size of Tokyo, and the results are visually presented to the user in the form of graphs and maps. Then, the optimal staffing allocation for stores in Tokyo is suggested.

[1849] Examples of prompt statements are as follows:

[1850] "Based on location data for Tokyo, collect in-store purchasing data, population information, and competitor information to conduct market analysis. Then, create an application that proposes appropriate staffing arrangements based on the results."

[1851] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1852] Step 1:

[1853] Data collection

[1854] The user specifies the target region and the required data type (purchase data, population information, economic data, etc.). The terminal collects the necessary data from the specified data source (public API, website, statistical database, etc.) and sends the collected data to the server. The server stores the received data in a central database.

[1855] Input: User-specified region and data type

[1856] Output: Regional data stored on the server

[1857] Step 2:

[1858] Data preprocessing

[1859] The server performs data cleansing on the received data. Specifically, it normalizes unstructured data, reduces noisy data, and detects and removes inaccurate or outlier values. Next, to impute missing data, it uses the mean or median, or applies imputation methods using machine learning models.

[1860] Input: Raw data stored in the central database

[1861] Output: Cleansed and imputed data

[1862] Step 3:

[1863] Feature extraction

[1864] The server extracts features from the pre-processed data. Specifically, it extracts demographic information (age, gender, income level, etc.), consumer behavior (purchase frequency, type of purchased goods, average purchase amount, etc.), and regional characteristics (infrastructure status, geographical characteristics, number of competing stores, etc.), and stores them as training data.

[1865] Input: Pre-processed data

[1866] Output: Training data with extracted features

[1867] Step 4:

[1868] Model training

[1869] The server uses the extracted features to train a generative artificial intelligence model. This process involves splitting the training data into training, validation, and test sets, selecting appropriate hyperparameters, and optimizing the model. The trained model is then used for subsequent analysis and prediction.

[1870] Input: Training data with extracted features

[1871] Output: Trained generative artificial intelligence model

[1872] Step 5:

[1873] Analysis and prediction

[1874] The server uses a pre-trained generative artificial intelligence model to analyze new data. This analysis predicts the market size and potential of untapped markets in each region, and these predictions are stored in a database.

[1875] Input: A trained generative AI model, and new data to be analyzed.

[1876] Output: Prediction results stored in the database

[1877] Step 6:

[1878] Visualization of results

[1879] The server generates data that visualizes the prediction results in formats such as graphs, maps, and dashboards. The terminal displays this visualized data to the user, allowing the user to visually confirm the analysis results.

[1880] Input: Prediction results stored in the database

[1881] Output: Visualized analysis results

[1882] Step 7:

[1883] Proposal for staffing arrangements

[1884] The server proposes an effective staffing plan based on the analysis results. The proposal takes into account the market size and purchasing power of each region. The terminal displays the proposed staffing plan to the user, allowing the user to create the optimal staffing plan.

[1885] Input: Visualized analysis results

[1886] Output: Proposed staffing arrangements

[1887] Step 8:

[1888] Real-time notifications

[1889] The server monitors market trends in a specific region in real time and notifies the user when a change is detected. This allows the user to respond quickly and accurately to market changes.

[1890] Input: Real-time monitored market data

[1891] Output: Notification of market changes to users

[1892] Step 9:

[1893] Use of electronic maps

[1894] The server generates data to display prediction results on an electronic map, and the terminal displays this to the user. This allows the user to visually perform store placement and market analysis.

[1895] Input: Prediction result

[1896] Output: Prediction results displayed on the electronic map

[1897] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1898] This invention combines a system that collects and analyzes regional data to perform effective market analysis and personnel allocation proposals with an emotion engine that recognizes user emotions. The processing of this system's program is described below in natural language.

[1899] 1. Data Collection

[1900] User: Specify the region to be analyzed and the required data types (e.g., purchasing data, demographic information, economic data).

[1901] Terminal: Collects necessary data from the specified data source and sends it to the server.

[1902] Server: Receives data and stores it in a central database. This ensures that all data is managed centrally.

[1903] 2. Data preprocessing

[1904] Server: Perform data cleansing to reduce unnecessary noise. Detect and delete or correct inaccurate or abnormal data.

[1905] Server: Performs processing to impute missing data. For example, it may use imputation methods such as imputation using the mean or median, or imputation using machine learning models.

[1906] Server: Updates the database to store the data after preprocessing is complete.

[1907] 3. Feature Extraction

[1908] Server: Extracts useful features from pre-processed data. Specifically, it extracts demographic information (age, gender, income level, etc.), consumer behavior (purchase frequency, type of purchased goods, average purchase amount, etc.), and regional characteristics (infrastructure status, geographical characteristics, number of competing companies, etc.).

[1909] Server: Organizes the extracted features and saves them as training data.

[1910] 4. Model training

[1911] Server: Trains a generative artificial intelligence model using the extracted features. The training process includes splitting the data (training set, validation set, test set).

[1912] Server: Evaluates the model's performance and adjusts hyperparameters as needed.

[1913] Server: Stores the trained model.

[1914] 5. Analysis and Prediction

[1915] Server: Performs analysis on new data using the trained model.

[1916] Server: Predicts market size and untapped market potential in each region.

[1917] Server: Organizes the prediction results and saves them to the database.

[1918] 6. Emotion recognition

[1919] Device: Recognizes the user's emotions using an emotion engine. For example, it analyzes emotions from facial expressions and voice using a camera and microphone.

[1920] Server: Stores recognized emotion data in a database and performs necessary processing to incorporate it into the analysis.

[1921] 7. Visualization of Results

[1922] Server: Combines prediction results and user sentiment data to generate data that can be visualized in an easy-to-understand format (e.g., graphs, maps, dashboards).

[1923] Terminal: Displays visualized data to the user. The user understands the specific analysis results by viewing graphs and maps.

[1924] 8. Emotion-based display adjustments

[1925] Server: Based on the emotion data recognized by the emotion engine, the server adjusts how the analysis results are displayed. For example, if the user is feeling stressed, the display is simplified.

[1926] 9. Proposal for staffing arrangements

[1927] Server: Based on the analysis results, propose effective staffing arrangements. These proposals will take into account regional market size, purchasing power, and user sentiment data.

[1928] Terminal: Displays the proposed staffing plan to the user. The user then develops the optimal staffing plan based on this.

[1929] Specific example

[1930] For example, if a user specifies Tokyo and Osaka as target regions, the device collects purchasing data, demographic information, and economic data for these regions. The server stores this data in a database, performs data cleansing and feature extraction, and trains a generative artificial intelligence model. The server then analyzes the new data to predict market size and untapped markets in Tokyo and Osaka. Along with the analysis results, the device recognizes the user's emotions using an emotion engine, and the server integrates this data and visualizes it appropriately. If the user is not feeling stressed, a detailed graphical interface is displayed; if they are feeling stressed, it switches to a simpler display. Finally, the server proposes a staffing plan and displays it to the user through the device. The user can then plan to effectively allocate 12 sales staff in Tokyo and 8 in Osaka.

[1931] Thus, the present invention provides a consistent series of processes, from the collection and analysis of regional data to sentiment recognition, visualization of results, and proposal of personnel allocation, and by taking user sentiment into consideration, it brings even greater convenience and accuracy.

[1932] The following describes the processing flow.

[1933] Step 1: Data Collection

[1934] User: Specify the region to be analyzed (e.g., Tokyo, Osaka) and the required data types (purchase data, demographic information, economic data).

[1935] Terminal: Collects necessary data from specified data sources (government statistics databases, commercial databases, web APIs, etc.) and sends it to the server.

[1936] Server: Receives data and stores it in a central database. This ensures that all data is managed centrally.

[1937] Step 2: Data Preprocessing

[1938] Server: Perform data cleansing to detect, delete, or correct inaccurate or abnormal data. For example, delete outliers in purchase data.

[1939] Server: Imputate missing data. Imputate missing data with the mean or median, or use a machine learning model to calculate predicted values.

[1940] Server: Saves the pre-processed data back to the database.

[1941] Step 3: Feature Extraction

[1942] Server: Extracts features from pre-processed data. Specifically, it extracts demographic information (age, gender, income level, etc.), consumer behavior (purchase frequency, type of purchased goods, average purchase amount, etc.), and regional characteristics (infrastructure status, geographical characteristics, number of competing companies, etc.).

[1943] Server: Organizes the extracted features and saves them as training data.

[1944] Step 4: Model Training

[1945] Server: Trains a generative artificial intelligence model using the extracted features. This process involves splitting the data into a training set, a validation set, and a test set.

[1946] Server: Evaluates the model's performance and adjusts hyperparameters as needed.

[1947] Server: Stores the trained model.

[1948] Step 5: Analysis and Prediction

[1949] Server: Uses a pre-trained model to analyze new data. The analysis predicts market size and untapped market potential in each region.

[1950] Server: Organizes the prediction results and saves them to the database.

[1951] Step 6: Emotion Recognition

[1952] Device: Recognizes the user's emotions using an emotion engine. Analyzes emotions from facial expressions and voice using the camera and microphone.

[1953] Server: Stores recognized emotion data in a database and performs necessary processing to incorporate it into the analysis.

[1954] Step 7: Visualizing the Results

[1955] Server: Combines prediction results and user sentiment data to generate data that can be visualized in an easy-to-understand format (e.g., graphs, maps, dashboards).

[1956] Terminal: Displays visualized data to the user. The user understands the specific analysis results by viewing graphs and maps.

[1957] Step 8: Emotion-based display adjustments

[1958] Server: Based on the emotion data recognized by the emotion engine, adjusts how the analysis results are displayed. For example, if the user is feeling stressed, simplify the display.

[1959] Step 9: Propose staffing arrangements

[1960] Server: Based on the analysis results, propose effective staffing arrangements. These proposals will take into account regional market size, purchasing power, and user sentiment data.

[1961] Terminal: Displays the proposed staffing plan to the user. The user then develops the optimal staffing plan based on this.

[1962] For example, if a user designates Tokyo and Osaka as target regions, the device collects purchasing data, demographic information, and economic data for these areas. The server stores this data in a database, performs data cleansing and feature extraction, and trains a generative artificial intelligence model. The server then analyzes the new data to predict market size and untapped markets in Tokyo and Osaka. Along with the analysis results, the device recognizes the user's emotions using an emotion engine, and the server integrates this data and visualizes it appropriately. If the user is not feeling stressed, a detailed graphical interface is displayed; if they are stressed, it switches to a simpler display. Finally, the server proposes a staffing plan and displays it to the user through the device. The user can then plan to effectively allocate 12 sales staff to Tokyo and 8 to Osaka.

[1963] (Example 2)

[1964] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1965] In modern society, effectively collecting and analyzing regional data and conducting market analysis is crucial for enhancing a company's competitiveness. However, current systems fragment the entire process from data collection to analysis and result display, making it difficult to perform efficiently. Furthermore, there is a lack of systems that provide advanced features such as displaying analysis results while considering user emotions and suggesting personnel allocation. This can lead to user stress and a risk of decreased business efficiency due to inability to implement optimal personnel allocation.

[1966] In Example 2, the identification processing by the identification processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for collecting regional data, means for storing the collected data in a database, means for performing data cleansing on the stored data, means for imputing missing data, means for extracting features from the data, means for training a generative artificial intelligence model using the extracted features, means for analyzing new data using the trained model and predicting market size, means for recognizing user emotions, means for reflecting the recognized emotion data in the analysis, means for visualizing and displaying the prediction results, means for adjusting the display method based on the emotion data, and means for proposing personnel allocation based on the analysis results. This enables the consistent execution of a series of processes, flexible display of analysis results in accordance with user emotions, and effective proposal of personnel allocation.

[1967] "Regional data" refers to various types of information related to a specific region, and specifically includes purchasing data, population information, and economic data.

[1968] A "database" is a system for centrally managing and storing collected data; examples include MySQL and PostgreSQL.

[1969] "Data cleansing" is the process of detecting, correcting, and removing inaccurate data and noise in order to improve the quality of the data.

[1970] "Missing data" refers to values ​​that are missing from a dataset, and a means of filling in these missing values ​​is required.

[1971] "Features" are elements of data extracted for use in analysis and model training, and specific examples include demographic information and consumer behavior.

[1972] A "generative artificial intelligence model" is a machine learning model that is trained on collected and pre-processed data to perform predictions and analyses.

[1973] "Emotion recognition" is a technology that detects a user's emotional state, using an emotion engine to analyze emotions from facial expressions and voice.

[1974] "Visualization" refers to displaying analysis results in a format that is easy for users to understand, and graphs and maps are commonly used for this purpose.

[1975] "Staffing" refers to a plan to effectively allocate the optimal number of staff based on analysis results.

[1976] The "display method" refers to the interface used to present analysis results to the user, and it is adjusted according to the user's emotions.

[1977] This invention is a system that collects and analyzes regional data to perform market analysis and propose personnel allocation, and in particular incorporates an emotion engine that recognizes user emotions. The following hardware and software are used in the implementation of this system.

[1978] The server is primarily responsible for data storage, preprocessing, feature extraction, model training, analysis, and visualization of results. The following software should be installed on the server:

[1979] Database management systems (e.g., MySQL, PostgreSQL)

[1980] Data analysis libraries (e.g., Pandas, NumPy)

[1981] Machine learning libraries (e.g., scikit-learn, TensorFlow, PyTorch)

[1982] Visualization libraries (e.g., Matplotlib, Plotly)

[1983] The device is responsible for data collection, emotion recognition, and partial data processing. Specifically, it is connected to a camera and microphone to recognize the user's emotions using an emotion engine. Affectiva's SDK is a suitable emotion engine to use.

[1984] The following is an example of the specific processing performed by this system.

[1985] Specific example

[1986] For example, suppose a user specifies Tokyo and Osaka as target regions. In this case,

[1987] 1. The device collects purchasing data, demographic information, and economic data related to these regions from APIs and websites. The device then sends the collected data to the server.

[1988] 2. The server saves the received data to the database. MySQL is used for database management.

[1989] 3. The server performs data cleansing, removing NaN values ​​and duplicate data. The Pandas library is used for this process.

[1990] 4. The server imputes missing data. For example, it can use scikit-learn's Inputter to impute missing values ​​with the mean.

[1991] 5. The server extracts useful features from the pre-processed data. These features include demographic information (age, gender, income level) and consumer behavior (purchase frequency, type of purchased goods, average purchase amount).

[1992] 6. Save the features as training data and train a generative artificial intelligence model. TensorFlow will be used to train the model.

[1993] 7. The server analyzes new data and predicts market size and untapped market potential in Tokyo and Osaka.

[1994] 8. The device uses an emotion engine to recognize the user's emotions. For example, it uses a camera to capture facial expressions and analyzes the emotional state.

[1995] 9. The server saves the prediction results and recognized emotion data to a database and incorporates them into the analysis results.

[1996] 10. The server visualizes the prediction results in the form of graphs or maps. For example, it can generate interactive graphs using the Plotly library.

[1997] 11. The device displays visualized data to the user. If the user is experiencing stress, measures are taken to simplify the display.

[1998] 12. Based on the analysis results, the server proposes an effective staffing plan. For example, it might plan to allocate 12 sales staff to Tokyo and 8 to Osaka.

[1999] 13. The terminal displays the proposed staffing plan to the user. The user can then use this to determine the optimal staffing arrangement.

[2000] This system uses a generative artificial intelligence model to perform market analysis while simultaneously incorporating user sentiment data to propose more appropriate information presentations and personnel allocation plans to users.

[2001] The flow of the specific processing in Example 2 will be explained using Figure 13.

[2002] Step 1:

[2003] On the system's administration screen, users specify the target region for analysis (e.g., Tokyo, Osaka) and the required data types (purchase data, demographic information, economic data). Based on this, a request for data collection is created.

[2004] Input: Target region and data type

[2005] Output: Data collection request

[2006] Step 2:

[2007] The device collects the necessary data from a specified data source (e.g., an API, website, database, etc.). Specifically, it uses a Python script to issue an HTTP request to the API and sends the retrieved data to the server in JSON format.

[2008] Input: Data collection request

[2009] Output: Collected data (JSON format)

[2010] Step 3:

[2011] The server stores the received data in a central database. A DBMS such as MySQL is used for this purpose. Specifically, data is registered in the database by issuing SQL INSERT statements.

[2012] Input: Collected data (JSON format)

[2013] Output: Data stored in the database

[2014] Step 4:

[2015] The server performs data cleansing. For example, it processes the dataframe using the Pandas library to remove NaN values ​​and duplicate data. It detects and corrects inaccurate data and noise to generate a clean dataset.

[2016] Input: Data retrieved from the database

[2017] Output: Cleansed data

[2018] Step 5:

[2019] The server will impute the missing data. Specifically, it will use scikit-learn's Inputter to impute missing values ​​with the mean or median, or it will use a machine learning model to perform the imputation.

[2020] Input: Cleansed data

[2021] Output: Interpolated data

[2022] Step 6:

[2023] The server extracts useful features from pre-processed data. Specifically, it extracts features based on demographic information, consumer behavior, and regional characteristics using Pandas or NumPy.

[2024] Input: Completed data

[2025] Output: Feature dataset

[2026] Step 7:

[2027] The server uses the extracted features to train a generative artificial intelligence model. Specifically, it splits the data into a training set, a validation set, and a test set, and trains the model using TensorFlow.

[2028] Input: Feature dataset

[2029] Output: Trained model

[2030] Step 8:

[2031] The server uses a pre-trained model to perform analysis on new data. Specifically, it inputs new data into the model and predicts market size.

[2032] Input: New data, trained model

[2033] Output: Forecast results (market size, potential)

[2034] Step 9:

[2035] The device uses an emotion engine to recognize the user's emotions. For example, it uses a camera to capture facial expressions and performs emotion analysis.

[2036] Input: User's facial expression, emotion engine

[2037] Output: Recognized emotion data

[2038] Step 10:

[2039] The server stores the recognized emotion data in a database and performs the necessary processing to incorporate it into the analysis. Specifically, the emotion data is incorporated into the analysis algorithm.

[2040] Input: Prediction results, sentiment data

[2041] Output: Emotion-reflected analysis results

[2042] Step 11:

[2043] The server visualizes the sentiment-infused analysis results in formats such as graphs, maps, and dashboards. For example, it can generate interactive graphs using the Plotly library.

[2044] Input: Sentiment-reflected analysis results

[2045] Output: Visualization data (graphs, maps)

[2046] Step 12:

[2047] The device displays visualized data to the user. If the user is experiencing stress, measures such as simplifying the display will be taken.

[2048] Input: Visualization data, user emotional state

[2049] Output: Screen displayed to the user

[2050] Step 13:

[2051] The server analyzes the data and proposes an effective staffing plan. For example, it might plan to deploy 12 sales staff in Tokyo and 8 in Osaka.

[2052] Input: Visualization data, analysis results

[2053] Output: Staffing proposal

[2054] Step 14:

[2055] The terminal displays the proposed staffing plan to the user. The user can then use this to determine the optimal staffing arrangement.

[2056] Input: Personnel allocation proposal

[2057] Output: Final screen displayed to the user

[2058] (Application Example 2)

[2059] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[2060] Conventional market analysis and staffing suggestion systems analyze data based solely on regional and purchasing data, failing to consider user emotions. This can lead to stress and confusion for users when receiving analysis results. Furthermore, there was no system that could acquire real-time emotional data and immediately suggest staffing and product placement based on it. To address these challenges, a system is needed that recognizes user emotions and adjusts the display of analysis results accordingly.

[2061] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for collecting regional data, means for storing the collected data in a database, means for performing data cleansing on the stored data, means for extracting features from the data, means for training a generative artificial intelligence model using the extracted features, means for analyzing new data using the trained model and predicting market size, means for visualizing and displaying the prediction results, means for proposing personnel allocation based on the analysis results, means for recognizing the user's emotions in real time, means for visualizing the analysis results including the recognized emotion data and adjusting the display method, and means for presenting the results to the user in real time using a device such as smart glasses. This makes it possible to present analysis results that take into account the user's emotions and to propose appropriate personnel allocation and product placement based on them.

[2062] "Regional data" is a general term for information including population data, economic data, and purchasing data for a specific geographical area.

[2063] A "database" is a digital system that centrally manages collected and stored data, enabling efficient searching and utilization.

[2064] "Data cleansing" is a processing method used to reduce inaccuracies and noise in data and prepare it for analysis.

[2065] "Features" are variables that represent important patterns or properties in data analysis and machine learning model training.

[2066] A "generative artificial intelligence model" is a type of machine learning model that generates useful information from data and makes predictions and decisions by learning patterns.

[2067] "Market size" refers to an indicator that shows the size of a market, such as the number of consumers or purchasing power in a particular region or market.

[2068] "Emotional data" refers to emotional information extracted from human facial expressions and voices collected using cameras and microphones.

[2069] "Smart glasses" are a type of wearable device, specifically glasses-shaped equipment that incorporates displays and sensors to show information in real time.

[2070] A "user" is a person or organization that uses this system and is the recipient of the analysis results and suggestions.

[2071] "Personnel allocation" refers to the plan or method of appropriately allocating staff or workers for the purpose of efficient business operations and service provision.

[2072] This invention relates to a system for collecting, analyzing, recognizing emotions in regional data, visualizing the results, and proposing personnel allocation. A key feature of this system is that it uses devices such as smart glasses to recognize the user's emotions in real time and reflects this in the analysis results and personnel allocation proposals.

[2073] 1. Data Collection

[2074] The server collects regional data from specified data sources and stores it in a database. This data includes diverse information such as purchasing data, population information, and economic data. This data collection allows the system to understand the market characteristics of each region.

[2075] 2. Data preprocessing

[2076] The server performs data cleansing on the collected data to reduce unwanted noise. Furthermore, it detects and removes or corrects inaccurate or anomalous data. Missing data is imputed using the mean or median, or through machine learning models.

[2077] 3. Feature Extraction

[2078] The server extracts useful features from the pre-processed data. These include demographic information (age, gender, income level, etc.), consumer behavior (purchase frequency, type of purchased goods, average purchase amount, etc.), and regional characteristics (infrastructure status, geographical features, number of competing companies, etc.). This extracts important data points that are then used for analysis.

[2079] 4. Model training

[2080] The server trains a generative AI model using the extracted features. This training process involves splitting the data (training set, validation set, test set). Once the training is complete, the model's performance is evaluated, and hyperparameters are adjusted as needed.

[2081] 5. Analysis and Prediction

[2082] The server uses a pre-trained model to analyze new data and predict market size and untapped market potential in each region. The prediction results are stored in a database.

[2083] 6. Emotion recognition

[2084] The device (smart glasses) uses a camera and microphone to capture the user's facial image and voice data, which are then analyzed by an emotion recognition engine. For example, it analyzes a customer's face in real time and obtains their emotion data. This emotion data is sent to a server and further incorporated into the analysis results.

[2085] 7. Visualization of Results

[2086] The server combines prediction results with user sentiment data and generates data to be visualized in an easy-to-understand format (graphs, maps, dashboards, etc.). The device (smart glasses) displays this visualized data to the user. The user can understand the analysis results while viewing graphs and maps.

[2087] 8. Emotion-based display adjustments

[2088] The server adjusts how the analysis results are displayed based on the emotional data recognized by the emotion recognition engine. For example, if the user is feeling stressed, the display may be simplified.

[2089] 9. Proposal for staffing arrangements

[2090] The server proposes an effective staffing plan based on the analysis results. This proposal takes into account regional market size, purchasing power, and even user sentiment data. The terminal (smart glasses) displays the proposed staffing plan to the user. The user can then develop an optimal staffing plan based on this.

[2091] Hardware and software to be used

[2092] Smart glasses: Recognizing customer emotions in real time (e.g., Google Glass, Vuzix Blade)

[2093] Server: Cloud infrastructure for data collection and analysis (e.g., AWS, Google Cloud)

[2094] Emotion recognition engine: (Examples: Microsoft Azure Face API, Google Cloud Vision API)

[2095] Database: (e.g., MySQL, MongoDB)

[2096] Machine learning libraries: (e.g., scikit-learn for Python, TensorFlow)

[2097] Example of a prompt:

[2098] "Using customer sentiment data and sales data, propose the optimal product placement strategy to increase store sales."

[2099] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[2100] Step 1:

[2101] The server collects regional data from specified data sources and stores it in a database.

[2102] Input: Purchase data, demographic information, economic data

[2103] Data processing: Data collection and organization

[2104] Output: Data organized into a unified format is stored in the database.

[2105] Step 2:

[2106] The server performs data cleansing on the collected data to reduce unwanted noise.

[2107] Input: Original data stored in the database

[2108] Data processing: Noise reduction, detection and correction of inaccurate data.

[2109] Output: Cleansed data

[2110] Step 3:

[2111] The server will fill in any missing values ​​in the data.

[2112] Input: Cleansed data

[2113] Data processing: Imputation of missing values ​​(using mean, median, or machine learning models)

[2114] Output: Completed and cleansed data

[2115] Step 4:

[2116] The server extracts useful features from the pre-processed data.

[2117] Input: Completed and cleansed data

[2118] Data processing: Feature extraction (demographic information, consumer behavior, regional characteristics, etc.)

[2119] Output: Feature data

[2120] Step 5:

[2121] The server uses the extracted features to train a generative AI model.

[2122] Input: Feature data

[2123] Data processing: Splitting into training, validation, and test sets; model training and evaluation.

[2124] Output: Trained AI model

[2125] Step 6:

[2126] The server uses a pre-trained model to analyze new data and predict market size.

[2127] Input: Trained AI model, new data

[2128] Data processing: Analysis and prediction

[2129] Output: Forecast data (market size, potential of untapped markets)

[2130] Step 7:

[2131] The device (smart glasses) uses a camera and microphone to capture the user's facial image and voice data, which are then analyzed by an emotion recognition engine.

[2132] Input: Customer's facial image, voice data

[2133] Data processing: Analysis using an emotion recognition engine

[2134] Output: User sentiment data

[2135] Step 8:

[2136] The server combines prediction results with user sentiment data and visualizes them in an easy-to-understand format (graphs, maps, dashboards, etc.).

[2137] Input: Prediction result data, user sentiment data

[2138] Data processing: Generation of visualized data (graphs, maps, dashboards)

[2139] Output: Visualization data

[2140] Step 9:

[2141] The device (smart glasses) displays the visualized analysis results to the user and adjusts the display method as needed.

[2142] Input: Visualization data, user sentiment data

[2143] Data processing: Adjusting display methods based on emotions

[2144] Output: Adjusted display data

[2145] Step 10:

[2146] Based on the analysis results, the server proposes an effective staffing allocation.

[2147] Input: Analysis results data, user sentiment data

[2148] Data calculation: Optimization calculation of personnel allocation

[2149] Output: Personnel allocation proposal data

[2150] Step 11:

[2151] The device (smart glasses) displays the proposed staffing plan to the user.

[2152] Input: Personnel allocation proposal data

[2153] Data processing: Real-time display

[2154] Output: Staffing proposals to be reviewed by the user

[2155] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[2156] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[2157] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[2158] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[2159] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[2160] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[2161] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[2162] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[2163] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[2164] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[2165] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[2166] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[2167] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[2168] 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.

[2169] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[2170] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[2171] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[2172] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a confi...

Claims

1. Means of collecting regional data, A means of storing the collected data in a database, A means of performing data cleansing on stored data, A means of extracting features from data, A method for training a generative artificial intelligence model using extracted features, A means of analyzing new data using a pre-trained model and predicting market size, A means of visualizing and displaying the prediction results, A means of proposing personnel allocation based on analysis results, A system that includes this.

2. The system according to claim 1, further comprising means for collecting purchasing data, population information, and economic data in a specific region.

3. The system according to claim 1, further comprising means for imputing missing values ​​in the data.

Images