System

Abstract

A system is provided.SOLUTION: The system includes a means for receiving and analyzing real consumption behavior data, a means for generating a virtual consumer model on the basis of the analyzed data, a means for integrating the generated virtual consumer model and constructing a virtual market, a means for simulating a business scenario in the constructed virtual market, and a means for outputting a simulation result.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] When modern companies make market forecasts and business plans, they face the problem of difficulty using traditional forecasting methods to accurately predict qualitative changes and the effects of new initiatives. Rapidly changing market environments require more accurate, microscopic forecasts, but current technology lacks the means to do so effectively. Furthermore, while the global market requires supply-demand adjustments and market forecasts that transcend the boundaries of individual companies, there are currently no appropriate tools to achieve this. [Means for solving the problem]

[0005] The present invention introduces a means for receiving and analyzing real consumer behavior data and a means for generating a virtual consumer model based on the analyzed data. It also provides a means for integrating the generated virtual consumer model to construct a virtual market. This makes it possible to simulate multiple business scenarios within the constructed virtual market, and also includes a means for outputting the simulation results. As a result, companies can predict qualitative changes and the effects of new initiatives with high accuracy, allowing them to flexibly respond to fluctuating market environments. Furthermore, by using real consumer behavior data obtained from a customer relationship management system, more realistic predictions are possible, allowing for effective simulation of scenarios for new product introductions, price changes, and advertising campaigns.

[0006] "Actual consumer behavior data" refers to actual consumer purchasing history and customer information obtained from a company's customer relationship management system or marketing database.

[0007] "Means for analysis" refers to the algorithms and software used to analyze the received data and extract consumer behavior patterns and characteristics.

[0008] "Virtual consumer model" refers to a virtual consumer profile based on predicted consumption behavior, generated from analyzed data.

[0009] A "virtual market" refers to a virtual space constructed by integrating multiple virtual consumer models to simulate a realistic market environment.

[0010] A "business scenario" refers to a specific marketing measure or business plan that a company plans, such as introducing a new product, changing prices, or conducting an advertising campaign.

[0011] "Simulation" refers to the process of running business scenarios within a virtual market and predicting the resulting behavior of virtual consumers and overall market fluctuations.

[0012] "Output means" refers to software or functionality for displaying the results of the simulation in a user interface or report format.

[0013] A "customer relationship management system" refers to a software system for managing customer information and efficiently building relationships with customers.

[0014] "New product introduction" refers to a marketing strategy that brings a new product to the market.

[0015] "Price change" refers to a marketing strategy that changes the price of an existing product or service.

[0016] An "advertising campaign" refers to promotional activities aimed at increasing consumer awareness of and demand for a particular product or service. [Brief explanation of the drawings]

[0017] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9]1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0018] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0019] First, the terms used in the following description will be explained.

[0020] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0021] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0022] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0023] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0024] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0025] [First embodiment]

[0026] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0027] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0028] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0029] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0030] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0031] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0032] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0033] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0034] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0035] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0036] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0037] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0038] The system according to the present invention performs a series of operations: it receives and analyzes real-world consumer behavior data, generates virtual consumer models, integrates them to build a virtual market, simulates business scenarios within the built virtual market, and outputs the results. Below, we will create a program for the system and explain its processing in detail.

[0039] Data entry and submission

[0040] 1. Users collect consumer data

[0041] Users collect consumer purchasing history, customer profiles, product information, etc. from a company's customer relationship management (CRM) system or marketing database. This data collection is done by extracting data from CSV files or directly from the database.

[0042] 2. The device displays the data entry screen.

[0043] The terminal provides a user interface and displays a form for the user to review and enter the collected consumer data, including information such as consumer ID, purchased product, purchase date, payment amount, age, gender, and region.

[0044] 3. The user enters and submits consumer data

[0045] The user enters the required information into the data input form on the terminal and clicks the submit button. The terminal converts the entered data into JSON format and sends it to the server.

[0046] Generating the Consumer Model

[0047] 4. The server receives and analyzes the data

[0048] The server receives the JSON data sent from the terminal. Specifically, it is designed to receive the data as an HTTP request so that it can reach the server.

[0049] The server analyzes the received data and performs data cleansing and preprocessing, which includes imputing missing data, handling outliers, and normalizing the data.

[0050] 5. The server generates a virtual consumer model

[0051] The server generates a virtual consumer model based on the analyzed data. Here, machine learning algorithms (e.g., clustering and regression models) are used to create a model that predicts consumer purchasing behavior. The generated virtual consumer model includes the consumer's purchasing history, interests, and future purchase predictions.

[0052] Building a virtual market

[0053] 6. The server aggregates multiple virtual consumer models

[0054] The server integrates the generated virtual consumer models to create a virtual market, which mimics a real-world market environment and includes information on the supply and demand of each product or service, as well as information on competitors.

[0055] Business scenario simulation

[0056] 7. The user sets up the business scenario

[0057] The user sets up a business scenario on the device. For example, a scenario could be "Introduce new product B on June 1, 2023 at a price of 2,000 yen." The user enters detailed information about the scenario and sends it to the server.

[0058] 8. The server receives the scenario and runs the simulation.

[0059] The server receives the scenario and runs a simulation in the virtual market based on that scenario, predicting things like sales fluctuations due to new product introductions, demand changes due to price changes, and the effects of advertising campaigns.

[0060] The simulation results include sales forecasts, profit margins, and changes in market share.

[0061] Output of simulation results

[0062] 9. The server generates the simulation results

[0063] The server generates the simulation results in JSON format, which include statistics on sales, profits, and market share after the scenario is executed.

[0064] 10. The device receives and displays the results

[0065] The terminal visualizes the simulation results received from the server and provides them to the user. The results are displayed in graphs and tables, and the user can use them to review strategies and make decisions.

[0066] Specific examples

[0067] New product introduction simulation example

[0068] The user sets up a scenario in which new product B will be introduced on June 1, 2023, at a price of 2,000 yen. The device sends this scenario to the server, which then runs a simulation within the virtual market. The results of the simulation, such as sales after the introduction of new product B, changes in market share with competing products, and the effectiveness of advertising campaigns, are calculated and displayed as graphs on the device. This allows the user to accurately predict the likelihood of success in introducing the new product and plan optimal marketing measures.

[0069] The above is a specific embodiment for carrying out the present invention. This system enables companies to perform detailed analysis and prediction of consumer behavior, and can be used as a tool to quickly respond to changing market conditions.

[0070] The processing flow will be explained below.

[0071] Step 1:

[0072] The user collects consumer data, such as consumer purchase history, customer profile, and product information, from a company's customer relationship management (CRM) system or marketing database. This data includes, for example, consumer ID, purchased product, purchase date, payment amount, age, gender, and region.

[0073] Step 2:

[0074] The terminal displays a data entry screen. The form includes fields for inputting information such as consumer ID, purchased item, purchase date, payment amount, age, gender, and region. The terminal provides an interface for the user to enter the collected data.

[0075] Step 3:

[0076] The user enters and submits consumer data. The user manually enters the required information into the data input form on the terminal and clicks the submit button. The terminal converts the entered data into JSON format and sends it to the server.

[0077] Step 4:

[0078] The server receives and analyzes the data. The server receives the JSON data sent from the device. After receiving the data, the server cleanses the data, filling in missing values ​​and processing outliers. For example, it excludes data with ages of 0 or 150.

[0079] Step 5:

[0080] The server generates a virtual consumer model based on the analyzed data. Specifically, it uses a machine learning algorithm to predict consumer purchasing behavior. For example, it uses a clustering algorithm to segment consumers and extract purchasing behavior patterns.

[0081] Step 6:

[0082] The server integrates multiple virtual consumer models. The server integrates the generated multiple virtual consumer models into a virtual market. This allows the virtual market to mimic a real market environment, including supply and demand for goods and services, as well as competitive information.

[0083] Step 7:

[0084] The user sets up a business scenario. The user inputs the business scenario into the terminal. For example, the user sets up a scenario such as "Introduce new product B on June 1, 2023 at a price of 2,000 yen." The terminal then sends this scenario to the server.

[0085] Step 8:

[0086] The server receives and analyzes the scenario. The server analyzes the received scenario data and prepares for the simulation. Based on the set scenario, the server prepares to predict fluctuations in the virtual market.

[0087] Step 9:

[0088] The server runs the simulation. The server applies a scenario within the virtual market and starts the simulation. For example, it calculates results such as estimated sales from the introduction of new product B, changes in market share with competitors' products, and the effectiveness of advertising campaigns.

[0089] Step 10:

[0090] The server generates the simulation results. After the simulation is completed, the server aggregates the results and outputs them in JSON format. The results include sales forecasts, profit margins, and market share fluctuation data.

[0091] Step 11:

[0092] The terminal receives the simulation results. The terminal receives the simulation results sent from the server. The received results are stored as they are and displayed on the user interface.

[0093] Step 12:

[0094] The terminal visualizes the results. The terminal visualizes the received results in graphs and tables and provides them to the user. The user analyzes the results and reviews their marketing strategies and business plans.

[0095] The above is a concrete processing flow of the system according to the present invention, which provides businesses with an effective tool for detailed analysis and prediction of consumer behavior and for quickly responding to changing market conditions.

[0096] Example 1

[0097] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0098] In a real market environment, it is difficult to effectively utilize consumer behavior data, quickly generate a virtual consumer model, and then simulate business scenarios based on that model. With conventional systems, data preparation, model generation, scenario setting, and simulation are all performed separately, which is extremely time-consuming and labor-intensive. Furthermore, when companies introduce new products, change prices, or plan advertising campaigns, there is no effective system that can comprehensively simulate these elements and quickly visualize and provide the results.

[0099] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0100] In this invention, the server includes means for receiving JSON data sent from the terminal, performing data cleansing and preprocessing, and generating a virtual consumer model; means for the server to integrate the generated virtual consumer model and build a virtual market; and means for the server to run a simulation within the virtual market based on a set scenario. This enables companies to quickly and effectively generate a virtual consumer model using consumer behavior data and simulate business scenarios in the virtual market. Furthermore, the simulation results can be quickly visualized to support strategy revisions and policy decisions.

[0101] "User" refers to an individual or corporation that uses the system to collect, input, and manage consumer behavior data and set scenarios.

[0102] "Device" refers to a computer device used by a user to input consumption behavior data and set up a scenario. Examples include PCs, tablets, and smartphones.

[0103] "Server" refers to a computer system that receives, analyzes, and processes data sent from terminals, generates virtual consumer models and virtual markets, and executes simulations.

[0104] "Consumer behavior data" refers to data that includes information about consumer behavior, such as consumer purchasing history, customer profiles, and product information.

[0105] "JSON" is an abbreviation for JavaScript Object Notation, and refers to a lightweight data format for data exchange. It represents structured data in a way that is easy for humans to read and machines to parse.

[0106] "Data cleansing" refers to the process of improving data quality by completing missing values, processing outliers, normalizing data, etc. as a preprocessing step for data analysis.

[0107] A "virtual consumer model" refers to a model that includes a consumer's purchasing history, interests, and future purchasing predictions, and is generated using a machine learning algorithm based on consumer behavior data.

[0108] A "virtual market" refers to a simulation environment that integrates multiple virtual consumer models and mimics a real market environment, including product supply and demand, pricing strategies, and competitor information.

[0109] A "business scenario" is a plan or strategy that a company sets up to simulate in a virtual market, including new product introductions, price changes, advertising campaigns, etc.

[0110] "Simulation" refers to the process of forecasting sales, analyzing the impact of price changes, measuring the effectiveness of advertising campaigns, etc., based on business scenarios set up within a virtual marketplace.

[0111] "Simulation results" refers to statistical data such as sales, profits, and market share obtained after executing a business scenario.

[0112] The system according to the present invention performs a series of operations: collects and analyzes real-world consumer behavior data to generate a virtual consumer model, integrates the data to build a virtual market, simulates business scenarios within the virtual market, and outputs the results. A specific embodiment of this system is described below.

[0113] Data entry and submission

[0114] Users collect consumer data

[0115] Users collect consumer behavior data from corporate customer relationship management (CRM) systems and marketing databases. Specifically, they use SQL queries to extract consumer purchasing histories and customer profiles from the databases and save them as CSV files.

[0116] The terminal displays a data entry screen.

[0117] The terminal displays a user interface for entering consumer data. The terminal can be a computer device such as a PC, tablet, or smartphone. The data entry form includes fields such as consumer ID, product purchased, purchase date, payment amount, age, gender, and region.

[0118] User enters and submits consumer data

[0119] The user manually enters data into the input form on the device and presses the submit button. The device converts the entered data into JSON format, generates an HTTP request, and sends it to the server. This is often done using a programming language such as JavaScript.

[0120] Generating the Consumer Model

[0121] The server receives and analyzes the data

[0122] The server receives the JSON data sent from the terminal and performs data cleansing using Python's Pandas library, among other tools. During this process, the server performs tasks such as filling in missing data, handling outliers, and normalizing the data.

[0123] The server generates a virtual consumer model.

[0124] The server generates a virtual consumer model based on the cleansed data. Specifically, it uses the machine learning library Scikit-learn to perform clustering and create a model that predicts consumer purchasing behavior.

[0125] Building a virtual market

[0126] The server aggregates multiple virtual consumer models.

[0127] The server integrates the generated virtual consumer models to create a virtual market, which includes information on product supply and demand, pricing, and competitors.

[0128] Business scenario simulation

[0129] The user sets up a business scenario

[0130] The user sets up a business scenario using the terminal interface. For example, the scenario "New product A will be introduced on January 5, 2024 at a price of 1,500 yen" is input and sent to the server.

[0131] The server receives the scenario and runs the simulation.

[0132] The server receives business scenarios and runs simulations within the virtual marketplace, which can include sales forecasting, analyzing the impact of price changes, and measuring the effectiveness of advertising campaigns, often using deep learning libraries such as TensorFlow and PyTorch.

[0133] Output of simulation results

[0134] The server generates the simulation results

[0135] The server generates simulation results in JSON format, including statistical data such as sales, profits, and market share.

[0136] The terminal receives and displays the results

[0137] The terminal visualizes the simulation results received from the server. For example, it uses the JavaScript library D3.js to draw a sales forecast graph and display it to the user, helping them to review strategies and decide on measures.

[0138] Specific examples

[0139] New product introduction simulation example

[0140] The user sets a scenario in which "New Product B will be introduced on June 1, 2023 at a price of 2,000 yen," and the device sends this scenario to the server. The server then runs a simulation within the virtual market and generates the results. Specific simulation results include sales after the introduction of New Product B, changes in market share with competing products, and the effectiveness of advertising campaigns. These results are displayed on the device as graphs and tables, allowing the user to plan marketing measures with high precision based on these results.

[0141] Examples of prompt statements

[0142] Below are some examples of prompt sentences to input into the generative AI model.

[0143] "You want to introduce new product B on June 1, 2023, at a price of 2,000 yen. Please conduct a market simulation and show the predicted changes in sales, profits, and market share after the introduction of new product B."

[0144] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0145] Step 1:

[0146] Users collect consumer data

[0147] Users collect consumer behavior data from a company's customer relationship management (CRM) system or marketing database. Specifically, they use SQL queries to extract consumer purchasing history and customer profiles from the database and save them as a CSV file. Input data includes consumer ID, purchased product, purchase date, payment amount, age, gender, region, etc. The output provides organized consumer behavior data.

[0148] Step 2:

[0149] The terminal displays a data entry screen.

[0150] The terminal displays a user interface for inputting consumer data. It allows the user to manually enter fields such as consumer ID, purchased product, purchase date, payment amount, age, gender, and region through a data input form. The input includes consumer behavior data collected by the user. The output includes the data entered by the user on the input screen.

[0151] Step 3:

[0152] User enters and submits consumer data

[0153] The user manually enters data into the input form on the device and presses the send button. The device converts the entered data into JSON format, generates an HTTP request, and sends it to the server. The input is consumption behavior data manually entered by the user. The output is the data converted into JSON format and sent to the server as an HTTP request.

[0154] Step 4:

[0155] The server receives and analyzes the data

[0156] The server receives JSON data sent from the device. It then performs data cleansing using the Python Pandas library. During this process, it performs tasks such as filling in missing data, handling outliers, and normalizing the data. The input is consumer behavior data in JSON format. The output is the cleansed data.

[0157] Step 5:

[0158] The server generates a virtual consumer model.

[0159] The server generates a virtual consumer model based on the preprocessed data. Specifically, it performs clustering using the Scikit-learn library to create consumer segments. The input is the cleansed consumer behavior data. The output is the generated virtual consumer model.

[0160] Step 6:

[0161] The server aggregates multiple virtual consumer models.

[0162] The server integrates the generated virtual consumer models to construct a virtual market. This virtual market includes information on product supply and demand, pricing, and competitors. The input is the multiple virtual consumer models. The output is the constructed virtual market.

[0163] Step 7:

[0164] The user sets up a business scenario

[0165] The user sets up a business scenario using the terminal interface. For example, they input a scenario such as "Introduce new product A on January 5, 2024 at a price of 1,500 yen" and send it to the server. The input includes detailed information about the business scenario. The output is the scenario sent to the server.

[0166] Step 8:

[0167] The server receives the scenario and runs the simulation.

[0168] The server receives business scenarios and runs simulations within a virtual market. It uses deep learning libraries such as TensorFlow and PyTorch to forecast sales, analyze the impact of price changes, and measure the effectiveness of advertising campaigns. The inputs are the business scenarios and virtual market data. The output is the simulation results.

[0169] Step 9:

[0170] The server generates the simulation results

[0171] The server generates the simulation results in JSON format. The results include statistical data such as sales, profit, market share, etc. As input, we have the detailed results of the simulation. As output, we get the simulation results in JSON format.

[0172] Step 10:

[0173] The terminal receives and displays the results

[0174] The terminal visualizes the simulation results received from the server. Using the JavaScript D3.js library, it draws a sales forecast graph and displays it to the user. The input is the simulation results in JSON format. The output is visualized data in graph and table format.

[0175] (Application example 1)

[0176] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0177] Conventional consumer behavior data analysis systems are limited to virtual market simulations, and have the problem of not being able to adequately predict and optimize for a variety of applications. Furthermore, for certain services, such as self-driving vehicles, there is a lack of systems that can analyze actual usage data, forecast demand, and provide optimal service routes. This has resulted in insufficient improvements in service efficiency and user satisfaction.

[0178] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0179] In this invention, the server includes means for receiving and analyzing real-world consumer behavior data, means for generating a virtual consumer model based on the analyzed data, means for integrating the generated virtual consumer model to build a virtual market, means for simulating business scenarios within the built virtual market, means for outputting simulation results, means for collecting and analyzing vehicle usage data, means for forecasting demand for on-demand services based on the usage data, and means for providing optimal service routes based on the demand forecast. This enables companies to use the results of simulations within the virtual market to forecast demand for various services and routes and to develop optimal strategies based on the forecasts.

[0180] "Actual consumer behavior data" refers to data on consumers' actual purchasing behavior and service usage. Specifically, it includes purchase history, usage history, customer profile, etc.

[0181] "Means of analysis" refers to the process of analyzing the received data and performing preprocessing such as information cleansing, handling outliers, and normalizing the data.

[0182] A "virtual consumer model" is a predictive model of consumer purchasing behavior and service usage behavior that is generated using machine learning algorithms based on real-world consumer behavior data.

[0183] A "virtual market" is a simulation environment that integrates multiple virtual consumer models and mimics a real market environment, including information on the supply and demand of each product or service, as well as competitor information.

[0184] A "business scenario" is a specific strategy or measure that a company is considering implementing. Examples include the introduction of a new product, a price change, or an advertising campaign.

[0185] "Simulation results" are statistical data such as sales forecasts, profit margins, and fluctuations in market share calculated as a result of executing a business scenario within a virtual market.

[0186] "Vehicle usage data" refers to data related to the usage history of autonomous vehicles and on-demand services, including ride history, fares, usage time, and user profiles.

[0187] "Demand forecasting for on-demand services" is the process of analyzing usage data to predict how much demand there will be for a particular service.

[0188] An "optimal service route" is a route or service plan that is optimal for a user and is provided based on a demand forecast.

[0189] This invention is a system for providing on-demand services provided by autonomous vehicles, which analyzes consumer behavior data of users and builds a virtual market to forecast demand for services and provide optimal routes. Below, we will explain in detail the processing of the program for realizing this system.

[0190] Data collection and analysis

[0191] First, the user collects information about their use of the autonomous vehicle service. This is done by extracting it from a CSV file or directly from a database. The terminal then displays a form for confirming and entering consumer data, and the user enters information such as user ID, route, usage time, payment amount, age, gender, and region. The entered data is converted to JSON format and sent to the server.

[0192] Creating a Virtual Consumer Model

[0193] The server receives the JSON data sent from the device and performs data cleansing and preprocessing. This includes filling in missing data, handling outliers, and normalizing the data. It then uses machine learning algorithms (such as clustering and regression models) to generate a virtual consumer model that predicts user behavior. The generated virtual consumer model includes the user's usage history, interests, and future usage predictions.

[0194] Building a virtual market

[0195] Next, the server integrates the generated virtual consumer models to create a virtual market. This virtual market mimics a real-world market environment, including the supply and demand of each product or service, as well as information on competitors. This makes it possible to carry out simulations within the virtual market.

[0196] Business scenario simulation

[0197] The user sets up a business scenario on the device, such as adding a new route or offering a discount service during a specific time period, and this scenario is input to the generative AI model as a prompt.

[0198] Prompt Sentence Examples

[0199] "New Route A will be introduced on October 1, 2023, with a discount campaign during the first month. We will provide demand and revenue forecasts after the introduction."

[0200] The server runs simulations within the virtual market to forecast demand, and the simulation results include data such as sales forecasts, personnel allocation, and optimal service routes.

[0201] Output of simulation results

[0202] Finally, the server generates the simulation results in JSON format and sends them to the terminal. The terminal visualizes the results received from the server and provides them to the user. The results are displayed in graphs and tables, allowing the user to review their strategy and make decisions based on them.

[0203] Hardware and software used

[0204] This system analyzes data using cloud servers (such as AWS or Google Cloud), primarily using MySQL or PostgreSQL as databases, and Python and libraries such as TensorFlow and Scikit-Learn for machine learning. The front end can be developed using frameworks such as React and Angular.

[0205] This system enables companies to use simulation results within a virtual market to forecast demand for various services and routes, and then develop optimal strategies based on that forecast.

[0206] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0207] Step 1:

[0208] Users collect usage data for the autonomous vehicle service. Specifically, information such as user ID, route, usage time, payment amount, age, gender, and region is collected using CSV files or database extraction. The collected data is then entered into a terminal.

[0209] Step 2:

[0210] The terminal displays a form for verifying and entering consumer data. The user enters the collected data into this form, confirms the input, and clicks the submit button. The input at this point is detailed information related to the user.

[0211] Step 3:

[0212] The terminal converts the input data into JSON format and sends it to the server. This converts the information entered by the user into a data structure and sends it via an HTTP request. The output is structured user data.

[0213] Step 4:

[0214] The server receives the JSON data sent from the terminal. After receiving it, the server performs data cleansing and preprocessing. Specifically, it completes missing data, processes outliers, and normalizes the data. The input is JSON data, and the output is the preprocessed, clean data.

[0215] Step 5:

[0216] The server generates a virtual consumer model based on the preprocessed data. Machine learning algorithms (e.g., clustering and regression models) are used to create a model that predicts consumption behavior, interests, and future behavior. The input is the preprocessed data, and the output is the virtual consumer model.

[0217] Step 6:

[0218] The server integrates multiple generated virtual consumer models to build a virtual market. This virtual market is a simulation environment that includes information on the supply and demand of each product or service, as well as information on competitors. The input is the virtual consumer model, and the output is the virtual market.

[0219] Step 7:

[0220] The user sets a business scenario on the terminal. For example, they set a specific prompt such as "New route A will be introduced on October 1, 2023, and a discount campaign will be held in the first month." The input is the business scenario, and the output is the prompt.

[0221] Step 8:

[0222] The server executes a simulation in the virtual market based on the received prompt. Specific simulation operations include demand forecasting, sales forecasting, personnel allocation, and calculation of optimal service routes after the introduction of new routes. The input is the prompt, and the output is the simulation results.

[0223] Step 9:

[0224] The server generates simulation results in JSON format. The simulation results include data such as demand forecasts, sales, and personnel allocation. These are sent to the terminal. The input is the simulation results, and the output is the result data in JSON format.

[0225] Step 10:

[0226] The terminal visualizes the simulation results received from the server and presents them to the user. The results are displayed in graphs and tables, and the user can use them to review strategies and make decisions. The input is the simulation results in JSON format, and the output is visualized data.

[0227] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0228] The system according to the present invention is characterized by the fact that it not only receives and analyzes real consumer behavior data, but also combines it with an emotion engine that recognizes user emotions to generate a more sophisticated virtual consumer model. Below, we will create a program for the system and explain its processing in detail.

[0229] Data entry and submission

[0230] 1. Users collect consumer data

[0231] Users collect consumer purchasing history, customer profiles, product information, etc. from companies' customer relationship management (CRM) systems and marketing databases, including consumer sentiment data, such as how consumers felt when they purchased a product and their feedback comments.

[0232] 2. The device displays the data entry screen.

[0233] The terminal provides a user interface and displays a form for the user to input the collected consumer data and emotion data, including the consumer ID, purchased item, purchase date, payment amount, age, gender, region, and emotion tag (e.g., joy, anger, sadness, etc.).

[0234] 3. The user enters and submits data

[0235] The user enters the required information into the data input form on the terminal and clicks the submit button. The terminal converts the entered data into JSON format and sends it to the server.

[0236] Generating the Consumer Model

[0237] 4. The server receives and analyzes the data

[0238] The server receives the JSON data sent from the device. After receiving the data, the server performs data cleansing, fills in missing values, and processes outliers. This also includes checking whether the emotion data is normal.

[0239] 5. The server generates a virtual consumer model

[0240] The server generates a virtual consumer model based on the analyzed consumer behavior data and emotion data. Specifically, it uses a machine learning algorithm to create a model that predicts the relationship between consumer purchasing behavior and emotions. The generated virtual consumer model includes the consumer's purchasing history, emotional state, interests, and future purchase predictions.

[0241] Building a virtual market

[0242] 6. The server aggregates multiple virtual consumer models

[0243] The server integrates multiple virtual consumer models to create a virtual market that mimics a real-world market environment, including supply and demand for each product or service, competitive information, and consumer sentiment data.

[0244] Business scenario simulation

[0245] 7. The user sets up the business scenario

[0246] The user sets up a business scenario on the device. For example, the scenario is set up as follows: "New product B will be introduced on June 1, 2023, at a price of 2,000 yen. Review sentiment at the time of introduction will be analyzed." The user enters detailed information about the scenario and sends it to the server.

[0247] 8. The server receives and analyzes the scenario.

[0248] The server analyzes the received scenario data and prepares for the simulation. Based on the set scenario, it prepares to predict fluctuations in the virtual market.

[0249] 9. The server runs the simulation

[0250] The server applies a scenario within the virtual market and starts a simulation. For example, it predicts sales from the introduction of new product B, market share fluctuations compared to competitors' products, and emotional reactions at the time of introduction. The emotion engine analyzes consumer reviews and feedback to determine positive or negative emotions.

[0251] Output of simulation results

[0252] 10. The server generates the simulation results

[0253] After the simulation is completed, the server aggregates the results and outputs them in JSON format, including sales forecasts, profit margins, market share fluctuation data, and sentiment analysis results.

[0254] 11. The terminal receives and displays the simulation results.

[0255] The terminal receives the simulation results sent from the server, and the received results are stored and displayed on the user interface.

[0256] 12. The terminal visualizes the results

[0257] The device visualizes the received results in graphs and tables and provides them to the user. Sentiment analysis results are displayed as the percentage of positive reactions, the percentage of negative reactions, etc. The user can analyze the results and revise their marketing strategies and business plans.

[0258] Specific examples

[0259] New product introduction scenario

[0260] The user sets up a scenario in which new product B will be introduced on June 1, 2023, at a price of 2,000 yen, and consumer review sentiment will be analyzed. The device sends this scenario to the server, which then runs a simulation within the virtual market. The simulation results, calculated as graphs on the device, include sales after the introduction of new product B, changes in market share compared to competing products, and sentiment analysis results based on consumer reviews. This allows the user to accurately predict new product introductions and changes in consumer sentiment, and plan optimal marketing measures.

[0261] The above is a specific embodiment for carrying out the present invention. This system enables companies to analyze and predict consumer behavior in detail, and to develop flexible marketing strategies that take consumer sentiment into account.

[0262] The processing flow will be explained below.

[0263] Step 1:

[0264] The user collects consumer data, such as consumer purchasing history, customer profile, product information, and emotional data, from a company's customer relationship management (CRM) system or marketing database. For example, data such as consumer ID, purchased product, purchase date, payment amount, age, gender, region, and emotional tag (happiness, anger, sadness, etc.) is collected.

[0265] Step 2:

[0266] The terminal displays a data entry screen. The form includes the consumer ID, purchased item, purchase date, payment amount, age, gender, region, and emotion tag (e.g., joy, anger, sadness, etc.). The terminal provides an interface for the user to enter the collected data.

[0267] Step 3:

[0268] The user enters data and submits it. The user manually enters the required information into the data input form on the terminal and clicks the submit button. The terminal converts the entered data into JSON format and sends it to the server.

[0269] Step 4:

[0270] The server receives and analyzes the data. The server receives the JSON data sent from the device. After receiving the data, the server cleanses the data, filling in missing values ​​and processing outliers. For example, it removes unnatural age data (such as 0 or 150 years old) and obviously incorrect emotion tags.

[0271] Step 5:

[0272] The server uses an emotion engine to analyze user emotions based on text, audio, or video input. For example, it analyzes consumer reviews and feedback comments using natural language processing (NLP) techniques and assigns positive, negative, or neutral emotion tags.

[0273] Step 6:

[0274] The server generates a virtual consumer model based on the analyzed consumer behavior data and emotion data. Specifically, a machine learning algorithm is used to create a model that predicts the relationship between consumer purchasing behavior and emotions. The generated virtual consumer model includes the consumer's purchasing history, emotional state, interests, and future purchase predictions.

[0275] Step 7:

[0276] The server integrates multiple virtual consumer models. The server then integrates the multiple virtual consumer models that have been generated to create a virtual market. This virtual market mimics a real market environment and includes information such as supply and demand for each product or service, competitive information, and consumer sentiment data.

[0277] Step 8:

[0278] The user sets up a business scenario. The user inputs the business scenario into the device. For example, the scenario is set as "New product B will be introduced on June 1, 2023, at a price of 2,000 yen. Review sentiment at this time will be analyzed." The device then sends this scenario to the server.

[0279] Step 9:

[0280] The server receives and analyzes the scenario. The server analyzes the received scenario data and prepares for the simulation. Based on the set scenario, the server prepares to predict fluctuations in the virtual market.

[0281] Step 10:

[0282] The server runs the simulation. The server applies a scenario within the virtual market and starts the simulation. For example, it predicts sales estimates for the introduction of new product B, changes in market share with competitors' products, and emotional reactions based on consumer reviews at the time of introduction. The emotion engine analyzes consumer reviews and feedback to determine positive or negative emotions.

[0283] Step 11:

[0284] The server generates the simulation results. After the simulation is completed, the server aggregates the results and outputs them in JSON format. The results include sales forecasts, profit margins, market share fluctuation data, as well as sentiment analysis results.

[0285] Step 12:

[0286] The terminal receives and displays the simulation results. The terminal receives the simulation results sent from the server. The received results are saved as they are and displayed on the user interface.

[0287] Step 13:

[0288] The device visualizes the results. The device visualizes the received results in graphs and tables and provides them to the user. Sentiment analysis results are displayed as the percentage of positive responses, the percentage of negative responses, etc. The user analyzes the results and reviews their marketing strategies and business plans.

[0289] Example 2

[0290] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0291] Simply analyzing real-world consumer behavior data is not enough to fully understand consumer emotions and the motivations behind their purchasing behavior. For this reason, there is a need for methods to generate more sophisticated virtual consumer models and predict consumer behavior with high accuracy. In particular, the insufficient analysis of emotional data is a challenge.

[0292] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0293] In this invention, the server includes means for receiving and analyzing real consumer behavior data, means for generating a virtual consumer model based on the analyzed data, and means for integrating the generated virtual consumer model to build a virtual market. This enables comprehensive analysis of consumer behavior and emotions by integrating means for recognizing and analyzing consumer emotions with means for refining the virtual consumer model based on emotion data, making it possible to make more accurate predictions and formulate marketing strategies.

[0294] "Consumer behavior data" refers to information about what products and services consumers have purchased, such as their purchasing history, customer profiles, and product information.

[0295] An "analytical means" is a method or device for extracting, processing, and analyzing data to understand its patterns and characteristics.

[0296] A "virtual consumer model" is a consumer behavior prediction model that is virtually created based on consumer behavior data, and includes the consumer's purchasing history, emotional state, interests, and future purchasing predictions.

[0297] A "virtual market" is a virtual environment that mimics a real market environment and is constructed by integrating virtual consumer models.

[0298] A "business scenario" is a scenario used to set assumptions for a company's marketing strategy or business plan, such as the introduction of a new product, price changes, or advertising campaigns.

[0299] A "simulating means" is a method or device for predicting fluctuations and results in a virtual market based on a set business scenario.

[0300] "Simulation results" are forecast results obtained by running a simulation based on a scenario, and include sales forecasts, profit margins, market share fluctuation data, and sentiment analysis results.

[0301] A "means for recognizing and analyzing sentiment" is a method or device for identifying positive or negative sentiment from consumer reviews and feedback comments and analyzing the data.

[0302] "Emotional data" is data that represents the emotions and feedback that consumers have about products and services.

[0303] A "refinement means" is a method or device for processing and improving existing data or models in more detail and accuracy.

[0304] The system according to the present invention is characterized by the fact that it not only receives and analyzes real consumer behavior data, but also combines it with an emotion engine that recognizes user emotions to generate a more sophisticated virtual consumer model. The processing of the system program is explained in detail below.

[0305] Hardware and software used

[0306] In this system, three entities, the server, the terminal, and the user, work together. The specific hardware and software used include the following:

[0307] Server: A high-performance data analysis server (e.g., AWS EC2, Google Cloud Platform, Microsoft Azure).

[0308] Device: A device such as a computer or smartphone used by a user.

[0309] Software: Machine learning algorithms (e.g., TensorFlow, PyTorch), database management systems (e.g., MySQL, PostgreSQL), sentiment analysis engines (e.g., IBM Watson, Microsoft Text Analytics).

[0310] Specific examples of system processing

[0311] 1. Users collect consumer data

[0312] Users collect consumer purchasing history, customer profiles, and product information from companies' customer relationship management (CRM) systems and marketing databases, including consumer sentiment and feedback comments when they purchase a product.

[0313] 2. The device displays the data entry screen.

[0314] The terminal displays a form for the user to enter the collected consumer and emotional data, including consumer ID, purchased item, purchase date, payment amount, age, gender, region, and emotional tag (e.g., joy, anger, sadness, etc.).

[0315] 3. The user enters and submits data

[0316] The user enters the required information into the data input form on the terminal and clicks the submit button. The terminal converts the entered data into JSON format and sends it to the server.

[0317] 4. The server receives and analyzes the data

[0318] The server receives the JSON data sent from the device. After receiving the data, the server cleanses the data, filling in missing values ​​and processing outliers. It verifies that the emotion data is normal and removes any invalid data.

[0319] 5. The server generates a virtual consumer model

[0320] The server uses a machine learning algorithm to generate a virtual consumer model based on the analyzed consumer behavior data and emotional data, which includes the consumer's purchasing history, emotional state, interests, and future purchasing predictions.

[0321] 6. The server aggregates multiple virtual consumer models

[0322] The server integrates the generated virtual consumer models to create a virtual market, which includes supply and demand for goods and services, competitive information, and consumer sentiment data.

[0323] Examples and prompts

[0324] For example, a user can set up a business scenario in which "New Product B will be introduced on June 1, 2023, at a price of 2,000 yen, and consumer review sentiment will be analyzed." The device sends this scenario to the server, which then runs a simulation within the virtual market. The results of the simulation are calculated and displayed as graphs on the device, including sales after the introduction of New Product B, changes in market share compared to competing products, and sentiment analysis results based on consumer reviews.

[0325] Prompt Sentence Examples

[0326] For example, consider the following prompt:

[0327] "New Product B will be introduced on June 1, 2023, at a price of 2,000 yen. Analyze review sentiment at the time of this introduction to see whether there are more positive or negative reactions."

[0328] The above is a specific embodiment for carrying out the present invention. This system enables companies to carry out detailed analysis and predictions that take into account consumer behavior and emotions, enabling the formulation of flexible marketing strategies.

[0329] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0330] Step 1:

[0331] Users collect consumer data

[0332] What it does: The user collects consumer purchasing history, customer profiles, product information, and sentiment data from the company's CRM system and marketing database.

[0333] Input: Consumer behavior and sentiment data from customer relationship management systems and marketing databases.

[0334] Output: The collected consumer behavior and sentiment data is prepared in a format for data input.

[0335] Step 2:

[0336] The terminal displays a data entry screen.

[0337] Specific behavior: The device provides a user interface and displays a form for inputting collected data, including consumer ID, purchased item, purchase date, payment amount, age, gender, region, and emotion tag (e.g., joy, anger, sadness, etc.).

[0338] Input: Consumer behavior and sentiment data collected by users.

[0339] Output: A data entry form is displayed on the terminal screen.

[0340] Step 3:

[0341] The user enters and submits data

[0342] Specific operation: The user enters the required information into the data input form on the terminal and clicks the submit button. The terminal converts the entered data into JSON format and sends it to the server.

[0343] Input: Consumer behavior and sentiment data entered into a data entry form.

[0344] Output: The data is converted to JSON format and sent to the server.

[0345] Step 4:

[0346] The server receives and analyzes the data

[0347] Specific operation: The server receives JSON data sent from the device. After receiving it, it performs data cleansing, fills in missing values, and processes outliers. It also checks whether the emotion data is normal and removes invalid data.

[0348] Input: JSON formatted consumer behavior and sentiment data sent from the device.

[0349] Output: Cleansed and accurate consumer behavior and sentiment data.

[0350] Step 5:

[0351] The server generates a virtual consumer model.

[0352] Specific operation: The server uses a machine learning algorithm to generate a virtual consumer model based on the analyzed consumer behavior data and emotion data. The generated model includes the consumer's purchase history, emotional state, interests, and future purchase predictions.

[0353] Input: Cleansed consumer behavior and sentiment data.

[0354] Output: A hypothetical consumer model.

[0355] Step 6:

[0356] The server aggregates multiple virtual consumer models.

[0357] Specific operation: The server integrates the generated multiple virtual consumer models to build a virtual market, which includes supply and demand for goods and services, competitive information, and consumer sentiment data.

[0358] Input: Multiple hypothetical consumer models.

[0359] Output: An integrated virtual marketplace.

[0360] Step 7:

[0361] The user sets up a business scenario

[0362] Specific operation: The user sets up a business scenario on the device. For example, they set up a scenario such as "Introduce new product B on June 1, 2023 at a price of 2,000 yen. Analyze review sentiment at the time of introduction." They then enter detailed scenario information and send it to the server.

[0363] Input: Business scenario (e.g. introduction of new product B, price, introduction date).

[0364] Output: Business scenario data sent to the server.

[0365] Step 8:

[0366] The server receives and analyzes the scenario.

[0367] Specific operation: The server analyzes the received scenario data and prepares for the simulation. Based on the set scenario, it prepares to predict fluctuations in the virtual market.

[0368] Input: Submitted business scenario data.

[0369] Output: Simulation ready.

[0370] Step 9:

[0371] The server runs the simulation

[0372] How it works: The server applies a scenario within the virtual market and starts a simulation. For example, it predicts sales from the introduction of new product B, changes in market share of competing products, and emotional reactions to the introduction. The emotion engine analyzes consumer reviews and feedback to determine positive or negative emotions.

[0373] Input: Simulation preparation data, review and feedback data to be analyzed for sentiment.

[0374] Output: Simulation results (sales forecast, market share change of competing products, predicted emotional response).

[0375] Step 10:

[0376] The server generates the simulation results

[0377] How it works: After the simulation is completed, the server aggregates the results and outputs them in JSON format, including sales forecasts, profit margins, market share fluctuation data, and sentiment analysis results.

[0378] Input: Various data after the simulation is run.

[0379] Output: Simulation results organized in JSON format.

[0380] Step 11:

[0381] The terminal receives and displays the simulation results.

[0382] Specific operation: The terminal receives the simulation results sent from the server, saves them, and displays them on the user interface.

[0383] Input: Simulation results sent from the server.

[0384] Output: Simulation results displayed on the terminal screen.

[0385] Step 12:

[0386] The terminal visualizes the results

[0387] Specific operation: The device visualizes the received results in graphs and tables and provides them to the user. Sentiment analysis results are displayed as the percentage of positive reactions, the percentage of negative reactions, etc. The user analyzes the results and revises their marketing strategy and business plan.

[0388] Input: Received simulation results.

[0389] Output: A visual summary in graphical and tabular form.

[0390] (Application example 2)

[0391] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0392] Modern purchasing behavior analysis does not adequately collect and analyze consumer emotional data, making it difficult to accurately grasp consumers' true needs and reactions. As a result, the accuracy of marketing strategies and product recommendations is limited, making it difficult to improve consumer satisfaction and maximize sales. There is a need to solve this problem, generate more accurate consumer models, and provide a system that can make optimal product recommendations.

[0393] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0394] In this invention, the server includes means for receiving and analyzing real consumer behavior data, means for generating a virtual consumer model based on the analyzed data, means for integrating the generated virtual consumer models to build a virtual market, means for collecting gaze and facial expression data and analyzing consumer emotions, means for recommending optimal products to users based on the analyzed emotion data, and means for outputting simulation results. This makes it possible to generate a sophisticated virtual consumer model based on consumer emotions, and to recommend optimal products and formulate marketing strategies.

[0395] "Actual consumer behavior data" refers to data such as the history, behavioral information, and payment information of consumers when they actually purchase products, and indicates the specific behavior of individual consumers.

[0396] "Means for analysis" refers to a device or program for analyzing collected data and extracting useful information.

[0397] A "virtual consumer model" is a model that is generated based on real consumer behavior data and mimics changes in consumer purchasing behavior and emotions.

[0398] The "means for integrating and constructing a virtual market" is a device or program for aggregating multiple virtual consumer models and generating a virtual market that mimics a real market environment.

[0399] A "means for simulating business scenarios" is a device or program for testing business plans such as the introduction of new products, price changes, and the implementation of information provision systems in a virtual market and predicting their effects.

[0400] "Gaze and facial expression data" refers to data that shows the eye movements and facial expressions of consumers when they view products, and is information used to analyze consumers' emotions and interests.

[0401] "Means for analyzing consumer emotions" refers to a device or program for recognizing and analyzing the emotional state of consumers based on gaze and facial expression data.

[0402] The "means for recommending optimal products to users" refers to a device or program that selects and suggests products that are likely to interest users based on analyzed emotional data and purchasing behavior data.

[0403] The "means for outputting the simulation results" is a device or program for displaying or reporting the results of the business scenario simulation.

[0404] This invention is a system that builds a virtual market by collecting and analyzing real-world consumer behavior data and consumer emotion data, and then generating and integrating a virtual consumer model based on that data. It also simulates business scenarios within that virtual market and outputs the simulation results. This system also includes functions to collect gaze and facial expression data, analyze consumer emotions, and recommend optimal products to users based on the analyzed emotion data.

[0405] 1. Data collection and transmission

[0406] The user wears the smart glasses and browses the products. The smart glasses' built-in camera and sensors collect the user's gaze and facial expression data in real time.

[0407] The collected data is converted into JSON format and sent to the server via the terminal.

[0408] 2. Data Reception and Analysis

[0409] The server receives the JSON data sent from the device and performs data cleansing on the received data, including filling in missing values ​​and processing outliers.

[0410] The server analyzes consumer emotions from gaze and facial expression data using software such as OpenCV, dlib, and the GazeTracking library.

[0411] 3. Creating a Virtual Consumer Model

[0412] The server generates a virtual consumer model based on the analyzed consumer behavior and emotion data, often using machine learning algorithms.

[0413] The generated virtual consumer model includes purchasing history, emotional state, interests, and future purchase predictions.

[0414] 4. Building a virtual market

[0415] The server integrates the generated virtual consumer models to create a virtual market, which includes supply and demand for each product or service, competitive information, and consumer sentiment data.

[0416] 5. Business scenario simulation

[0417] Users set up business scenarios such as new product introductions, price changes, and advertising campaigns on their terminals.

[0418] The server predicts and runs simulations of virtual market fluctuations based on set scenarios, resulting in forecasts of sales, profit margins, market share fluctuations, and consumer emotional responses.

[0419] 6. Output of simulation results

[0420] The server outputs the simulation results in JSON format, and the terminal receives and visualizes them.

[0421] The user reviews their marketing strategy and business plan based on the output results.

[0422] Specific examples

[0423] If a user wears smart glasses and smiles when looking at Product A, the application will analyze their facial expression and gaze and recommend "Recommended Product A." Below is a specific example of a prompt sentence to input to the generative AI model.

[0424] Example prompt sentence:

[0425] When a user wears smart glasses and smiles when looking at product A, the application should analyze their facial expressions and gaze and recommend the most suitable product in real time.

[0426] The above is a specific embodiment for carrying out the invention. This system allows for a precise understanding of consumer emotions, making it possible to recommend optimal products and formulate marketing strategies.

[0427] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0428] Step 1:

[0429] A user puts on the smart glasses and browses the products.

[0430] (input)

[0431] User gaze and facial expression data.

[0432] (process)

[0433] The smart glasses' built-in cameras and sensors collect the user's gaze and facial expression data.

[0434] (output)

[0435] Real-time gaze and facial expression data is captured.

[0436] Step 2:

[0437] The data collected by the device is converted into JSON format and sent to the server.

[0438] (input)

[0439] Gaze and facial expression data.

[0440] (process)

[0441] The terminal converts the data into JSON format and sends it to the server over the network.

[0442] (output)

[0443] Gaze and facial expression data in JSON format is sent to the server.

[0444] Step 3:

[0445] The server performs data cleansing on the data received.

[0446] (input)

[0447] Gaze and expression data in JSON format.

[0448] (process)

[0449] The server cleanses the data by imputing missing values ​​and handling outliers.

[0450] (output)

[0451] Cleansed gaze and facial expression data.

[0452] Step 4:

[0453] The server analyzes consumer emotions from gaze and facial expression data.

[0454] (input)

[0455] Cleansed gaze and facial expression data.

[0456] (process)

[0457] Emotions are analyzed using software such as OpenCV, dlib, and GazeTracking. Specifically, changes in gaze direction and facial expressions are detected to determine the emotional state.

[0458] (output)

[0459] Consumer sentiment data.

[0460] Step 5:

[0461] The server generates a virtual consumer model based on the analyzed data.

[0462] (input)

[0463] Analyzed consumer sentiment data and consumption behavior data.

[0464] (process)

[0465] Use machine learning algorithms to create models that predict the relationship between consumer purchasing behavior and emotions.

[0466] (output)

[0467] Virtual consumer model.

[0468] Step 6:

[0469] The server integrates the generated virtual consumer models and constructs a virtual market.

[0470] (input)

[0471] Multiple virtual consumer models.

[0472] (process)

[0473] The server integrates multiple virtual consumer models to create a virtual market, which includes supply and demand for each product or service, competitive information, and consumer sentiment data.

[0474] (output)

[0475] Hypothetical market data.

[0476] Step 7:

[0477] The user sets up a business scenario on the terminal.

[0478] (input)

[0479] Scenario data for new product introductions, price changes, advertising campaigns, etc.

[0480] (process)

[0481] The user uses the interface on the terminal to input detailed information about the scenario and transmits it to the server.

[0482] (output)

[0483] The scenario data is sent to the server.

[0484] Step 8:

[0485] The server analyzes the received scenario data and executes a simulation.

[0486] (input)

[0487] Scenario and hypothetical market data.

[0488] (process)

[0489] Based on a set scenario, a simulation is run to predict fluctuations in a virtual market.

[0490] (output)

[0491] Simulation result data.

[0492] Step 9:

[0493] The server outputs the simulation results in JSON format, and the terminal receives and visualizes them.

[0494] (input)

[0495] Simulation result data.

[0496] (process)

[0497] The server outputs the simulation results in JSON format, and the terminal receives them and visualizes them in graphs or tables.

[0498] (output)

[0499] Visualized simulation results.

[0500] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0501] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0502] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0503] [Second embodiment]

[0504] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0505] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0506] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0507] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0508] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0509] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0510] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0511] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0512] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0513] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0514] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0515] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0516] The system according to the present invention performs a series of operations: it receives and analyzes real-world consumer behavior data, generates virtual consumer models, integrates them to build a virtual market, simulates business scenarios within the built virtual market, and outputs the results. Below, we will create a program for the system and explain its processing in detail.

[0517] Data entry and submission

[0518] 1. Users collect consumer data

[0519] Users collect consumer purchasing history, customer profiles, product information, etc. from a company's customer relationship management (CRM) system or marketing database. This data collection is done by extracting data from CSV files or directly from the database.

[0520] 2. The device displays the data entry screen.

[0521] The terminal provides a user interface and displays a form for the user to review and enter the collected consumer data, including information such as consumer ID, purchased product, purchase date, payment amount, age, gender, and region.

[0522] 3. The user enters and submits consumer data

[0523] The user enters the required information into the data input form on the terminal and clicks the submit button. The terminal converts the entered data into JSON format and sends it to the server.

[0524] Generating the Consumer Model

[0525] 4. The server receives and analyzes the data

[0526] The server receives the JSON data sent from the terminal. Specifically, it is designed to receive the data as an HTTP request so that it can reach the server.

[0527] The server analyzes the received data and performs data cleansing and preprocessing, which includes imputing missing data, handling outliers, and normalizing the data.

[0528] 5. The server generates a virtual consumer model

[0529] The server generates a virtual consumer model based on the analyzed data. Here, machine learning algorithms (e.g., clustering and regression models) are used to create a model that predicts consumer purchasing behavior. The generated virtual consumer model includes the consumer's purchasing history, interests, and future purchase predictions.

[0530] Building a virtual market

[0531] 6. The server aggregates multiple virtual consumer models

[0532] The server integrates the generated virtual consumer models to create a virtual market, which mimics a real-world market environment and includes information on the supply and demand of each product or service, as well as information on competitors.

[0533] Business scenario simulation

[0534] 7. The user sets up the business scenario

[0535] The user sets up a business scenario on the device. For example, a scenario could be "Introduce new product B on June 1, 2023 at a price of 2,000 yen." The user enters detailed information about the scenario and sends it to the server.

[0536] 8. The server receives the scenario and runs the simulation.

[0537] The server receives the scenario and runs a simulation in the virtual market based on that scenario, predicting things like sales fluctuations due to new product introductions, demand changes due to price changes, and the effects of advertising campaigns.

[0538] The simulation results include sales forecasts, profit margins, and changes in market share.

[0539] Output of simulation results

[0540] 9. The server generates the simulation results

[0541] The server generates the simulation results in JSON format, which include statistics on sales, profits, and market share after the scenario is executed.

[0542] 10. The device receives and displays the results

[0543] The terminal visualizes the simulation results received from the server and provides them to the user. The results are displayed in graphs and tables, and the user can use them to review strategies and make decisions.

[0544] Specific examples

[0545] New product introduction simulation example

[0546] The user sets up a scenario in which new product B will be introduced on June 1, 2023, at a price of 2,000 yen. The device sends this scenario to the server, which then runs a simulation within the virtual market. The results of the simulation, such as sales after the introduction of new product B, changes in market share with competing products, and the effectiveness of advertising campaigns, are calculated and displayed as graphs on the device. This allows the user to accurately predict the likelihood of success in introducing the new product and plan optimal marketing measures.

[0547] The above is a specific embodiment for carrying out the present invention. This system enables companies to perform detailed analysis and prediction of consumer behavior, and can be used as a tool to quickly respond to changing market conditions.

[0548] The processing flow will be explained below.

[0549] Step 1:

[0550] The user collects consumer data, such as consumer purchase history, customer profile, and product information, from a company's customer relationship management (CRM) system or marketing database. This data includes, for example, consumer ID, purchased product, purchase date, payment amount, age, gender, and region.

[0551] Step 2:

[0552] The terminal displays a data entry screen. The form includes fields for inputting information such as consumer ID, purchased item, purchase date, payment amount, age, gender, and region. The terminal provides an interface for the user to enter the collected data.

[0553] Step 3:

[0554] The user enters and submits consumer data. The user manually enters the required information into the data input form on the terminal and clicks the submit button. The terminal converts the entered data into JSON format and sends it to the server.

[0555] Step 4:

[0556] The server receives and analyzes the data. The server receives the JSON data sent from the device. After receiving the data, the server cleanses the data, filling in missing values ​​and processing outliers. For example, it excludes data with ages of 0 or 150.

[0557] Step 5:

[0558] The server generates a virtual consumer model based on the analyzed data. Specifically, it uses a machine learning algorithm to predict consumer purchasing behavior. For example, it uses a clustering algorithm to segment consumers and extract purchasing behavior patterns.

[0559] Step 6:

[0560] The server integrates multiple virtual consumer models. The server integrates the generated multiple virtual consumer models into a virtual market. This allows the virtual market to mimic a real market environment, including supply and demand for goods and services, as well as competitive information.

[0561] Step 7:

[0562] The user sets up a business scenario. The user inputs the business scenario into the terminal. For example, the user sets up a scenario such as "Introduce new product B on June 1, 2023 at a price of 2,000 yen." The terminal then sends this scenario to the server.

[0563] Step 8:

[0564] The server receives and analyzes the scenario. The server analyzes the received scenario data and prepares for the simulation. Based on the set scenario, the server prepares to predict fluctuations in the virtual market.

[0565] Step 9:

[0566] The server runs the simulation. The server applies a scenario within the virtual market and starts the simulation. For example, it calculates results such as estimated sales from the introduction of new product B, changes in market share with competitors' products, and the effectiveness of advertising campaigns.

[0567] Step 10:

[0568] The server generates the simulation results. After the simulation is completed, the server aggregates the results and outputs them in JSON format. The results include sales forecasts, profit margins, and market share fluctuation data.

[0569] Step 11:

[0570] The terminal receives the simulation results. The terminal receives the simulation results sent from the server. The received results are stored as they are and displayed on the user interface.

[0571] Step 12:

[0572] The terminal visualizes the results. The terminal visualizes the received results in graphs and tables and provides them to the user. The user analyzes the results and reviews their marketing strategies and business plans.

[0573] The above is a concrete processing flow of the system according to the present invention, which provides businesses with an effective tool for detailed analysis and prediction of consumer behavior and for quickly responding to changing market conditions.

[0574] Example 1

[0575] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0576] In a real market environment, it is difficult to effectively utilize consumer behavior data, quickly generate a virtual consumer model, and then simulate business scenarios based on that model. With conventional systems, data preparation, model generation, scenario setting, and simulation are all performed separately, which is extremely time-consuming and labor-intensive. Furthermore, when companies introduce new products, change prices, or plan advertising campaigns, there is no effective system that can comprehensively simulate these elements and quickly visualize and provide the results.

[0577] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0578] In this invention, the server includes means for receiving JSON data sent from the terminal, performing data cleansing and preprocessing, and generating a virtual consumer model; means for the server to integrate the generated virtual consumer model and build a virtual market; and means for the server to run a simulation within the virtual market based on a set scenario. This enables companies to quickly and effectively generate a virtual consumer model using consumer behavior data and simulate business scenarios in the virtual market. Furthermore, the simulation results can be quickly visualized to support strategy revisions and policy decisions.

[0579] "User" refers to an individual or corporation that uses the system to collect, input, and manage consumer behavior data and set scenarios.

[0580] "Device" refers to a computer device used by a user to input consumption behavior data and set up a scenario. Examples include PCs, tablets, and smartphones.

[0581] "Server" refers to a computer system that receives, analyzes, and processes data sent from terminals, generates virtual consumer models and virtual markets, and executes simulations.

[0582] "Consumer behavior data" refers to data that includes information about consumer behavior, such as consumer purchasing history, customer profiles, and product information.

[0583] "JSON" is an abbreviation for JavaScript Object Notation, and refers to a lightweight data format for data exchange. It represents structured data in a way that is easy for humans to read and machines to parse.

[0584] "Data cleansing" refers to the process of improving data quality by completing missing values, processing outliers, normalizing data, etc. as a preprocessing step for data analysis.

[0585] A "virtual consumer model" refers to a model that includes a consumer's purchasing history, interests, and future purchasing predictions, and is generated using a machine learning algorithm based on consumer behavior data.

[0586] A "virtual market" refers to a simulation environment that integrates multiple virtual consumer models and mimics a real market environment, including product supply and demand, pricing strategies, and competitor information.

[0587] A "business scenario" is a plan or strategy that a company sets up to simulate in a virtual market, including new product introductions, price changes, advertising campaigns, etc.

[0588] "Simulation" refers to the process of forecasting sales, analyzing the impact of price changes, measuring the effectiveness of advertising campaigns, etc., based on business scenarios set up within a virtual marketplace.

[0589] "Simulation results" refers to statistical data such as sales, profits, and market share obtained after executing a business scenario.

[0590] The system according to the present invention performs a series of operations: collects and analyzes real-world consumer behavior data to generate a virtual consumer model, integrates the data to build a virtual market, simulates business scenarios within the virtual market, and outputs the results. A specific embodiment of this system is described below.

[0591] Data entry and submission

[0592] Users collect consumer data

[0593] Users collect consumer behavior data from corporate customer relationship management (CRM) systems and marketing databases. Specifically, they use SQL queries to extract consumer purchasing histories and customer profiles from the databases and save them as CSV files.

[0594] The terminal displays a data entry screen.

[0595] The terminal displays a user interface for entering consumer data. The terminal can be a computer device such as a PC, tablet, or smartphone. The data entry form includes fields such as consumer ID, product purchased, purchase date, payment amount, age, gender, and region.

[0596] User enters and submits consumer data

[0597] The user manually enters data into the input form on the device and presses the submit button. The device converts the entered data into JSON format, generates an HTTP request, and sends it to the server. This is often done using a programming language such as JavaScript.

[0598] Generating the Consumer Model

[0599] The server receives and analyzes the data

[0600] The server receives the JSON data sent from the terminal and performs data cleansing using Python's Pandas library, among other tools. During this process, the server performs tasks such as filling in missing data, handling outliers, and normalizing the data.

[0601] The server generates a virtual consumer model.

[0602] The server generates a virtual consumer model based on the cleansed data. Specifically, it uses the machine learning library Scikit-learn to perform clustering and create a model that predicts consumer purchasing behavior.

[0603] Building a virtual market

[0604] The server aggregates multiple virtual consumer models.

[0605] The server integrates the generated virtual consumer models to create a virtual market, which includes information on product supply and demand, pricing, and competitors.

[0606] Business scenario simulation

[0607] The user sets up a business scenario

[0608] The user sets up a business scenario using the terminal interface. For example, the scenario "New product A will be introduced on January 5, 2024 at a price of 1,500 yen" is input and sent to the server.

[0609] The server receives the scenario and runs the simulation.

[0610] The server receives business scenarios and runs simulations within the virtual marketplace, which can include sales forecasting, analyzing the impact of price changes, and measuring the effectiveness of advertising campaigns, often using deep learning libraries such as TensorFlow and PyTorch.

[0611] Output of simulation results

[0612] The server generates the simulation results

[0613] The server generates simulation results in JSON format, including statistical data such as sales, profits, and market share.

[0614] The terminal receives and displays the results

[0615] The terminal visualizes the simulation results received from the server. For example, it uses the JavaScript library D3.js to draw a sales forecast graph and display it to the user, helping them to review strategies and decide on measures.

[0616] Specific examples

[0617] New product introduction simulation example

[0618] The user sets a scenario in which "New Product B will be introduced on June 1, 2023 at a price of 2,000 yen," and the device sends this scenario to the server. The server then runs a simulation within the virtual market and generates the results. Specific simulation results include sales after the introduction of New Product B, changes in market share with competing products, and the effectiveness of advertising campaigns. These results are displayed on the device as graphs and tables, allowing the user to plan marketing measures with high precision based on these results.

[0619] Examples of prompt statements

[0620] Below are some examples of prompt sentences to input into the generative AI model.

[0621] "You want to introduce new product B on June 1, 2023, at a price of 2,000 yen. Please conduct a market simulation and show the predicted changes in sales, profits, and market share after the introduction of new product B."

[0622] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0623] Step 1:

[0624] Users collect consumer data

[0625] Users collect consumer behavior data from a company's customer relationship management (CRM) system or marketing database. Specifically, they use SQL queries to extract consumer purchasing history and customer profiles from the database and save them as a CSV file. Input data includes consumer ID, purchased product, purchase date, payment amount, age, gender, region, etc. The output provides organized consumer behavior data.

[0626] Step 2:

[0627] The terminal displays a data entry screen.

[0628] The terminal displays a user interface for inputting consumer data. It allows the user to manually enter fields such as consumer ID, purchased product, purchase date, payment amount, age, gender, and region through a data input form. The input includes consumer behavior data collected by the user. The output includes the data entered by the user on the input screen.

[0629] Step 3:

[0630] User enters and submits consumer data

[0631] The user manually enters data into the input form on the device and presses the send button. The device converts the entered data into JSON format, generates an HTTP request, and sends it to the server. The input is consumption behavior data manually entered by the user. The output is the data converted into JSON format and sent to the server as an HTTP request.

[0632] Step 4:

[0633] The server receives and analyzes the data

[0634] The server receives JSON data sent from the device. It then performs data cleansing using the Python Pandas library. During this process, it performs tasks such as filling in missing data, handling outliers, and normalizing the data. The input is consumer behavior data in JSON format. The output is the cleansed data.

[0635] Step 5:

[0636] The server generates a virtual consumer model.

[0637] The server generates a virtual consumer model based on the preprocessed data. Specifically, it performs clustering using the Scikit-learn library to create consumer segments. The input is the cleansed consumer behavior data. The output is the generated virtual consumer model.

[0638] Step 6:

[0639] The server aggregates multiple virtual consumer models.

[0640] The server integrates the generated virtual consumer models to construct a virtual market. This virtual market includes information on product supply and demand, pricing, and competitors. The input is the multiple virtual consumer models. The output is the constructed virtual market.

[0641] Step 7:

[0642] The user sets up a business scenario

[0643] The user sets up a business scenario using the terminal interface. For example, they input a scenario such as "Introduce new product A on January 5, 2024 at a price of 1,500 yen" and send it to the server. The input includes detailed information about the business scenario. The output is the scenario sent to the server.

[0644] Step 8:

[0645] The server receives the scenario and runs the simulation.

[0646] The server receives business scenarios and runs simulations within a virtual market. It uses deep learning libraries such as TensorFlow and PyTorch to forecast sales, analyze the impact of price changes, and measure the effectiveness of advertising campaigns. The inputs are the business scenarios and virtual market data. The output is the simulation results.

[0647] Step 9:

[0648] The server generates the simulation results

[0649] The server generates the simulation results in JSON format. The results include statistical data such as sales, profit, market share, etc. As input, we have the detailed results of the simulation. As output, we get the simulation results in JSON format.

[0650] Step 10:

[0651] The terminal receives and displays the results

[0652] The terminal visualizes the simulation results received from the server. Using the JavaScript D3.js library, it draws a sales forecast graph and displays it to the user. The input is the simulation results in JSON format. The output is visualized data in graph and table format.

[0653] (Application example 1)

[0654] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0655] Conventional consumer behavior data analysis systems are limited to virtual market simulations, and have the problem of not being able to adequately predict and optimize for a variety of applications. Furthermore, for certain services, such as self-driving vehicles, there is a lack of systems that can analyze actual usage data, forecast demand, and provide optimal service routes. This has resulted in insufficient improvements in service efficiency and user satisfaction.

[0656] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0657] In this invention, the server includes means for receiving and analyzing real-world consumer behavior data, means for generating a virtual consumer model based on the analyzed data, means for integrating the generated virtual consumer model to build a virtual market, means for simulating business scenarios within the built virtual market, means for outputting simulation results, means for collecting and analyzing vehicle usage data, means for forecasting demand for on-demand services based on the usage data, and means for providing optimal service routes based on the demand forecast. This enables companies to use the results of simulations within the virtual market to forecast demand for various services and routes and to develop optimal strategies based on the forecasts.

[0658] "Actual consumer behavior data" refers to data on consumers' actual purchasing behavior and service usage. Specifically, it includes purchase history, usage history, customer profile, etc.

[0659] "Means of analysis" refers to the process of analyzing the received data and performing preprocessing such as information cleansing, handling outliers, and normalizing the data.

[0660] A "virtual consumer model" is a predictive model of consumer purchasing behavior and service usage behavior that is generated using machine learning algorithms based on real-world consumer behavior data.

[0661] A "virtual market" is a simulation environment that integrates multiple virtual consumer models and mimics a real market environment, including information on the supply and demand of each product or service, as well as competitor information.

[0662] A "business scenario" is a specific strategy or measure that a company is considering implementing. Examples include the introduction of a new product, a price change, or an advertising campaign.

[0663] "Simulation results" are statistical data such as sales forecasts, profit margins, and fluctuations in market share calculated as a result of executing a business scenario within a virtual market.

[0664] "Vehicle usage data" refers to data related to the usage history of autonomous vehicles and on-demand services, including ride history, fares, usage time, and user profiles.

[0665] "Demand forecasting for on-demand services" is the process of analyzing usage data to predict how much demand there will be for a particular service.

[0666] An "optimal service route" is a route or service plan that is optimal for a user and is provided based on a demand forecast.

[0667] This invention is a system for providing on-demand services provided by autonomous vehicles, which analyzes consumer behavior data of users and builds a virtual market to forecast demand for services and provide optimal routes. Below, we will explain in detail the processing of the program for realizing this system.

[0668] Data collection and analysis

[0669] First, the user collects information about their use of the autonomous vehicle service. This is done by extracting it from a CSV file or directly from a database. The terminal then displays a form for confirming and entering consumer data, and the user enters information such as user ID, route, usage time, payment amount, age, gender, and region. The entered data is converted to JSON format and sent to the server.

[0670] Creating a Virtual Consumer Model

[0671] The server receives the JSON data sent from the device and performs data cleansing and preprocessing. This includes filling in missing data, handling outliers, and normalizing the data. It then uses machine learning algorithms (such as clustering and regression models) to generate a virtual consumer model that predicts user behavior. The generated virtual consumer model includes the user's usage history, interests, and future usage predictions.

[0672] Building a virtual market

[0673] Next, the server integrates the generated virtual consumer models to create a virtual market. This virtual market mimics a real-world market environment, including the supply and demand of each product or service, as well as information on competitors. This makes it possible to carry out simulations within the virtual market.

[0674] Business scenario simulation

[0675] The user sets up a business scenario on the device, such as adding a new route or offering a discount service during a specific time period, and this scenario is input to the generative AI model as a prompt.

[0676] Prompt Sentence Examples

[0677] "New Route A will be introduced on October 1, 2023, with a discount campaign during the first month. We will provide demand and revenue forecasts after the introduction."

[0678] The server runs simulations within the virtual market to forecast demand, and the simulation results include data such as sales forecasts, personnel allocation, and optimal service routes.

[0679] Output of simulation results

[0680] Finally, the server generates the simulation results in JSON format and sends them to the terminal. The terminal visualizes the results received from the server and provides them to the user. The results are displayed in graphs and tables, allowing the user to review their strategy and make decisions based on them.

[0681] Hardware and software used

[0682] This system analyzes data using cloud servers (such as AWS or Google Cloud), primarily using MySQL or PostgreSQL as databases, and Python and libraries such as TensorFlow and Scikit-Learn for machine learning. The front end can be developed using frameworks such as React and Angular.

[0683] This system enables companies to use simulation results within a virtual market to forecast demand for various services and routes, and then develop optimal strategies based on that forecast.

[0684] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0685] Step 1:

[0686] Users collect usage data for the autonomous vehicle service. Specifically, information such as user ID, route, usage time, payment amount, age, gender, and region is collected using CSV files or database extraction. The collected data is then entered into a terminal.

[0687] Step 2:

[0688] The terminal displays a form for verifying and entering consumer data. The user enters the collected data into this form, confirms the input, and clicks the submit button. The input at this point is detailed information related to the user.

[0689] Step 3:

[0690] The terminal converts the input data into JSON format and sends it to the server. This converts the information entered by the user into a data structure and sends it via an HTTP request. The output is structured user data.

[0691] Step 4:

[0692] The server receives the JSON data sent from the terminal. After receiving it, the server performs data cleansing and preprocessing. Specifically, it completes missing data, processes outliers, and normalizes the data. The input is JSON data, and the output is the preprocessed, clean data.

[0693] Step 5:

[0694] The server generates a virtual consumer model based on the preprocessed data. Machine learning algorithms (e.g., clustering and regression models) are used to create a model that predicts consumption behavior, interests, and future behavior. The input is the preprocessed data, and the output is the virtual consumer model.

[0695] Step 6:

[0696] The server integrates multiple generated virtual consumer models to build a virtual market. This virtual market is a simulation environment that includes information on the supply and demand of each product or service, as well as information on competitors. The input is the virtual consumer model, and the output is the virtual market.

[0697] Step 7:

[0698] The user sets a business scenario on the terminal. For example, they set a specific prompt such as "New route A will be introduced on October 1, 2023, and a discount campaign will be held in the first month." The input is the business scenario, and the output is the prompt.

[0699] Step 8:

[0700] The server executes a simulation in the virtual market based on the received prompt. Specific simulation operations include demand forecasting, sales forecasting, personnel allocation, and calculation of optimal service routes after the introduction of new routes. The input is the prompt, and the output is the simulation results.

[0701] Step 9:

[0702] The server generates simulation results in JSON format. The simulation results include data such as demand forecasts, sales, and personnel allocation. These are sent to the terminal. The input is the simulation results, and the output is the result data in JSON format.

[0703] Step 10:

[0704] The terminal visualizes the simulation results received from the server and presents them to the user. The results are displayed in graphs and tables, and the user can use them to review strategies and make decisions. The input is the simulation results in JSON format, and the output is visualized data.

[0705] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0706] The system according to the present invention is characterized by the fact that it not only receives and analyzes real consumer behavior data, but also combines it with an emotion engine that recognizes user emotions to generate a more sophisticated virtual consumer model. Below, we will create a program for the system and explain its processing in detail.

[0707] Data entry and submission

[0708] 1. Users collect consumer data

[0709] Users collect consumer purchasing history, customer profiles, product information, etc. from companies' customer relationship management (CRM) systems and marketing databases, including consumer sentiment data, such as consumer sentiment and feedback comments when purchasing a product.

[0710] 2. The device displays the data entry screen.

[0711] The terminal provides a user interface and displays a form for the user to input the collected consumer data and emotion data, including the consumer ID, purchased item, purchase date, payment amount, age, gender, region, and emotion tag (e.g., joy, anger, sadness, etc.).

[0712] 3. The user enters and submits data

[0713] The user enters the required information into the data input form on the terminal and clicks the submit button. The terminal converts the entered data into JSON format and sends it to the server.

[0714] Generating the Consumer Model

[0715] 4. The server receives and analyzes the data

[0716] The server receives the JSON data sent from the device. After receiving the data, the server performs data cleansing, fills in missing values, and processes outliers. This also includes checking whether the emotion data is normal.

[0717] 5. The server generates a virtual consumer model

[0718] The server generates a virtual consumer model based on the analyzed consumer behavior data and emotion data. Specifically, it uses a machine learning algorithm to create a model that predicts the relationship between consumer purchasing behavior and emotions. The generated virtual consumer model includes the consumer's purchasing history, emotional state, interests, and future purchase predictions.

[0719] Building a virtual market

[0720] 6. The server aggregates multiple virtual consumer models

[0721] The server integrates multiple virtual consumer models to create a virtual market that mimics a real-world market environment, including supply and demand for each product or service, competitive information, and consumer sentiment data.

[0722] Business scenario simulation

[0723] 7. The user sets up the business scenario

[0724] The user sets up a business scenario on the device. For example, the scenario is set up as follows: "New product B will be introduced on June 1, 2023, at a price of 2,000 yen. Review sentiment at the time of introduction will be analyzed." The user enters detailed information about the scenario and sends it to the server.

[0725] 8. The server receives and analyzes the scenario.

[0726] The server analyzes the received scenario data and prepares for the simulation. Based on the set scenario, it prepares to predict fluctuations in the virtual market.

[0727] 9. The server runs the simulation

[0728] The server applies a scenario within the virtual market and starts a simulation. For example, it predicts sales from the introduction of new product B, market share fluctuations compared to competitors' products, and emotional reactions at the time of introduction. The emotion engine analyzes consumer reviews and feedback to determine positive or negative emotions.

[0729] Output of simulation results

[0730] 10. The server generates the simulation results

[0731] After the simulation is completed, the server aggregates the results and outputs them in JSON format, including sales forecasts, profit margins, market share fluctuation data, and sentiment analysis results.

[0732] 11. The terminal receives and displays the simulation results.

[0733] The terminal receives the simulation results sent from the server, and the received results are stored and displayed on the user interface.

[0734] 12. The terminal visualizes the results

[0735] The device visualizes the received results in graphs and tables and provides them to the user. Sentiment analysis results are displayed as the percentage of positive reactions, the percentage of negative reactions, etc. The user can analyze the results and revise their marketing strategies and business plans.

[0736] Specific examples

[0737] New product introduction scenario

[0738] The user sets up a scenario in which new product B will be introduced on June 1, 2023, at a price of 2,000 yen, and consumer review sentiment will be analyzed. The device sends this scenario to the server, which then runs a simulation within the virtual market. The simulation results, calculated as graphs on the device, include sales after the introduction of new product B, changes in market share compared to competing products, and sentiment analysis results based on consumer reviews. This allows the user to accurately predict new product introductions and changes in consumer sentiment, and plan optimal marketing measures.

[0739] The above is a specific embodiment for carrying out the present invention. This system enables companies to analyze and predict consumer behavior in detail, and to develop flexible marketing strategies that take consumer sentiment into account.

[0740] The processing flow will be explained below.

[0741] Step 1:

[0742] The user collects consumer data, such as consumer purchasing history, customer profile, product information, and emotional data, from a company's customer relationship management (CRM) system or marketing database. For example, data such as consumer ID, purchased product, purchase date, payment amount, age, gender, region, and emotional tag (happiness, anger, sadness, etc.) is collected.

[0743] Step 2:

[0744] The terminal displays a data entry screen. The form includes the consumer ID, purchased item, purchase date, payment amount, age, gender, region, and emotion tag (e.g., joy, anger, sadness, etc.). The terminal provides an interface for the user to enter the collected data.

[0745] Step 3:

[0746] The user enters data and submits it. The user manually enters the required information into the data input form on the terminal and clicks the submit button. The terminal converts the entered data into JSON format and sends it to the server.

[0747] Step 4:

[0748] The server receives and analyzes the data. The server receives the JSON data sent from the device. After receiving the data, the server cleanses the data, filling in missing values ​​and processing outliers. For example, it removes unnatural age data (such as 0 or 150 years old) and obviously incorrect emotion tags.

[0749] Step 5:

[0750] The server uses an emotion engine to analyze user emotions based on text, audio, or video input. For example, it analyzes consumer reviews and feedback comments using natural language processing (NLP) techniques and assigns positive, negative, or neutral emotion tags.

[0751] Step 6:

[0752] The server generates a virtual consumer model based on the analyzed consumer behavior data and emotion data. Specifically, a machine learning algorithm is used to create a model that predicts the relationship between consumer purchasing behavior and emotions. The generated virtual consumer model includes the consumer's purchasing history, emotional state, interests, and future purchase predictions.

[0753] Step 7:

[0754] The server integrates multiple virtual consumer models. The server then integrates the multiple virtual consumer models that have been generated to create a virtual market. This virtual market mimics a real market environment and includes information such as supply and demand for each product or service, competitive information, and consumer sentiment data.

[0755] Step 8:

[0756] The user sets up a business scenario. The user inputs the business scenario into the device. For example, the scenario is set as "New product B will be introduced on June 1, 2023, at a price of 2,000 yen. Review sentiment at this time will be analyzed." The device then sends this scenario to the server.

[0757] Step 9:

[0758] The server receives and analyzes the scenario. The server analyzes the received scenario data and prepares for the simulation. Based on the set scenario, the server prepares to predict fluctuations in the virtual market.

[0759] Step 10:

[0760] The server runs the simulation. The server applies a scenario within the virtual market and starts the simulation. For example, it predicts sales estimates for the introduction of new product B, changes in market share with competitors' products, and emotional reactions based on consumer reviews at the time of introduction. The emotion engine analyzes consumer reviews and feedback to determine positive or negative emotions.

[0761] Step 11:

[0762] The server generates the simulation results. After the simulation is completed, the server aggregates the results and outputs them in JSON format. The results include sales forecasts, profit margins, market share fluctuation data, as well as sentiment analysis results.

[0763] Step 12:

[0764] The terminal receives and displays the simulation results. The terminal receives the simulation results sent from the server. The received results are saved as they are and displayed on the user interface.

[0765] Step 13:

[0766] The device visualizes the results. The device visualizes the received results in graphs and tables and provides them to the user. Sentiment analysis results are displayed as the percentage of positive responses, the percentage of negative responses, etc. The user analyzes the results and reviews their marketing strategies and business plans.

[0767] Example 2

[0768] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0769] Simply analyzing real-world consumer behavior data is not enough to fully understand consumer emotions and the motivations behind their purchasing behavior. For this reason, there is a need for methods to generate more sophisticated virtual consumer models and predict consumer behavior with high accuracy. In particular, the insufficient analysis of emotional data is a challenge.

[0770] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0771] In this invention, the server includes means for receiving and analyzing real consumer behavior data, means for generating a virtual consumer model based on the analyzed data, and means for integrating the generated virtual consumer model to build a virtual market. This enables comprehensive analysis of consumer behavior and emotions by integrating means for recognizing and analyzing consumer emotions with means for refining the virtual consumer model based on emotion data, making it possible to make more accurate predictions and formulate marketing strategies.

[0772] "Consumer behavior data" refers to information about what products and services consumers have purchased, such as their purchasing history, customer profiles, and product information.

[0773] An "analytical means" is a method or device for extracting, processing, and analyzing data to understand its patterns and characteristics.

[0774] A "virtual consumer model" is a consumer behavior prediction model that is virtually created based on consumer behavior data, and includes the consumer's purchasing history, emotional state, interests, and future purchasing predictions.

[0775] A "virtual market" is a virtual environment that mimics a real market environment and is constructed by integrating virtual consumer models.

[0776] A "business scenario" is a scenario used to set assumptions for a company's marketing strategy or business plan, such as the introduction of a new product, price changes, or advertising campaigns.

[0777] A "simulating means" is a method or device for predicting fluctuations and results in a virtual market based on a set business scenario.

[0778] "Simulation results" are forecast results obtained by running a simulation based on a scenario, and include sales forecasts, profit margins, market share fluctuation data, and sentiment analysis results.

[0779] A "means for recognizing and analyzing sentiment" is a method or device for identifying positive or negative sentiment from consumer reviews and feedback comments and analyzing the data.

[0780] "Emotional data" is data that represents the emotions and feedback that consumers have about products and services.

[0781] A "refinement means" is a method or device for processing and improving existing data or models in more detail and accuracy.

[0782] The system according to the present invention is characterized by the fact that it not only receives and analyzes real consumer behavior data, but also combines it with an emotion engine that recognizes user emotions to generate a more sophisticated virtual consumer model. The processing of the system program is explained in detail below.

[0783] Hardware and software used

[0784] In this system, three entities, the server, the terminal, and the user, work together. The specific hardware and software used include the following:

[0785] Server: A high-performance data analysis server (e.g., AWS EC2, Google Cloud Platform, Microsoft Azure).

[0786] Device: A device such as a computer or smartphone used by a user.

[0787] Software: Machine learning algorithms (e.g., TensorFlow, PyTorch), database management systems (e.g., MySQL, PostgreSQL), sentiment analysis engines (e.g., IBM Watson, Microsoft Text Analytics).

[0788] Specific examples of system processing

[0789] 1. Users collect consumer data

[0790] Users collect consumer purchasing history, customer profiles, and product information from companies' customer relationship management (CRM) systems and marketing databases, including consumer sentiment and feedback comments when they purchase a product.

[0791] 2. The device displays the data entry screen.

[0792] The terminal displays a form for the user to enter the collected consumer and emotional data, including consumer ID, purchased item, purchase date, payment amount, age, gender, region, and emotional tag (e.g., joy, anger, sadness, etc.).

[0793] 3. The user enters and submits data

[0794] The user enters the required information into the data input form on the terminal and clicks the submit button. The terminal converts the entered data into JSON format and sends it to the server.

[0795] 4. The server receives and analyzes the data

[0796] The server receives the JSON data sent from the device. After receiving the data, the server cleanses the data, filling in missing values ​​and processing outliers. It verifies that the emotion data is normal and removes any invalid data.

[0797] 5. The server generates a virtual consumer model

[0798] The server uses a machine learning algorithm to generate a virtual consumer model based on the analyzed consumer behavior data and emotional data, which includes the consumer's purchasing history, emotional state, interests, and future purchasing predictions.

[0799] 6. The server aggregates multiple virtual consumer models

[0800] The server integrates the generated virtual consumer models to create a virtual market, which includes supply and demand for goods and services, competitive information, and consumer sentiment data.

[0801] Examples and prompts

[0802] For example, a user can set up a business scenario in which "New Product B will be introduced on June 1, 2023, at a price of 2,000 yen, and consumer review sentiment will be analyzed." The device sends this scenario to the server, which then runs a simulation within the virtual market. The results of the simulation are calculated and displayed as graphs on the device, including sales after the introduction of New Product B, changes in market share compared to competing products, and sentiment analysis results based on consumer reviews.

[0803] Prompt Sentence Examples

[0804] For example, consider the following prompt:

[0805] "New Product B will be introduced on June 1, 2023, at a price of 2,000 yen. Analyze review sentiment at the time of this introduction to see whether there are more positive or negative reactions."

[0806] The above is a specific embodiment for carrying out the present invention. This system enables companies to carry out detailed analysis and predictions that take into account consumer behavior and emotions, enabling the formulation of flexible marketing strategies.

[0807] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0808] Step 1:

[0809] Users collect consumer data

[0810] What it does: The user collects consumer purchasing history, customer profiles, product information, and sentiment data from the company's CRM system and marketing database.

[0811] Input: Consumer behavior and sentiment data from customer relationship management systems and marketing databases.

[0812] Output: The collected consumer behavior and sentiment data is prepared in a format for data input.

[0813] Step 2:

[0814] The terminal displays a data entry screen.

[0815] Specific behavior: The device provides a user interface and displays a form for inputting collected data, including consumer ID, purchased item, purchase date, payment amount, age, gender, region, and emotion tag (e.g., joy, anger, sadness, etc.).

[0816] Input: Consumer behavior and sentiment data collected by users.

[0817] Output: A data entry form is displayed on the terminal screen.

[0818] Step 3:

[0819] The user enters and submits data

[0820] Specific operation: The user enters the required information into the data input form on the terminal and clicks the submit button. The terminal converts the entered data into JSON format and sends it to the server.

[0821] Input: Consumer behavior and sentiment data entered into a data entry form.

[0822] Output: The data is converted to JSON format and sent to the server.

[0823] Step 4:

[0824] The server receives and analyzes the data

[0825] Specific operation: The server receives JSON data sent from the device. After receiving it, it performs data cleansing, fills in missing values, and processes outliers. It also checks whether the emotion data is normal and removes invalid data.

[0826] Input: JSON formatted consumer behavior and sentiment data sent from the device.

[0827] Output: Cleansed and accurate consumer behavior and sentiment data.

[0828] Step 5:

[0829] The server generates a virtual consumer model.

[0830] Specific operation: The server uses a machine learning algorithm to generate a virtual consumer model based on the analyzed consumer behavior data and emotion data. The generated model includes the consumer's purchase history, emotional state, interests, and future purchase predictions.

[0831] Input: Cleansed consumer behavior and sentiment data.

[0832] Output: A hypothetical consumer model.

[0833] Step 6:

[0834] The server aggregates multiple virtual consumer models.

[0835] Specific operation: The server integrates the generated multiple virtual consumer models to build a virtual market, which includes supply and demand for goods and services, competitive information, and consumer sentiment data.

[0836] Input: Multiple hypothetical consumer models.

[0837] Output: An integrated virtual marketplace.

[0838] Step 7:

[0839] The user sets up a business scenario

[0840] Specific operation: The user sets up a business scenario on the device. For example, they set up a scenario such as "Introduce new product B on June 1, 2023 at a price of 2,000 yen. Analyze review sentiment at the time of introduction." They then enter detailed scenario information and send it to the server.

[0841] Input: Business scenario (e.g. introduction of new product B, price, introduction date).

[0842] Output: Business scenario data sent to the server.

[0843] Step 8:

[0844] The server receives and analyzes the scenario.

[0845] Specific operation: The server analyzes the received scenario data and prepares for the simulation. Based on the set scenario, it prepares to predict fluctuations in the virtual market.

[0846] Input: Submitted business scenario data.

[0847] Output: Simulation ready.

[0848] Step 9:

[0849] The server runs the simulation

[0850] How it works: The server applies a scenario within the virtual market and starts a simulation. For example, it predicts sales from the introduction of new product B, changes in market share of competing products, and emotional reactions to the introduction. The emotion engine analyzes consumer reviews and feedback to determine positive or negative emotions.

[0851] Input: Simulation preparation data, review and feedback data to be analyzed for sentiment.

[0852] Output: Simulation results (sales forecast, market share change of competing products, predicted emotional response).

[0853] Step 10:

[0854] The server generates the simulation results

[0855] How it works: After the simulation is completed, the server aggregates the results and outputs them in JSON format, including sales forecasts, profit margins, market share fluctuation data, and sentiment analysis results.

[0856] Input: Various data after the simulation is run.

[0857] Output: Simulation results organized in JSON format.

[0858] Step 11:

[0859] The terminal receives and displays the simulation results.

[0860] Specific operation: The terminal receives the simulation results sent from the server, saves them, and displays them on the user interface.

[0861] Input: Simulation results sent from the server.

[0862] Output: Simulation results displayed on the terminal screen.

[0863] Step 12:

[0864] The terminal visualizes the results

[0865] Specific operation: The device visualizes the received results in graphs and tables and provides them to the user. Sentiment analysis results are displayed as the percentage of positive reactions, the percentage of negative reactions, etc. The user analyzes the results and revises their marketing strategy and business plan.

[0866] Input: Received simulation results.

[0867] Output: A visual summary in graphical and tabular form.

[0868] (Application example 2)

[0869] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0870] Modern purchasing behavior analysis does not adequately collect and analyze consumer emotional data, making it difficult to accurately grasp consumers' true needs and reactions. As a result, the accuracy of marketing strategies and product recommendations is limited, making it difficult to improve consumer satisfaction and maximize sales. There is a need to solve this problem, generate more accurate consumer models, and provide a system that can make optimal product recommendations.

[0871] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0872] In this invention, the server includes means for receiving and analyzing real consumer behavior data, means for generating a virtual consumer model based on the analyzed data, means for integrating the generated virtual consumer models to build a virtual market, means for collecting gaze and facial expression data and analyzing consumer emotions, means for recommending optimal products to users based on the analyzed emotion data, and means for outputting simulation results. This makes it possible to generate a sophisticated virtual consumer model based on consumer emotions, and to recommend optimal products and formulate marketing strategies.

[0873] "Actual consumer behavior data" refers to data such as the history, behavioral information, and payment information of consumers when they actually purchase products, and indicates the specific behavior of individual consumers.

[0874] "Means for analysis" refers to a device or program for analyzing collected data and extracting useful information.

[0875] A "virtual consumer model" is a model that is generated based on real consumer behavior data and mimics changes in consumer purchasing behavior and emotions.

[0876] The "means for integrating and constructing a virtual market" is a device or program for aggregating multiple virtual consumer models and generating a virtual market that mimics a real market environment.

[0877] A "means for simulating business scenarios" is a device or program for testing business plans such as the introduction of new products, price changes, and the implementation of information provision systems in a virtual market and predicting their effects.

[0878] "Gaze and facial expression data" refers to data that shows the eye movements and facial expressions of consumers when they view products, and is information used to analyze consumers' emotions and interests.

[0879] "Means for analyzing consumer emotions" refers to a device or program for recognizing and analyzing the emotional state of consumers based on gaze and facial expression data.

[0880] The "means for recommending optimal products to users" refers to a device or program that selects and suggests products that are likely to interest users based on analyzed emotional data and purchasing behavior data.

[0881] The "means for outputting the simulation results" is a device or program for displaying or reporting the results of the business scenario simulation.

[0882] This invention is a system that builds a virtual market by collecting and analyzing real-world consumer behavior data and consumer emotion data, and then generating and integrating a virtual consumer model based on that data. It also simulates business scenarios within that virtual market and outputs the simulation results. This system also includes functions to collect gaze and facial expression data, analyze consumer emotions, and recommend optimal products to users based on the analyzed emotion data.

[0883] 1. Data collection and transmission

[0884] The user wears the smart glasses and browses the products. The smart glasses' built-in camera and sensors collect the user's gaze and facial expression data in real time.

[0885] The collected data is converted into JSON format and sent to the server via the terminal.

[0886] 2. Data Reception and Analysis

[0887] The server receives the JSON data sent from the device and performs data cleansing on the received data, including filling in missing values ​​and processing outliers.

[0888] The server analyzes consumer emotions from gaze and facial expression data using software such as OpenCV, dlib, and the GazeTracking library.

[0889] 3. Creating a Virtual Consumer Model

[0890] The server generates a virtual consumer model based on the analyzed consumer behavior and emotion data, often using machine learning algorithms.

[0891] The generated virtual consumer model includes purchasing history, emotional state, interests, and future purchase predictions.

[0892] 4. Building a virtual market

[0893] The server integrates the generated virtual consumer models to create a virtual market, which includes supply and demand for each product or service, competitive information, and consumer sentiment data.

[0894] 5. Business scenario simulation

[0895] Users set up business scenarios such as new product introductions, price changes, and advertising campaigns on their terminals.

[0896] The server predicts and runs simulations of virtual market fluctuations based on set scenarios, resulting in forecasts of sales, profit margins, market share fluctuations, and consumer emotional responses.

[0897] 6. Output of simulation results

[0898] The server outputs the simulation results in JSON format, and the terminal receives and visualizes them.

[0899] The user reviews their marketing strategy and business plan based on the output results.

[0900] Specific examples

[0901] If a user wears smart glasses and smiles when looking at Product A, the application will analyze their facial expression and gaze and recommend "Recommended Product A." Below is a specific example of a prompt sentence to input to the generative AI model.

[0902] Example prompt sentence:

[0903] When a user wears smart glasses and smiles when looking at product A, the application should analyze their facial expressions and gaze and recommend the most suitable product in real time.

[0904] The above is a specific embodiment for carrying out the invention. This system allows for a precise understanding of consumer emotions, making it possible to recommend optimal products and formulate marketing strategies.

[0905] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0906] Step 1:

[0907] A user puts on the smart glasses and browses the products.

[0908] (input)

[0909] User gaze and facial expression data.

[0910] (process)

[0911] The smart glasses' built-in cameras and sensors collect the user's gaze and facial expression data.

[0912] (output)

[0913] Real-time gaze and facial expression data is captured.

[0914] Step 2:

[0915] The data collected by the device is converted into JSON format and sent to the server.

[0916] (input)

[0917] Gaze and facial expression data.

[0918] (process)

[0919] The terminal converts the data into JSON format and sends it to the server over the network.

[0920] (output)

[0921] Gaze and facial expression data in JSON format is sent to the server.

[0922] Step 3:

[0923] The server performs data cleansing on the data received.

[0924] (input)

[0925] Gaze and expression data in JSON format.

[0926] (process)

[0927] The server cleanses the data by imputing missing values ​​and handling outliers.

[0928] (output)

[0929] Cleansed gaze and facial expression data.

[0930] Step 4:

[0931] The server analyzes consumer emotions from gaze and facial expression data.

[0932] (input)

[0933] Cleansed gaze and facial expression data.

[0934] (process)

[0935] Emotions are analyzed using software such as OpenCV, dlib, and GazeTracking. Specifically, changes in gaze direction and facial expressions are detected to determine the emotional state.

[0936] (output)

[0937] Consumer sentiment data.

[0938] Step 5:

[0939] The server generates a virtual consumer model based on the analyzed data.

[0940] (input)

[0941] Analyzed consumer sentiment data and consumption behavior data.

[0942] (process)

[0943] Use machine learning algorithms to create models that predict the relationship between consumer purchasing behavior and emotions.

[0944] (output)

[0945] Virtual consumer model.

[0946] Step 6:

[0947] The server integrates the generated virtual consumer models and constructs a virtual market.

[0948] (input)

[0949] Multiple virtual consumer models.

[0950] (process)

[0951] The server integrates multiple virtual consumer models to create a virtual market, which includes supply and demand for each product or service, competitive information, and consumer sentiment data.

[0952] (output)

[0953] Hypothetical market data.

[0954] Step 7:

[0955] The user sets up a business scenario on the terminal.

[0956] (input)

[0957] Scenario data for new product introductions, price changes, advertising campaigns, etc.

[0958] (process)

[0959] The user uses the interface on the terminal to input detailed information about the scenario and transmits it to the server.

[0960] (output)

[0961] The scenario data is sent to the server.

[0962] Step 8:

[0963] The server analyzes the received scenario data and executes a simulation.

[0964] (input)

[0965] Scenario and hypothetical market data.

[0966] (process)

[0967] Based on a set scenario, a simulation is run to predict fluctuations in a virtual market.

[0968] (output)

[0969] Simulation result data.

[0970] Step 9:

[0971] The server outputs the simulation results in JSON format, and the terminal receives and visualizes them.

[0972] (input)

[0973] Simulation result data.

[0974] (process)

[0975] The server outputs the simulation results in JSON format, and the terminal receives them and visualizes them in graphs or tables.

[0976] (output)

[0977] Visualized simulation results.

[0978] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0979] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0980] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0981] [Third embodiment]

[0982] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0983] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0984] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0985] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0986] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0987] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0988] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0989] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0990] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0991] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0992] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0993] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0994] The system according to the present invention performs a series of operations: it receives and analyzes real-world consumer behavior data, generates virtual consumer models, integrates them to build a virtual market, simulates business scenarios within the built virtual market, and outputs the results. Below, we will create a program for the system and explain its processing in detail.

[0995] Data entry and submission

[0996] 1. Users collect consumer data

[0997] Users collect consumer purchasing history, customer profiles, product information, etc. from a company's customer relationship management (CRM) system or marketing database. This data collection is done by extracting data from CSV files or directly from the database.

[0998] 2. The device displays the data entry screen.

[0999] The terminal provides a user interface and displays a form for the user to review and enter the collected consumer data, including information such as consumer ID, purchased product, purchase date, payment amount, age, gender, and region.

[1000] 3. The user enters and submits consumer data

[1001] The user enters the required information into the data input form on the terminal and clicks the submit button. The terminal converts the entered data into JSON format and sends it to the server.

[1002] Generating the Consumer Model

[1003] 4. The server receives and analyzes the data

[1004] The server receives the JSON data sent from the terminal. Specifically, it is designed to receive the data as an HTTP request so that it can reach the server.

[1005] The server analyzes the received data and performs data cleansing and preprocessing, which includes imputing missing data, handling outliers, and normalizing the data.

[1006] 5. The server generates a virtual consumer model

[1007] The server generates a virtual consumer model based on the analyzed data. Here, machine learning algorithms (e.g., clustering and regression models) are used to create a model that predicts consumer purchasing behavior. The generated virtual consumer model includes the consumer's purchasing history, interests, and future purchase predictions.

[1008] Building a virtual market

[1009] 6. The server aggregates multiple virtual consumer models

[1010] The server integrates the generated virtual consumer models to create a virtual market, which mimics a real-world market environment and includes information on the supply and demand of each product or service, as well as information on competitors.

[1011] Business scenario simulation

[1012] 7. The user sets up the business scenario

[1013] The user sets up a business scenario on the device. For example, a scenario could be "Introduce new product B on June 1, 2023 at a price of 2,000 yen." The user enters detailed information about the scenario and sends it to the server.

[1014] 8. The server receives the scenario and runs the simulation.

[1015] The server receives the scenario and runs a simulation in the virtual market based on that scenario, predicting things like sales fluctuations due to new product introductions, demand changes due to price changes, and the effects of advertising campaigns.

[1016] The simulation results include sales forecasts, profit margins, and changes in market share.

[1017] Output of simulation results

[1018] 9. The server generates the simulation results

[1019] The server generates the simulation results in JSON format, which include statistics on sales, profits, and market share after the scenario is executed.

[1020] 10. The device receives and displays the results

[1021] The terminal visualizes the simulation results received from the server and provides them to the user. The results are displayed in graphs and tables, and the user can use them to review strategies and make decisions.

[1022] Specific examples

[1023] New product introduction simulation example

[1024] The user sets up a scenario in which new product B will be introduced on June 1, 2023, at a price of 2,000 yen. The device sends this scenario to the server, which then runs a simulation within the virtual market. The results of the simulation, such as sales after the introduction of new product B, changes in market share with competing products, and the effectiveness of advertising campaigns, are calculated and displayed as graphs on the device. This allows the user to accurately predict the likelihood of success in introducing the new product and plan optimal marketing measures.

[1025] The above is a specific embodiment for carrying out the present invention. This system enables companies to perform detailed analysis and prediction of consumer behavior, and can be used as a tool to quickly respond to changing market conditions.

[1026] The processing flow will be explained below.

[1027] Step 1:

[1028] The user collects consumer data, such as consumer purchase history, customer profile, and product information, from a company's customer relationship management (CRM) system or marketing database. This data includes, for example, consumer ID, purchased product, purchase date, payment amount, age, gender, and region.

[1029] Step 2:

[1030] The terminal displays a data entry screen. The form includes fields for inputting information such as consumer ID, purchased item, purchase date, payment amount, age, gender, and region. The terminal provides an interface for the user to enter the collected data.

[1031] Step 3:

[1032] The user enters and submits consumer data. The user manually enters the required information into the data input form on the terminal and clicks the submit button. The terminal converts the entered data into JSON format and sends it to the server.

[1033] Step 4:

[1034] The server receives and analyzes the data. The server receives the JSON data sent from the device. After receiving the data, the server cleanses the data, filling in missing values ​​and processing outliers. For example, it excludes data with ages of 0 or 150.

[1035] Step 5:

[1036] The server generates a virtual consumer model based on the analyzed data. Specifically, it uses a machine learning algorithm to predict consumer purchasing behavior. For example, it uses a clustering algorithm to segment consumers and extract purchasing behavior patterns.

[1037] Step 6:

[1038] The server integrates multiple virtual consumer models. The server integrates the generated multiple virtual consumer models into a virtual market. This allows the virtual market to mimic a real market environment, including supply and demand for goods and services, as well as competitive information.

[1039] Step 7:

[1040] The user sets up a business scenario. The user inputs the business scenario into the terminal. For example, the user sets up a scenario such as "Introduce new product B on June 1, 2023 at a price of 2,000 yen." The terminal then sends this scenario to the server.

[1041] Step 8:

[1042] The server receives and analyzes the scenario. The server analyzes the received scenario data and prepares for the simulation. Based on the set scenario, the server prepares to predict fluctuations in the virtual market.

[1043] Step 9:

[1044] The server runs the simulation. The server applies a scenario within the virtual market and starts the simulation. For example, it calculates results such as estimated sales from the introduction of new product B, changes in market share with competitors' products, and the effectiveness of advertising campaigns.

[1045] Step 10:

[1046] The server generates the simulation results. After the simulation is completed, the server aggregates the results and outputs them in JSON format. The results include sales forecasts, profit margins, and market share fluctuation data.

[1047] Step 11:

[1048] The terminal receives the simulation results. The terminal receives the simulation results sent from the server. The received results are stored as they are and displayed on the user interface.

[1049] Step 12:

[1050] The terminal visualizes the results. The terminal visualizes the received results in graphs and tables and provides them to the user. The user analyzes the results and reviews their marketing strategies and business plans.

[1051] The above is a concrete processing flow of the system according to the present invention, which provides businesses with an effective tool for detailed analysis and prediction of consumer behavior and for quickly responding to changing market conditions.

[1052] Example 1

[1053] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1054] In a real market environment, it is difficult to effectively utilize consumer behavior data, quickly generate a virtual consumer model, and then simulate business scenarios based on that model. With conventional systems, data preparation, model generation, scenario setting, and simulation are all performed separately, which is extremely time-consuming and labor-intensive. Furthermore, when companies introduce new products, change prices, or plan advertising campaigns, there is no effective system that can comprehensively simulate these elements and quickly visualize and provide the results.

[1055] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1056] In this invention, the server includes means for receiving JSON data sent from the terminal, performing data cleansing and preprocessing, and generating a virtual consumer model; means for the server to integrate the generated virtual consumer model and build a virtual market; and means for the server to run a simulation within the virtual market based on a set scenario. This enables companies to quickly and effectively generate a virtual consumer model using consumer behavior data and simulate business scenarios in the virtual market. Furthermore, the simulation results can be quickly visualized to support strategy revisions and policy decisions.

[1057] "User" refers to an individual or corporation that uses the system to collect, input, and manage consumer behavior data and set scenarios.

[1058] "Device" refers to a computer device used by a user to input consumption behavior data and set up a scenario. Examples include PCs, tablets, and smartphones.

[1059] "Server" refers to a computer system that receives, analyzes, and processes data sent from terminals, generates virtual consumer models and virtual markets, and executes simulations.

[1060] "Consumer behavior data" refers to data that includes information about consumer behavior, such as consumer purchasing history, customer profiles, and product information.

[1061] "JSON" is an abbreviation for JavaScript Object Notation, and refers to a lightweight data format for data exchange. It represents structured data in a way that is easy for humans to read and machines to parse.

[1062] "Data cleansing" refers to the process of improving data quality by completing missing values, processing outliers, normalizing data, etc. as a preprocessing step for data analysis.

[1063] A "virtual consumer model" refers to a model that includes a consumer's purchasing history, interests, and future purchasing predictions, and is generated using a machine learning algorithm based on consumer behavior data.

[1064] A "virtual market" refers to a simulation environment that integrates multiple virtual consumer models and mimics a real market environment, including product supply and demand, pricing strategies, and competitor information.

[1065] A "business scenario" is a plan or strategy that a company sets up to simulate in a virtual market, including new product introductions, price changes, advertising campaigns, etc.

[1066] "Simulation" refers to the process of forecasting sales, analyzing the impact of price changes, measuring the effectiveness of advertising campaigns, etc., based on business scenarios set up within a virtual marketplace.

[1067] "Simulation results" refers to statistical data such as sales, profits, and market share obtained after executing a business scenario.

[1068] The system according to the present invention performs a series of operations: collects and analyzes real-world consumer behavior data to generate a virtual consumer model, integrates the data to build a virtual market, simulates business scenarios within the virtual market, and outputs the results. A specific embodiment of this system is described below.

[1069] Data entry and submission

[1070] Users collect consumer data

[1071] Users collect consumer behavior data from corporate customer relationship management (CRM) systems and marketing databases. Specifically, they use SQL queries to extract consumer purchasing histories and customer profiles from the databases and save them as CSV files.

[1072] The terminal displays a data entry screen.

[1073] The terminal displays a user interface for entering consumer data. The terminal can be a computer device such as a PC, tablet, or smartphone. The data entry form includes fields such as consumer ID, product purchased, purchase date, payment amount, age, gender, and region.

[1074] User enters and submits consumer data

[1075] The user manually enters data into the input form on the device and presses the submit button. The device converts the entered data into JSON format, generates an HTTP request, and sends it to the server. This is often done using a programming language such as JavaScript.

[1076] Generating the Consumer Model

[1077] The server receives and analyzes the data

[1078] The server receives the JSON data sent from the terminal and performs data cleansing using Python's Pandas library, among other tools. During this process, the server performs tasks such as filling in missing data, handling outliers, and normalizing the data.

[1079] The server generates a virtual consumer model.

[1080] The server generates a virtual consumer model based on the cleansed data. Specifically, it uses the machine learning library Scikit-learn to perform clustering and create a model that predicts consumer purchasing behavior.

[1081] Building a virtual market

[1082] The server aggregates multiple virtual consumer models.

[1083] The server integrates the generated virtual consumer models to create a virtual market, which includes information on product supply and demand, pricing, and competitors.

[1084] Business scenario simulation

[1085] The user sets up a business scenario

[1086] The user sets up a business scenario using the terminal interface. For example, the scenario "New product A will be introduced on January 5, 2024 at a price of 1,500 yen" is input and sent to the server.

[1087] The server receives the scenario and runs the simulation.

[1088] The server receives business scenarios and runs simulations within the virtual marketplace, which can include sales forecasting, analyzing the impact of price changes, and measuring the effectiveness of advertising campaigns, often using deep learning libraries such as TensorFlow and PyTorch.

[1089] Output of simulation results

[1090] The server generates the simulation results

[1091] The server generates simulation results in JSON format, including statistical data such as sales, profits, and market share.

[1092] The terminal receives and displays the results

[1093] The terminal visualizes the simulation results received from the server. For example, it uses the JavaScript library D3.js to draw a sales forecast graph and display it to the user, helping them to review strategies and decide on measures.

[1094] Specific examples

[1095] New product introduction simulation example

[1096] The user sets a scenario in which "New Product B will be introduced on June 1, 2023 at a price of 2,000 yen," and the device sends this scenario to the server. The server then runs a simulation within the virtual market and generates the results. Specific simulation results include sales after the introduction of New Product B, changes in market share with competing products, and the effectiveness of advertising campaigns. These results are displayed on the device as graphs and tables, allowing the user to plan marketing measures with high precision based on these results.

[1097] Examples of prompt statements

[1098] Below are some examples of prompt sentences to input into the generative AI model.

[1099] "You want to introduce new product B on June 1, 2023, at a price of 2,000 yen. Please conduct a market simulation and show the predicted changes in sales, profits, and market share after the introduction of new product B."

[1100] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1101] Step 1:

[1102] Users collect consumer data

[1103] Users collect consumer behavior data from a company's customer relationship management (CRM) system or marketing database. Specifically, they use SQL queries to extract consumer purchasing history and customer profiles from the database and save them as a CSV file. Input data includes consumer ID, purchased product, purchase date, payment amount, age, gender, region, etc. The output provides organized consumer behavior data.

[1104] Step 2:

[1105] The terminal displays a data entry screen.

[1106] The terminal displays a user interface for inputting consumer data. It allows the user to manually enter fields such as consumer ID, purchased product, purchase date, payment amount, age, gender, and region through a data input form. The input includes consumer behavior data collected by the user. The output includes the data entered by the user on the input screen.

[1107] Step 3:

[1108] User enters and submits consumer data

[1109] The user manually enters data into the input form on the device and presses the send button. The device converts the entered data into JSON format, generates an HTTP request, and sends it to the server. The input is consumption behavior data manually entered by the user. The output is the data converted into JSON format and sent to the server as an HTTP request.

[1110] Step 4:

[1111] The server receives and analyzes the data

[1112] The server receives JSON data sent from the device. It then performs data cleansing using the Python Pandas library. During this process, it performs tasks such as filling in missing data, handling outliers, and normalizing the data. The input is consumer behavior data in JSON format. The output is the cleansed data.

[1113] Step 5:

[1114] The server generates a virtual consumer model.

[1115] The server generates a virtual consumer model based on the preprocessed data. Specifically, it performs clustering using the Scikit-learn library to create consumer segments. The input is the cleansed consumer behavior data. The output is the generated virtual consumer model.

[1116] Step 6:

[1117] The server aggregates multiple virtual consumer models.

[1118] The server integrates the generated virtual consumer models to construct a virtual market. This virtual market includes information on product supply and demand, pricing, and competitors. The input is the multiple virtual consumer models. The output is the constructed virtual market.

[1119] Step 7:

[1120] The user sets up a business scenario

[1121] The user sets up a business scenario using the terminal interface. For example, they input a scenario such as "Introduce new product A on January 5, 2024 at a price of 1,500 yen" and send it to the server. The input includes detailed information about the business scenario. The output is the scenario sent to the server.

[1122] Step 8:

[1123] The server receives the scenario and runs the simulation.

[1124] The server receives business scenarios and runs simulations within a virtual market. It uses deep learning libraries such as TensorFlow and PyTorch to forecast sales, analyze the impact of price changes, and measure the effectiveness of advertising campaigns. The inputs are the business scenarios and virtual market data. The output is the simulation results.

[1125] Step 9:

[1126] The server generates the simulation results

[1127] The server generates the simulation results in JSON format. The results include statistical data such as sales, profit, market share, etc. As input, we have the detailed results of the simulation. As output, we get the simulation results in JSON format.

[1128] Step 10:

[1129] The terminal receives and displays the results

[1130] The terminal visualizes the simulation results received from the server. Using the JavaScript D3.js library, it draws a sales forecast graph and displays it to the user. The input is the simulation results in JSON format. The output is visualized data in graph and table format.

[1131] (Application example 1)

[1132] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1133] Conventional consumer behavior data analysis systems are limited to virtual market simulations, and have the problem of not being able to adequately predict and optimize for a variety of applications. Furthermore, for certain services, such as self-driving vehicles, there is a lack of systems that can analyze actual usage data, forecast demand, and provide optimal service routes. This has resulted in insufficient improvements in service efficiency and user satisfaction.

[1134] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1135] In this invention, the server includes means for receiving and analyzing real-world consumer behavior data, means for generating a virtual consumer model based on the analyzed data, means for integrating the generated virtual consumer model to build a virtual market, means for simulating business scenarios within the built virtual market, means for outputting simulation results, means for collecting and analyzing vehicle usage data, means for forecasting demand for on-demand services based on the usage data, and means for providing optimal service routes based on the demand forecast. This enables companies to use the results of simulations within the virtual market to forecast demand for various services and routes and to develop optimal strategies based on the forecasts.

[1136] "Actual consumer behavior data" refers to data on consumers' actual purchasing behavior and service usage. Specifically, it includes purchase history, usage history, customer profile, etc.

[1137] "Means of analysis" refers to the process of analyzing the received data and performing preprocessing such as information cleansing, handling outliers, and normalizing the data.

[1138] A "virtual consumer model" is a predictive model of consumer purchasing behavior and service usage behavior that is generated using machine learning algorithms based on real-world consumer behavior data.

[1139] A "virtual market" is a simulation environment that integrates multiple virtual consumer models and mimics a real market environment, including information on the supply and demand of each product or service, as well as competitor information.

[1140] A "business scenario" is a specific strategy or measure that a company is considering implementing. Examples include the introduction of a new product, a price change, or an advertising campaign.

[1141] "Simulation results" are statistical data such as sales forecasts, profit margins, and fluctuations in market share calculated as a result of executing a business scenario within a virtual market.

[1142] "Vehicle usage data" refers to data related to the usage history of autonomous vehicles and on-demand services, including ride history, fares, usage time, and user profiles.

[1143] "Demand forecasting for on-demand services" is the process of analyzing usage data to predict how much demand there will be for a particular service.

[1144] An "optimal service route" is a route or service plan that is optimal for a user and is provided based on a demand forecast.

[1145] This invention is a system for providing on-demand services provided by autonomous vehicles, which analyzes consumer behavior data of users and builds a virtual market to forecast demand for services and provide optimal routes. Below, we will explain in detail the processing of the program for realizing this system.

[1146] Data collection and analysis

[1147] First, the user collects information about their use of the autonomous vehicle service. This is done by extracting it from a CSV file or directly from a database. The terminal then displays a form for confirming and entering consumer data, and the user enters information such as user ID, route, usage time, payment amount, age, gender, and region. The entered data is converted to JSON format and sent to the server.

[1148] Creating a Virtual Consumer Model

[1149] The server receives the JSON data sent from the device and performs data cleansing and preprocessing. This includes filling in missing data, handling outliers, and normalizing the data. It then uses machine learning algorithms (such as clustering and regression models) to generate a virtual consumer model that predicts user behavior. The generated virtual consumer model includes the user's usage history, interests, and future usage predictions.

[1150] Building a virtual market

[1151] Next, the server integrates the generated virtual consumer models to create a virtual market. This virtual market mimics a real-world market environment, including the supply and demand of each product or service, as well as information on competitors. This makes it possible to carry out simulations within the virtual market.

[1152] Business scenario simulation

[1153] The user sets up a business scenario on the device, such as adding a new route or offering a discount service during a specific time period, and this scenario is input to the generative AI model as a prompt.

[1154] Prompt Sentence Examples

[1155] "New Route A will be introduced on October 1, 2023, with a discount campaign during the first month. We will provide demand and revenue forecasts after the introduction."

[1156] The server runs simulations within the virtual market to forecast demand, and the simulation results include data such as sales forecasts, personnel allocation, and optimal service routes.

[1157] Output of simulation results

[1158] Finally, the server generates the simulation results in JSON format and sends them to the terminal. The terminal visualizes the results received from the server and provides them to the user. The results are displayed in graphs and tables, allowing the user to review their strategy and make decisions based on them.

[1159] Hardware and software used

[1160] This system analyzes data using cloud servers (such as AWS or Google Cloud), primarily using MySQL or PostgreSQL as databases, and Python and libraries such as TensorFlow and Scikit-Learn for machine learning. The front end can be developed using frameworks such as React and Angular.

[1161] This system enables companies to use simulation results within a virtual market to forecast demand for various services and routes, and then develop optimal strategies based on that forecast.

[1162] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1163] Step 1:

[1164] Users collect usage data for the autonomous vehicle service. Specifically, information such as user ID, route, usage time, payment amount, age, gender, and region is collected using CSV files or database extraction. The collected data is then entered into a terminal.

[1165] Step 2:

[1166] The terminal displays a form for verifying and entering consumer data. The user enters the collected data into this form, confirms the input, and clicks the submit button. The input at this point is detailed information related to the user.

[1167] Step 3:

[1168] The terminal converts the input data into JSON format and sends it to the server. This converts the information entered by the user into a data structure and sends it via an HTTP request. The output is structured user data.

[1169] Step 4:

[1170] The server receives the JSON data sent from the terminal. After receiving it, the server performs data cleansing and preprocessing. Specifically, it completes missing data, processes outliers, and normalizes the data. The input is JSON data, and the output is the preprocessed, clean data.

[1171] Step 5:

[1172] The server generates a virtual consumer model based on the preprocessed data. Machine learning algorithms (e.g., clustering and regression models) are used to create a model that predicts consumption behavior, interests, and future behavior. The input is the preprocessed data, and the output is the virtual consumer model.

[1173] Step 6:

[1174] The server integrates multiple generated virtual consumer models to build a virtual market. This virtual market is a simulation environment that includes information on the supply and demand of each product or service, as well as information on competitors. The input is the virtual consumer model, and the output is the virtual market.

[1175] Step 7:

[1176] The user sets a business scenario on the terminal. For example, they set a specific prompt such as "New route A will be introduced on October 1, 2023, and a discount campaign will be held in the first month." The input is the business scenario, and the output is the prompt.

[1177] Step 8:

[1178] The server executes a simulation in the virtual market based on the received prompt. Specific simulation operations include demand forecasting, sales forecasting, personnel allocation, and calculation of optimal service routes after the introduction of new routes. The input is the prompt, and the output is the simulation results.

[1179] Step 9:

[1180] The server generates simulation results in JSON format. The simulation results include data such as demand forecasts, sales, and personnel allocation. These are sent to the terminal. The input is the simulation results, and the output is the result data in JSON format.

[1181] Step 10:

[1182] The terminal visualizes the simulation results received from the server and presents them to the user. The results are displayed in graphs and tables, and the user can use them to review strategies and make decisions. The input is the simulation results in JSON format, and the output is visualized data.

[1183] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1184] The system according to the present invention is characterized by the fact that it not only receives and analyzes real consumer behavior data, but also combines it with an emotion engine that recognizes user emotions to generate a more sophisticated virtual consumer model. Below, we will create a program for the system and explain its processing in detail.

[1185] Data entry and submission

[1186] 1. Users collect consumer data

[1187] Users collect consumer purchasing history, customer profiles, product information, etc. from companies' customer relationship management (CRM) systems and marketing databases, including consumer sentiment data, such as consumer sentiment and feedback comments when purchasing a product.

[1188] 2. The device displays the data entry screen.

[1189] The terminal provides a user interface and displays a form for the user to input the collected consumer data and emotion data, including the consumer ID, purchased item, purchase date, payment amount, age, gender, region, and emotion tag (e.g., joy, anger, sadness, etc.).

[1190] 3. The user enters and submits data

[1191] The user enters the required information into the data input form on the terminal and clicks the submit button. The terminal converts the entered data into JSON format and sends it to the server.

[1192] Generating the Consumer Model

[1193] 4. The server receives and analyzes the data

[1194] The server receives the JSON data sent from the device. After receiving the data, the server performs data cleansing, fills in missing values, and processes outliers. This also includes checking whether the emotion data is normal.

[1195] 5. The server generates a virtual consumer model

[1196] The server generates a virtual consumer model based on the analyzed consumer behavior data and emotion data. Specifically, it uses a machine learning algorithm to create a model that predicts the relationship between consumer purchasing behavior and emotions. The generated virtual consumer model includes the consumer's purchasing history, emotional state, interests, and future purchase predictions.

[1197] Building a virtual market

[1198] 6. The server aggregates multiple virtual consumer models

[1199] The server integrates multiple virtual consumer models to create a virtual market that mimics a real-world market environment, including supply and demand for each product or service, competitive information, and consumer sentiment data.

[1200] Business scenario simulation

[1201] 7. The user sets up the business scenario

[1202] The user sets up a business scenario on the device. For example, the scenario is set up as follows: "New product B will be introduced on June 1, 2023, at a price of 2,000 yen. Review sentiment at the time of introduction will be analyzed." The user enters detailed information about the scenario and sends it to the server.

[1203] 8. The server receives and analyzes the scenario.

[1204] The server analyzes the received scenario data and prepares for the simulation. Based on the set scenario, it prepares to predict fluctuations in the virtual market.

[1205] 9. The server runs the simulation

[1206] The server applies a scenario within the virtual market and starts a simulation. For example, it predicts sales from the introduction of new product B, market share fluctuations compared to competitors' products, and emotional reactions at the time of introduction. The emotion engine analyzes consumer reviews and feedback to determine positive or negative emotions.

[1207] Output of simulation results

[1208] 10. The server generates the simulation results

[1209] After the simulation is completed, the server aggregates the results and outputs them in JSON format, including sales forecasts, profit margins, market share fluctuation data, and sentiment analysis results.

[1210] 11. The terminal receives and displays the simulation results.

[1211] The terminal receives the simulation results sent from the server, and the received results are stored and displayed on the user interface.

[1212] 12. The terminal visualizes the results

[1213] The device visualizes the received results in graphs and tables and provides them to the user. Sentiment analysis results are displayed as the percentage of positive reactions, the percentage of negative reactions, etc. The user can analyze the results and revise their marketing strategies and business plans.

[1214] Specific examples

[1215] New product introduction scenario

[1216] The user sets up a scenario in which new product B will be introduced on June 1, 2023, at a price of 2,000 yen, and consumer review sentiment will be analyzed. The device sends this scenario to the server, which then runs a simulation within the virtual market. The simulation results, calculated as graphs on the device, include sales after the introduction of new product B, changes in market share compared to competing products, and sentiment analysis results based on consumer reviews. This allows the user to accurately predict new product introductions and changes in consumer sentiment, and plan optimal marketing measures.

[1217] The above is a specific embodiment for carrying out the present invention. This system enables companies to analyze and predict consumer behavior in detail, and to develop flexible marketing strategies that take consumer sentiment into account.

[1218] The processing flow will be explained below.

[1219] Step 1:

[1220] The user collects consumer data, such as consumer purchasing history, customer profile, product information, and emotional data, from a company's customer relationship management (CRM) system or marketing database. For example, data such as consumer ID, purchased product, purchase date, payment amount, age, gender, region, and emotional tag (happiness, anger, sadness, etc.) is collected.

[1221] Step 2:

[1222] The terminal displays a data entry screen. The form includes the consumer ID, purchased item, purchase date, payment amount, age, gender, region, and emotion tag (e.g., joy, anger, sadness, etc.). The terminal provides an interface for the user to enter the collected data.

[1223] Step 3:

[1224] The user enters data and submits it. The user manually enters the required information into the data input form on the terminal and clicks the submit button. The terminal converts the entered data into JSON format and sends it to the server.

[1225] Step 4:

[1226] The server receives and analyzes the data. The server receives the JSON data sent from the device. After receiving the data, the server cleanses the data, filling in missing values ​​and processing outliers. For example, it removes unnatural age data (such as 0 or 150 years old) and obviously incorrect emotion tags.

[1227] Step 5:

[1228] The server uses an emotion engine to analyze user emotions based on text, audio, or video input. For example, it analyzes consumer reviews and feedback comments using natural language processing (NLP) techniques and assigns positive, negative, or neutral emotion tags.

[1229] Step 6:

[1230] The server generates a virtual consumer model based on the analyzed consumer behavior data and emotion data. Specifically, a machine learning algorithm is used to create a model that predicts the relationship between consumer purchasing behavior and emotions. The generated virtual consumer model includes the consumer's purchasing history, emotional state, interests, and future purchase predictions.

[1231] Step 7:

[1232] The server integrates multiple virtual consumer models. The server then integrates the multiple virtual consumer models that have been generated to create a virtual market. This virtual market mimics a real market environment and includes information such as supply and demand for each product or service, competitive information, and consumer sentiment data.

[1233] Step 8:

[1234] The user sets up a business scenario. The user inputs the business scenario into the device. For example, the scenario is set as "New product B will be introduced on June 1, 2023, at a price of 2,000 yen. Review sentiment at this time will be analyzed." The device then sends this scenario to the server.

[1235] Step 9:

[1236] The server receives and analyzes the scenario. The server analyzes the received scenario data and prepares for the simulation. Based on the set scenario, the server prepares to predict fluctuations in the virtual market.

[1237] Step 10:

[1238] The server runs the simulation. The server applies a scenario within the virtual market and starts the simulation. For example, it predicts sales estimates for the introduction of new product B, changes in market share with competitors' products, and emotional reactions based on consumer reviews at the time of introduction. The emotion engine analyzes consumer reviews and feedback to determine positive or negative emotions.

[1239] Step 11:

[1240] The server generates the simulation results. After the simulation is completed, the server aggregates the results and outputs them in JSON format. The results include sales forecasts, profit margins, market share fluctuation data, as well as sentiment analysis results.

[1241] Step 12:

[1242] The terminal receives and displays the simulation results. The terminal receives the simulation results sent from the server. The received results are saved as they are and displayed on the user interface.

[1243] Step 13:

[1244] The device visualizes the results. The device visualizes the received results in graphs and tables and provides them to the user. Sentiment analysis results are displayed as the percentage of positive responses, the percentage of negative responses, etc. The user analyzes the results and reviews their marketing strategies and business plans.

[1245] Example 2

[1246] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1247] Simply analyzing real-world consumer behavior data is not enough to fully understand consumer emotions and the motivations behind their purchasing behavior. For this reason, there is a need for methods to generate more sophisticated virtual consumer models and predict consumer behavior with high accuracy. In particular, the insufficient analysis of emotional data is a challenge.

[1248] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1249] In this invention, the server includes means for receiving and analyzing real consumer behavior data, means for generating a virtual consumer model based on the analyzed data, and means for integrating the generated virtual consumer model to build a virtual market. This enables comprehensive analysis of consumer behavior and emotions by integrating means for recognizing and analyzing consumer emotions with means for refining the virtual consumer model based on emotion data, making it possible to make more accurate predictions and formulate marketing strategies.

[1250] "Consumer behavior data" refers to information about what products and services consumers have purchased, such as their purchasing history, customer profiles, and product information.

[1251] An "analytical means" is a method or device for extracting, processing, and analyzing data to understand its patterns and characteristics.

[1252] A "virtual consumer model" is a consumer behavior prediction model that is virtually created based on consumer behavior data, and includes the consumer's purchasing history, emotional state, interests, and future purchasing predictions.

[1253] A "virtual market" is a virtual environment that mimics a real market environment and is constructed by integrating virtual consumer models.

[1254] A "business scenario" is a scenario used to set assumptions for a company's marketing strategy or business plan, such as the introduction of a new product, price changes, or advertising campaigns.

[1255] A "simulating means" is a method or device for predicting fluctuations and results in a virtual market based on a set business scenario.

[1256] "Simulation results" are forecast results obtained by running a simulation based on a scenario, and include sales forecasts, profit margins, market share fluctuation data, and sentiment analysis results.

[1257] A "means for recognizing and analyzing sentiment" is a method or device for identifying positive or negative sentiment from consumer reviews and feedback comments and analyzing the data.

[1258] "Emotional data" is data that represents the emotions and feedback that consumers have about products and services.

[1259] A "refinement means" is a method or device for processing and improving existing data or models in more detail and accuracy.

[1260] The system according to the present invention is characterized by the fact that it not only receives and analyzes real consumer behavior data, but also combines it with an emotion engine that recognizes user emotions to generate a more sophisticated virtual consumer model. The processing of the system program is explained in detail below.

[1261] Hardware and software used

[1262] In this system, three entities, the server, the terminal, and the user, work together. The specific hardware and software used include the following:

[1263] Server: A high-performance data analysis server (e.g., AWS EC2, Google Cloud Platform, Microsoft Azure).

[1264] Device: A device such as a computer or smartphone used by a user.

[1265] Software: Machine learning algorithms (e.g., TensorFlow, PyTorch), database management systems (e.g., MySQL, PostgreSQL), sentiment analysis engines (e.g., IBM Watson, Microsoft Text Analytics).

[1266] Specific examples of system processing

[1267] 1. Users collect consumer data

[1268] Users collect consumer purchasing history, customer profiles, and product information from companies' customer relationship management (CRM) systems and marketing databases, including consumer sentiment and feedback comments when they purchase a product.

[1269] 2. The device displays the data entry screen.

[1270] The terminal displays a form for the user to enter the collected consumer and emotional data, including consumer ID, purchased item, purchase date, payment amount, age, gender, region, and emotional tag (e.g., joy, anger, sadness, etc.).

[1271] 3. The user enters and submits data

[1272] The user enters the required information into the data input form on the terminal and clicks the submit button. The terminal converts the entered data into JSON format and sends it to the server.

[1273] 4. The server receives and analyzes the data

[1274] The server receives the JSON data sent from the device. After receiving the data, the server cleanses the data, filling in missing values ​​and processing outliers. It verifies that the emotion data is normal and removes any invalid data.

[1275] 5. The server generates a virtual consumer model

[1276] The server uses a machine learning algorithm to generate a virtual consumer model based on the analyzed consumer behavior data and emotional data, which includes the consumer's purchasing history, emotional state, interests, and future purchasing predictions.

[1277] 6. The server aggregates multiple virtual consumer models

[1278] The server integrates the generated virtual consumer models to create a virtual market, which includes supply and demand for goods and services, competitive information, and consumer sentiment data.

[1279] Examples and prompts

[1280] For example, a user can set up a business scenario in which "New Product B will be introduced on June 1, 2023, at a price of 2,000 yen, and consumer review sentiment will be analyzed." The device sends this scenario to the server, which then runs a simulation within the virtual market. The results of the simulation are calculated and displayed as graphs on the device, including sales after the introduction of New Product B, changes in market share compared to competing products, and sentiment analysis results based on consumer reviews.

[1281] Prompt Sentence Examples

[1282] For example, consider the following prompt:

[1283] "New Product B will be introduced on June 1, 2023, at a price of 2,000 yen. Analyze review sentiment at the time of this introduction to see whether there are more positive or negative reactions."

[1284] The above is a specific embodiment for carrying out the present invention. This system enables companies to carry out detailed analysis and predictions that take into account consumer behavior and emotions, enabling the formulation of flexible marketing strategies.

[1285] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1286] Step 1:

[1287] Users collect consumer data

[1288] What it does: The user collects consumer purchasing history, customer profiles, product information, and sentiment data from the company's CRM system and marketing database.

[1289] Input: Consumer behavior and sentiment data from customer relationship management systems and marketing databases.

[1290] Output: The collected consumer behavior and sentiment data is prepared in a format for data input.

[1291] Step 2:

[1292] The terminal displays a data entry screen.

[1293] Specific behavior: The device provides a user interface and displays a form for inputting collected data, including consumer ID, purchased item, purchase date, payment amount, age, gender, region, and emotion tag (e.g., joy, anger, sadness, etc.).

[1294] Input: Consumer behavior and sentiment data collected by users.

[1295] Output: A data entry form is displayed on the terminal screen.

[1296] Step 3:

[1297] The user enters and submits data

[1298] Specific operation: The user enters the required information into the data input form on the terminal and clicks the submit button. The terminal converts the entered data into JSON format and sends it to the server.

[1299] Input: Consumer behavior and sentiment data entered into a data entry form.

[1300] Output: The data is converted to JSON format and sent to the server.

[1301] Step 4:

[1302] The server receives and analyzes the data

[1303] Specific operation: The server receives JSON data sent from the device. After receiving it, it performs data cleansing, fills in missing values, and processes outliers. It also checks whether the emotion data is normal and removes invalid data.

[1304] Input: JSON formatted consumer behavior and sentiment data sent from the device.

[1305] Output: Cleansed and accurate consumer behavior and sentiment data.

[1306] Step 5:

[1307] The server generates a virtual consumer model.

[1308] Specific operation: The server uses a machine learning algorithm to generate a virtual consumer model based on the analyzed consumer behavior data and emotion data. The generated model includes the consumer's purchase history, emotional state, interests, and future purchase predictions.

[1309] Input: Cleansed consumer behavior and sentiment data.

[1310] Output: A hypothetical consumer model.

[1311] Step 6:

[1312] The server aggregates multiple virtual consumer models.

[1313] Specific operation: The server integrates the generated multiple virtual consumer models to build a virtual market, which includes supply and demand for goods and services, competitive information, and consumer sentiment data.

[1314] Input: Multiple hypothetical consumer models.

[1315] Output: An integrated virtual marketplace.

[1316] Step 7:

[1317] The user sets up a business scenario

[1318] Specific operation: The user sets up a business scenario on the device. For example, they set up a scenario such as "Introduce new product B on June 1, 2023 at a price of 2,000 yen. Analyze review sentiment at the time of introduction." They then enter detailed scenario information and send it to the server.

[1319] Input: Business scenario (e.g. introduction of new product B, price, introduction date).

[1320] Output: Business scenario data sent to the server.

[1321] Step 8:

[1322] The server receives and analyzes the scenario.

[1323] Specific operation: The server analyzes the received scenario data and prepares for the simulation. Based on the set scenario, it prepares to predict fluctuations in the virtual market.

[1324] Input: Submitted business scenario data.

[1325] Output: Simulation ready.

[1326] Step 9:

[1327] The server runs the simulation

[1328] How it works: The server applies a scenario within the virtual market and starts a simulation. For example, it predicts sales from the introduction of new product B, changes in market share of competing products, and emotional reactions to the introduction. The emotion engine analyzes consumer reviews and feedback to determine positive or negative emotions.

[1329] Input: Simulation preparation data, review and feedback data to be analyzed for sentiment.

[1330] Output: Simulation results (sales forecast, market share change of competing products, predicted emotional response).

[1331] Step 10:

[1332] The server generates the simulation results

[1333] How it works: After the simulation is completed, the server aggregates the results and outputs them in JSON format, including sales forecasts, profit margins, market share fluctuation data, and sentiment analysis results.

[1334] Input: Various data after the simulation is run.

[1335] Output: Simulation results organized in JSON format.

[1336] Step 11:

[1337] The terminal receives and displays the simulation results.

[1338] Specific operation: The terminal receives the simulation results sent from the server, saves them, and displays them on the user interface.

[1339] Input: Simulation results sent from the server.

[1340] Output: Simulation results displayed on the terminal screen.

[1341] Step 12:

[1342] The terminal visualizes the results

[1343] Specific operation: The device visualizes the received results in graphs and tables and provides them to the user. Sentiment analysis results are displayed as the percentage of positive reactions, the percentage of negative reactions, etc. The user analyzes the results and revises their marketing strategy and business plan.

[1344] Input: Received simulation results.

[1345] Output: A visual summary in graphical and tabular form.

[1346] (Application example 2)

[1347] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1348] Modern purchasing behavior analysis does not adequately collect and analyze consumer emotional data, making it difficult to accurately grasp consumers' true needs and reactions. As a result, the accuracy of marketing strategies and product recommendations is limited, making it difficult to improve consumer satisfaction and maximize sales. There is a need to solve this problem, generate more accurate consumer models, and provide a system that can make optimal product recommendations.

[1349] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1350] In this invention, the server includes means for receiving and analyzing real consumer behavior data, means for generating a virtual consumer model based on the analyzed data, means for integrating the generated virtual consumer models to build a virtual market, means for collecting gaze and facial expression data and analyzing consumer emotions, means for recommending optimal products to users based on the analyzed emotion data, and means for outputting simulation results. This makes it possible to generate a sophisticated virtual consumer model based on consumer emotions, and to recommend optimal products and formulate marketing strategies.

[1351] "Actual consumer behavior data" refers to data such as the history, behavioral information, and payment information of consumers when they actually purchase products, and indicates the specific behavior of individual consumers.

[1352] "Means for analysis" refers to a device or program for analyzing collected data and extracting useful information.

[1353] A "virtual consumer model" is a model that is generated based on real consumer behavior data and mimics changes in consumer purchasing behavior and emotions.

[1354] The "means for integrating and constructing a virtual market" is a device or program for aggregating multiple virtual consumer models and generating a virtual market that mimics a real market environment.

[1355] A "means for simulating business scenarios" is a device or program for testing business plans such as the introduction of new products, price changes, and the implementation of information provision systems in a virtual market and predicting their effects.

[1356] "Gaze and facial expression data" refers to data that shows the eye movements and facial expressions of consumers when they view products, and is information used to analyze consumers' emotions and interests.

[1357] "Means for analyzing consumer emotions" refers to a device or program for recognizing and analyzing the emotional state of consumers based on gaze and facial expression data.

[1358] The "means for recommending optimal products to users" refers to a device or program that selects and suggests products that are likely to interest users based on analyzed emotional data and purchasing behavior data.

[1359] The "means for outputting the simulation results" is a device or program for displaying or reporting the results of the business scenario simulation.

[1360] This invention is a system that builds a virtual market by collecting and analyzing real-world consumer behavior data and consumer emotion data, and then generating and integrating a virtual consumer model based on that data. It also simulates business scenarios within that virtual market and outputs the simulation results. This system also includes functions to collect gaze and facial expression data, analyze consumer emotions, and recommend optimal products to users based on the analyzed emotion data.

[1361] 1. Data collection and transmission

[1362] The user wears the smart glasses and browses the products. The smart glasses' built-in camera and sensors collect the user's gaze and facial expression data in real time.

[1363] The collected data is converted into JSON format and sent to the server via the terminal.

[1364] 2. Data Reception and Analysis

[1365] The server receives the JSON data sent from the device and performs data cleansing on the received data, including filling in missing values ​​and processing outliers.

[1366] The server analyzes consumer emotions from gaze and facial expression data using software such as OpenCV, dlib, and the GazeTracking library.

[1367] 3. Creating a Virtual Consumer Model

[1368] The server generates a virtual consumer model based on the analyzed consumer behavior and emotion data, often using machine learning algorithms.

[1369] The generated virtual consumer model includes purchasing history, emotional state, interests, and future purchase predictions.

[1370] 4. Building a virtual market

[1371] The server integrates the generated virtual consumer models to create a virtual market, which includes supply and demand for each product or service, competitive information, and consumer sentiment data.

[1372] 5. Business scenario simulation

[1373] Users set up business scenarios such as new product introductions, price changes, and advertising campaigns on their terminals.

[1374] The server predicts and runs simulations of virtual market fluctuations based on set scenarios, resulting in forecasts of sales, profit margins, market share fluctuations, and consumer emotional responses.

[1375] 6. Output of simulation results

[1376] The server outputs the simulation results in JSON format, and the terminal receives and visualizes them.

[1377] The user reviews their marketing strategy and business plan based on the output results.

[1378] Specific examples

[1379] If a user wears smart glasses and smiles when looking at Product A, the application will analyze their facial expression and gaze and recommend "Recommended Product A." Below is a specific example of a prompt sentence to input to the generative AI model.

[1380] Example prompt sentence:

[1381] When a user wears smart glasses and smiles when looking at product A, the application should analyze their facial expressions and gaze and recommend the most suitable product in real time.

[1382] The above is a specific embodiment for carrying out the invention. This system allows for a precise understanding of consumer emotions, making it possible to recommend optimal products and formulate marketing strategies.

[1383] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1384] Step 1:

[1385] A user puts on the smart glasses and browses the products.

[1386] (input)

[1387] User gaze and facial expression data.

[1388] (process)

[1389] The smart glasses' built-in cameras and sensors collect the user's gaze and facial expression data.

[1390] (output)

[1391] Real-time gaze and facial expression data is captured.

[1392] Step 2:

[1393] The data collected by the device is converted into JSON format and sent to the server.

[1394] (input)

[1395] Gaze and facial expression data.

[1396] (process)

[1397] The terminal converts the data into JSON format and sends it to the server over the network.

[1398] (output)

[1399] Gaze and facial expression data in JSON format is sent to the server.

[1400] Step 3:

[1401] The server performs data cleansing on the data received.

[1402] (input)

[1403] Gaze and expression data in JSON format.

[1404] (process)

[1405] The server cleanses the data by imputing missing values ​​and handling outliers.

[1406] (output)

[1407] Cleansed gaze and facial expression data.

[1408] Step 4:

[1409] The server analyzes consumer emotions from gaze and facial expression data.

[1410] (input)

[1411] Cleansed gaze and facial expression data.

[1412] (process)

[1413] Emotions are analyzed using software such as OpenCV, dlib, and GazeTracking. Specifically, changes in gaze direction and facial expressions are detected to determine the emotional state.

[1414] (output)

[1415] Consumer sentiment data.

[1416] Step 5:

[1417] The server generates a virtual consumer model based on the analyzed data.

[1418] (input)

[1419] Analyzed consumer sentiment data and consumption behavior data.

[1420] (process)

[1421] Use machine learning algorithms to create models that predict the relationship between consumer purchasing behavior and emotions.

[1422] (output)

[1423] Virtual consumer model.

[1424] Step 6:

[1425] The server integrates the generated virtual consumer models and constructs a virtual market.

[1426] (input)

[1427] Multiple virtual consumer models.

[1428] (process)

[1429] The server integrates multiple virtual consumer models to create a virtual market, which includes supply and demand for each product or service, competitive information, and consumer sentiment data.

[1430] (output)

[1431] Hypothetical market data.

[1432] Step 7:

[1433] The user sets up a business scenario on the terminal.

[1434] (input)

[1435] Scenario data for new product introductions, price changes, advertising campaigns, etc.

[1436] (process)

[1437] The user uses the interface on the terminal to input detailed information about the scenario and transmits it to the server.

[1438] (output)

[1439] The scenario data is sent to the server.

[1440] Step 8:

[1441] The server analyzes the received scenario data and executes a simulation.

[1442] (input)

[1443] Scenario and hypothetical market data.

[1444] (process)

[1445] Based on a set scenario, a simulation is run to predict fluctuations in a virtual market.

[1446] (output)

[1447] Simulation result data.

[1448] Step 9:

[1449] The server outputs the simulation results in JSON format, and the terminal receives and visualizes them.

[1450] (input)

[1451] Simulation result data.

[1452] (process)

[1453] The server outputs the simulation results in JSON format, and the terminal receives them and visualizes them in graphs or tables.

[1454] (output)

[1455] Visualized simulation results.

[1456] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1457] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1458] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1459] [Fourth embodiment]

[1460] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1461] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1462] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1463] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1464] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1465] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1466] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1467] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1468] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1469] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1470] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1471] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1472] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1473] The system according to the present invention performs a series of operations: it receives and analyzes real-world consumer behavior data, generates virtual consumer models, integrates them to build a virtual market, simulates business scenarios within the built virtual market, and outputs the results. Below, we will create a program for the system and explain its processing in detail.

[1474] Data entry and submission

[1475] 1. Users collect consumer data

[1476] Users collect consumer purchasing history, customer profiles, product information, etc. from a company's customer relationship management (CRM) system or marketing database. This data collection is done by extracting data from CSV files or directly from the database.

[1477] 2. The device displays the data entry screen.

[1478] The terminal provides a user interface and displays a form for the user to review and enter the collected consumer data, including information such as consumer ID, purchased product, purchase date, payment amount, age, gender, and region.

[1479] 3. The user enters and submits consumer data

[1480] The user enters the required information into the data input form on the terminal and clicks the submit button. The terminal converts the entered data into JSON format and sends it to the server.

[1481] Generating the Consumer Model

[1482] 4. The server receives and analyzes the data

[1483] The server receives the JSON data sent from the terminal. Specifically, it is designed to receive the data as an HTTP request so that it can reach the server.

[1484] The server analyzes the received data and performs data cleansing and preprocessing, which includes imputing missing data, handling outliers, and normalizing the data.

[1485] 5. The server generates a virtual consumer model

[1486] The server generates a virtual consumer model based on the analyzed data. Here, machine learning algorithms (e.g., clustering and regression models) are used to create a model that predicts consumer purchasing behavior. The generated virtual consumer model includes the consumer's purchasing history, interests, and future purchase predictions.

[1487] Building a virtual market

[1488] 6. The server aggregates multiple virtual consumer models

[1489] The server integrates the generated virtual consumer models to create a virtual market, which mimics a real-world market environment and includes information on the supply and demand of each product or service, as well as information on competitors.

[1490] Business scenario simulation

[1491] 7. The user sets up the business scenario

[1492] The user sets up a business scenario on the device. For example, a scenario could be "Introduce new product B on June 1, 2023 at a price of 2,000 yen." The user enters detailed information about the scenario and sends it to the server.

[1493] 8. The server receives the scenario and runs the simulation.

[1494] The server receives the scenario and runs a simulation in the virtual market based on that scenario, predicting things like sales fluctuations due to new product introductions, demand changes due to price changes, and the effects of advertising campaigns.

[1495] The simulation results include sales forecasts, profit margins, and changes in market share.

[1496] Output of simulation results

[1497] 9. The server generates the simulation results

[1498] The server generates the simulation results in JSON format, which include statistics on sales, profits, and market share after the scenario is executed.

[1499] 10. The device receives and displays the results

[1500] The terminal visualizes the simulation results received from the server and provides them to the user. The results are displayed in graphs and tables, and the user can use them to review strategies and make decisions.

[1501] Specific examples

[1502] New product introduction simulation example

[1503] The user sets up a scenario in which new product B will be introduced on June 1, 2023, at a price of 2,000 yen. The device sends this scenario to the server, which then runs a simulation within the virtual market. The results of the simulation, such as sales after the introduction of new product B, changes in market share with competing products, and the effectiveness of advertising campaigns, are calculated and displayed as graphs on the device. This allows the user to accurately predict the likelihood of success in introducing the new product and plan optimal marketing measures.

[1504] The above is a specific embodiment for carrying out the present invention. This system enables companies to perform detailed analysis and prediction of consumer behavior, and can be used as a tool to quickly respond to changing market conditions.

[1505] The processing flow will be explained below.

[1506] Step 1:

[1507] The user collects consumer data, such as consumer purchase history, customer profile, and product information, from a company's customer relationship management (CRM) system or marketing database. This data includes, for example, consumer ID, purchased product, purchase date, payment amount, age, gender, and region.

[1508] Step 2:

[1509] The terminal displays a data entry screen. The form includes fields for inputting information such as consumer ID, purchased item, purchase date, payment amount, age, gender, and region. The terminal provides an interface for the user to enter the collected data.

[1510] Step 3:

[1511] The user enters and submits consumer data. The user manually enters the required information into the data input form on the terminal and clicks the submit button. The terminal converts the entered data into JSON format and sends it to the server.

[1512] Step 4:

[1513] The server receives and analyzes the data. The server receives the JSON data sent from the device. After receiving the data, the server cleanses the data, filling in missing values ​​and processing outliers. For example, it excludes data with ages of 0 or 150.

[1514] Step 5:

[1515] The server generates a virtual consumer model based on the analyzed data. Specifically, it uses a machine learning algorithm to predict consumer purchasing behavior. For example, it uses a clustering algorithm to segment consumers and extract purchasing behavior patterns.

[1516] Step 6:

[1517] The server integrates multiple virtual consumer models. The server integrates the generated multiple virtual consumer models into a virtual market. This allows the virtual market to mimic a real market environment, including supply and demand for goods and services, as well as competitive information.

[1518] Step 7:

[1519] The user sets up a business scenario. The user inputs the business scenario into the terminal. For example, the user sets up a scenario such as "Introduce new product B on June 1, 2023 at a price of 2,000 yen." The terminal then sends this scenario to the server.

[1520] Step 8:

[1521] The server receives and analyzes the scenario. The server analyzes the received scenario data and prepares for the simulation. Based on the set scenario, the server prepares to predict fluctuations in the virtual market.

[1522] Step 9:

[1523] The server runs the simulation. The server applies a scenario within the virtual market and starts the simulation. For example, it calculates results such as estimated sales from the introduction of new product B, changes in market share with competitors' products, and the effectiveness of advertising campaigns.

[1524] Step 10:

[1525] The server generates the simulation results. After the simulation is completed, the server aggregates the results and outputs them in JSON format. The results include sales forecasts, profit margins, and market share fluctuation data.

[1526] Step 11:

[1527] The terminal receives the simulation results. The terminal receives the simulation results sent from the server. The received results are stored as they are and displayed on the user interface.

[1528] Step 12:

[1529] The terminal visualizes the results. The terminal visualizes the received results in graphs and tables and provides them to the user. The user analyzes the results and reviews their marketing strategies and business plans.

[1530] The above is a concrete processing flow of the system according to the present invention, which provides businesses with an effective tool for detailed analysis and prediction of consumer behavior and for quickly responding to changing market conditions.

[1531] Example 1

[1532] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1533] In a real market environment, it is difficult to effectively utilize consumer behavior data, quickly generate a virtual consumer model, and then simulate business scenarios based on that model. With conventional systems, data preparation, model generation, scenario setting, and simulation are all performed separately, which is extremely time-consuming and labor-intensive. Furthermore, when companies introduce new products, change prices, or plan advertising campaigns, there is no effective system that can comprehensively simulate these elements and quickly visualize and provide the results.

[1534] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1535] In this invention, the server includes means for receiving JSON data sent from the terminal, performing data cleansing and preprocessing, and generating a virtual consumer model; means for the server to integrate the generated virtual consumer model and build a virtual market; and means for the server to run a simulation within the virtual market based on a set scenario. This enables companies to quickly and effectively generate a virtual consumer model using consumer behavior data and simulate business scenarios in the virtual market. Furthermore, the simulation results can be quickly visualized to support strategy revisions and policy decisions.

[1536] "User" refers to an individual or corporation that uses the system to collect, input, and manage consumer behavior data and set scenarios.

[1537] "Device" refers to a computer device used by a user to input consumption behavior data and set up a scenario. Examples include PCs, tablets, and smartphones.

[1538] "Server" refers to a computer system that receives, analyzes, and processes data sent from terminals, generates virtual consumer models and virtual markets, and executes simulations.

[1539] "Consumer behavior data" refers to data that includes information about consumer behavior, such as consumer purchasing history, customer profiles, and product information.

[1540] "JSON" is an abbreviation for JavaScript Object Notation, and refers to a lightweight data format for data exchange. It represents structured data in a way that is easy for humans to read and machines to parse.

[1541] "Data cleansing" refers to the process of improving data quality by completing missing values, processing outliers, normalizing data, etc. as a preprocessing step for data analysis.

[1542] A "virtual consumer model" refers to a model that includes a consumer's purchasing history, interests, and future purchasing predictions, and is generated using a machine learning algorithm based on consumer behavior data.

[1543] A "virtual market" refers to a simulation environment that integrates multiple virtual consumer models and mimics a real market environment, including product supply and demand, pricing strategies, and competitor information.

[1544] A "business scenario" is a plan or strategy that a company sets up to simulate in a virtual market, including new product introductions, price changes, advertising campaigns, etc.

[1545] "Simulation" refers to the process of forecasting sales, analyzing the impact of price changes, measuring the effectiveness of advertising campaigns, etc., based on business scenarios set up within a virtual marketplace.

[1546] "Simulation results" refers to statistical data such as sales, profits, and market share obtained after executing a business scenario.

[1547] The system according to the present invention performs a series of operations: collects and analyzes real-world consumer behavior data to generate a virtual consumer model, integrates the data to build a virtual market, simulates business scenarios within the virtual market, and outputs the results. A specific embodiment of this system is described below.

[1548] Data entry and submission

[1549] Users collect consumer data

[1550] Users collect consumer behavior data from corporate customer relationship management (CRM) systems and marketing databases. Specifically, they use SQL queries to extract consumer purchasing histories and customer profiles from the databases and save them as CSV files.

[1551] The terminal displays a data entry screen.

[1552] The terminal displays a user interface for entering consumer data. The terminal can be a computer device such as a PC, tablet, or smartphone. The data entry form includes fields such as consumer ID, product purchased, purchase date, payment amount, age, gender, and region.

[1553] User enters and submits consumer data

[1554] The user manually enters data into the input form on the device and presses the submit button. The device converts the entered data into JSON format, generates an HTTP request, and sends it to the server. This is often done using a programming language such as JavaScript.

[1555] Generating the Consumer Model

[1556] The server receives and analyzes the data

[1557] The server receives the JSON data sent from the terminal and performs data cleansing using Python's Pandas library, among other tools. During this process, the server performs tasks such as filling in missing data, handling outliers, and normalizing the data.

[1558] The server generates a virtual consumer model.

[1559] The server generates a virtual consumer model based on the cleansed data. Specifically, it uses the machine learning library Scikit-learn to perform clustering and create a model that predicts consumer purchasing behavior.

[1560] Building a virtual market

[1561] The server aggregates multiple virtual consumer models.

[1562] The server integrates the generated virtual consumer models to create a virtual market, which includes information on product supply and demand, pricing, and competitors.

[1563] Business scenario simulation

[1564] The user sets up a business scenario

[1565] The user sets up a business scenario using the terminal interface. For example, the scenario "New product A will be introduced on January 5, 2024 at a price of 1,500 yen" is input and sent to the server.

[1566] The server receives the scenario and runs the simulation.

[1567] The server receives business scenarios and runs simulations within the virtual marketplace, which can include sales forecasting, analyzing the impact of price changes, and measuring the effectiveness of advertising campaigns, often using deep learning libraries such as TensorFlow and PyTorch.

[1568] Output of simulation results

[1569] The server generates the simulation results

[1570] The server generates simulation results in JSON format, including statistical data such as sales, profits, and market share.

[1571] The terminal receives and displays the results

[1572] The terminal visualizes the simulation results received from the server. For example, it uses the JavaScript library D3.js to draw a sales forecast graph and display it to the user, helping them to review strategies and decide on measures.

[1573] Specific examples

[1574] New product introduction simulation example

[1575] The user sets a scenario in which "New Product B will be introduced on June 1, 2023 at a price of 2,000 yen," and the device sends this scenario to the server. The server then runs a simulation within the virtual market and generates the results. Specific simulation results include sales after the introduction of New Product B, changes in market share with competing products, and the effectiveness of advertising campaigns. These results are displayed on the device as graphs and tables, allowing the user to plan marketing measures with high precision based on these results.

[1576] Examples of prompt statements

[1577] Below are some examples of prompt sentences to input into the generative AI model.

[1578] "You want to introduce new product B on June 1, 2023, at a price of 2,000 yen. Please conduct a market simulation and show the predicted changes in sales, profits, and market share after the introduction of new product B."

[1579] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1580] Step 1:

[1581] Users collect consumer data

[1582] Users collect consumer behavior data from a company's customer relationship management (CRM) system or marketing database. Specifically, they use SQL queries to extract consumer purchasing history and customer profiles from the database and save them as a CSV file. Input data includes consumer ID, purchased product, purchase date, payment amount, age, gender, region, etc. The output provides organized consumer behavior data.

[1583] Step 2:

[1584] The terminal displays a data entry screen.

[1585] The terminal displays a user interface for inputting consumer data. It allows the user to manually enter fields such as consumer ID, purchased product, purchase date, payment amount, age, gender, and region through a data input form. The input includes consumer behavior data collected by the user. The output includes the data entered by the user on the input screen.

[1586] Step 3:

[1587] User enters and submits consumer data

[1588] The user manually enters data into the input form on the device and presses the send button. The device converts the entered data into JSON format, generates an HTTP request, and sends it to the server. The input is consumption behavior data manually entered by the user. The output is the data converted into JSON format and sent to the server as an HTTP request.

[1589] Step 4:

[1590] The server receives and analyzes the data

[1591] The server receives JSON data sent from the device. It then performs data cleansing using the Python Pandas library. During this process, it performs tasks such as filling in missing data, handling outliers, and normalizing the data. The input is consumer behavior data in JSON format. The output is the cleansed data.

[1592] Step 5:

[1593] The server generates a virtual consumer model.

[1594] The server generates a virtual consumer model based on the preprocessed data. Specifically, it performs clustering using the Scikit-learn library to create consumer segments. The input is the cleansed consumer behavior data. The output is the generated virtual consumer model.

[1595] Step 6:

[1596] The server aggregates multiple virtual consumer models.

[1597] The server integrates the generated virtual consumer models to construct a virtual market. This virtual market includes information on product supply and demand, pricing, and competitors. The input is the multiple virtual consumer models. The output is the constructed virtual market.

[1598] Step 7:

[1599] The user sets up a business scenario

[1600] The user sets up a business scenario using the terminal interface. For example, they input a scenario such as "Introduce new product A on January 5, 2024 at a price of 1,500 yen" and send it to the server. The input includes detailed information about the business scenario. The output is the scenario sent to the server.

[1601] Step 8:

[1602] The server receives the scenario and runs the simulation.

[1603] The server receives business scenarios and runs simulations within a virtual market. It uses deep learning libraries such as TensorFlow and PyTorch to forecast sales, analyze the impact of price changes, and measure the effectiveness of advertising campaigns. The inputs are the business scenarios and virtual market data. The output is the simulation results.

[1604] Step 9:

[1605] The server generates the simulation results

[1606] The server generates the simulation results in JSON format. The results include statistical data such as sales, profit, market share, etc. As input, we have the detailed results of the simulation. As output, we get the simulation results in JSON format.

[1607] Step 10:

[1608] The terminal receives and displays the results

[1609] The terminal visualizes the simulation results received from the server. Using the JavaScript D3.js library, it draws a sales forecast graph and displays it to the user. The input is the simulation results in JSON format. The output is visualized data in graph and table format.

[1610] (Application example 1)

[1611] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1612] Conventional consumer behavior data analysis systems are limited to virtual market simulations, and have the problem of not being able to adequately predict and optimize for a variety of applications. Furthermore, for certain services, such as self-driving vehicles, there is a lack of systems that can analyze actual usage data, forecast demand, and provide optimal service routes. This has resulted in insufficient improvements in service efficiency and user satisfaction.

[1613] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1614] In this invention, the server includes means for receiving and analyzing real-world consumer behavior data, means for generating a virtual consumer model based on the analyzed data, means for integrating the generated virtual consumer model to build a virtual market, means for simulating business scenarios within the built virtual market, means for outputting simulation results, means for collecting and analyzing vehicle usage data, means for forecasting demand for on-demand services based on the usage data, and means for providing optimal service routes based on the demand forecast. This enables companies to use the results of simulations within the virtual market to forecast demand for various services and routes and to develop optimal strategies based on the forecasts.

[1615] "Actual consumer behavior data" refers to data on consumers' actual purchasing behavior and service usage. Specifically, it includes purchase history, usage history, customer profile, etc.

[1616] "Means of analysis" refers to the process of analyzing the received data and performing preprocessing such as information cleansing, handling outliers, and normalizing the data.

[1617] A "virtual consumer model" is a predictive model of consumer purchasing behavior and service usage behavior that is generated using machine learning algorithms based on real-world consumer behavior data.

[1618] A "virtual market" is a simulation environment that integrates multiple virtual consumer models and mimics a real market environment, including information on the supply and demand of each product or service, as well as competitor information.

[1619] A "business scenario" is a specific strategy or measure that a company is considering implementing. Examples include the introduction of a new product, a price change, or an advertising campaign.

[1620] "Simulation results" are statistical data such as sales forecasts, profit margins, and fluctuations in market share calculated as a result of executing a business scenario within a virtual market.

[1621] "Vehicle usage data" refers to data related to the usage history of autonomous vehicles and on-demand services, including ride history, fares, usage time, and user profiles.

[1622] "Demand forecasting for on-demand services" is the process of analyzing usage data to predict how much demand there will be for a particular service.

[1623] An "optimal service route" is a route or service plan that is optimal for a user and is provided based on a demand forecast.

[1624] This invention is a system for providing on-demand services provided by autonomous vehicles, which analyzes consumer behavior data of users and builds a virtual market to forecast demand for services and provide optimal routes. Below, we will explain in detail the processing of the program for realizing this system.

[1625] Data collection and analysis

[1626] First, the user collects information about their use of the autonomous vehicle service. This is done by extracting it from a CSV file or directly from a database. The terminal then displays a form for confirming and entering consumer data, and the user enters information such as user ID, route, usage time, payment amount, age, gender, and region. The entered data is converted to JSON format and sent to the server.

[1627] Creating a Virtual Consumer Model

[1628] The server receives the JSON data sent from the device and performs data cleansing and preprocessing. This includes filling in missing data, handling outliers, and normalizing the data. It then uses machine learning algorithms (such as clustering and regression models) to generate a virtual consumer model that predicts user behavior. The generated virtual consumer model includes the user's usage history, interests, and future usage predictions.

[1629] Building a virtual market

[1630] Next, the server integrates the generated virtual consumer models to create a virtual market. This virtual market mimics a real-world market environment, including the supply and demand of each product or service, as well as information on competitors. This makes it possible to carry out simulations within the virtual market.

[1631] Business scenario simulation

[1632] The user sets up a business scenario on the device, such as adding a new route or offering a discount service during a specific time period, and this scenario is input to the generative AI model as a prompt.

[1633] Prompt Sentence Examples

[1634] "New Route A will be introduced on October 1, 2023, with a discount campaign during the first month. We will provide demand and revenue forecasts after the introduction."

[1635] The server runs simulations within the virtual market to forecast demand, and the simulation results include data such as sales forecasts, personnel allocation, and optimal service routes.

[1636] Output of simulation results

[1637] Finally, the server generates the simulation results in JSON format and sends them to the terminal. The terminal visualizes the results received from the server and provides them to the user. The results are displayed in graphs and tables, allowing the user to review their strategy and make decisions based on them.

[1638] Hardware and software used

[1639] This system analyzes data using cloud servers (such as AWS or Google Cloud), primarily using MySQL or PostgreSQL as databases, and Python and libraries such as TensorFlow and Scikit-Learn for machine learning. The front end can be developed using frameworks such as React and Angular.

[1640] This system enables companies to use simulation results within a virtual market to forecast demand for various services and routes, and then develop optimal strategies based on that forecast.

[1641] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1642] Step 1:

[1643] Users collect usage data for the autonomous vehicle service. Specifically, information such as user ID, route, usage time, payment amount, age, gender, and region is collected using CSV files or database extraction. The collected data is then entered into a terminal.

[1644] Step 2:

[1645] The terminal displays a form for verifying and entering consumer data. The user enters the collected data into this form, confirms the input, and clicks the submit button. The input at this point is detailed information related to the user.

[1646] Step 3:

[1647] The terminal converts the input data into JSON format and sends it to the server. This converts the information entered by the user into a data structure and sends it via an HTTP request. The output is structured user data.

[1648] Step 4:

[1649] The server receives the JSON data sent from the terminal. After receiving it, the server performs data cleansing and preprocessing. Specifically, it completes missing data, processes outliers, and normalizes the data. The input is JSON data, and the output is the preprocessed, clean data.

[1650] Step 5:

[1651] The server generates a virtual consumer model based on the preprocessed data. Machine learning algorithms (e.g., clustering and regression models) are used to create a model that predicts consumption behavior, interests, and future behavior. The input is the preprocessed data, and the output is the virtual consumer model.

[1652] Step 6:

[1653] The server integrates multiple generated virtual consumer models to build a virtual market. This virtual market is a simulation environment that includes information on the supply and demand of each product or service, as well as information on competitors. The input is the virtual consumer model, and the output is the virtual market.

[1654] Step 7:

[1655] The user sets a business scenario on the terminal. For example, they set a specific prompt such as "New route A will be introduced on October 1, 2023, and a discount campaign will be held in the first month." The input is the business scenario, and the output is the prompt.

[1656] Step 8:

[1657] The server executes a simulation in the virtual market based on the received prompt. Specific simulation operations include demand forecasting, sales forecasting, personnel allocation, and calculation of optimal service routes after the introduction of new routes. The input is the prompt, and the output is the simulation results.

[1658] Step 9:

[1659] The server generates simulation results in JSON format. The simulation results include data such as demand forecasts, sales, and personnel allocation. These are sent to the terminal. The input is the simulation results, and the output is the result data in JSON format.

[1660] Step 10:

[1661] The terminal visualizes the simulation results received from the server and presents them to the user. The results are displayed in graphs and tables, and the user can use them to review strategies and make decisions. The input is the simulation results in JSON format, and the output is visualized data.

[1662] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1663] The system according to the present invention is characterized by the fact that it not only receives and analyzes real consumer behavior data, but also combines it with an emotion engine that recognizes user emotions to generate a more sophisticated virtual consumer model. Below, we will create a program for the system and explain its processing in detail.

[1664] Data entry and submission

[1665] 1. Users collect consumer data

[1666] Users collect consumer purchasing history, customer profiles, product information, etc. from companies' customer relationship management (CRM) systems and marketing databases, including consumer sentiment data, such as consumer sentiment and feedback comments when purchasing a product.

[1667] 2. The device displays the data entry screen.

[1668] The terminal provides a user interface and displays a form for the user to input the collected consumer data and emotion data, including the consumer ID, purchased item, purchase date, payment amount, age, gender, region, and emotion tag (e.g., joy, anger, sadness, etc.).

[1669] 3. The user enters and submits data

[1670] The user enters the required information into the data input form on the terminal and clicks the submit button. The terminal converts the entered data into JSON format and sends it to the server.

[1671] Generating the Consumer Model

[1672] 4. The server receives and analyzes the data

[1673] The server receives the JSON data sent from the device. After receiving the data, the server performs data cleansing, fills in missing values, and processes outliers. This also includes checking whether the emotion data is normal.

[1674] 5. The server generates a virtual consumer model

[1675] The server generates a virtual consumer model based on the analyzed consumer behavior data and emotion data. Specifically, it uses a machine learning algorithm to create a model that predicts the relationship between consumer purchasing behavior and emotions. The generated virtual consumer model includes the consumer's purchasing history, emotional state, interests, and future purchase predictions.

[1676] Building a virtual market

[1677] 6. The server aggregates multiple virtual consumer models

[1678] The server integrates multiple virtual consumer models to create a virtual market that mimics a real-world market environment, including supply and demand for each product or service, competitive information, and consumer sentiment data.

[1679] Business scenario simulation

[1680] 7. The user sets up the business scenario

[1681] The user sets up a business scenario on the device. For example, the scenario is set up as follows: "New product B will be introduced on June 1, 2023, at a price of 2,000 yen. Review sentiment at the time of introduction will be analyzed." The user enters detailed information about the scenario and sends it to the server.

[1682] 8. The server receives and analyzes the scenario.

[1683] The server analyzes the received scenario data and prepares for the simulation. Based on the set scenario, it prepares to predict fluctuations in the virtual market.

[1684] 9. The server runs the simulation

[1685] The server applies a scenario within the virtual market and starts a simulation. For example, it predicts sales from the introduction of new product B, market share fluctuations compared to competitors' products, and emotional reactions at the time of introduction. The emotion engine analyzes consumer reviews and feedback to determine positive or negative emotions.

[1686] Output of simulation results

[1687] 10. The server generates the simulation results

[1688] After the simulation is completed, the server aggregates the results and outputs them in JSON format, including sales forecasts, profit margins, market share fluctuation data, and sentiment analysis results.

[1689] 11. The terminal receives and displays the simulation results.

[1690] The terminal receives the simulation results sent from the server, and the received results are stored and displayed on the user interface.

[1691] 12. The terminal visualizes the results

[1692] The device visualizes the received results in graphs and tables and provides them to the user. Sentiment analysis results are displayed as the percentage of positive reactions, the percentage of negative reactions, etc. The user can analyze the results and revise their marketing strategies and business plans.

[1693] Specific examples

[1694] New product introduction scenario

[1695] The user sets up a scenario in which new product B will be introduced on June 1, 2023, at a price of 2,000 yen, and consumer review sentiment will be analyzed. The device sends this scenario to the server, which then runs a simulation within the virtual market. The simulation results, calculated as graphs on the device, include sales after the introduction of new product B, changes in market share compared to competing products, and sentiment analysis results based on consumer reviews. This allows the user to accurately predict new product introductions and changes in consumer sentiment, and plan optimal marketing measures.

[1696] The above is a specific embodiment for carrying out the present invention. This system enables companies to analyze and predict consumer behavior in detail, and to develop flexible marketing strategies that take consumer sentiment into account.

[1697] The processing flow will be explained below.

[1698] Step 1:

[1699] The user collects consumer data, such as consumer purchasing history, customer profile, product information, and emotional data, from a company's customer relationship management (CRM) system or marketing database. For example, data such as consumer ID, purchased product, purchase date, payment amount, age, gender, region, and emotional tag (happiness, anger, sadness, etc.) is collected.

[1700] Step 2:

[1701] The terminal displays a data entry screen. The form includes the consumer ID, purchased item, purchase date, payment amount, age, gender, region, and emotion tag (e.g., joy, anger, sadness, etc.). The terminal provides an interface for the user to enter the collected data.

[1702] Step 3:

[1703] The user enters data and submits it. The user manually enters the required information into the data input form on the terminal and clicks the submit button. The terminal converts the entered data into JSON format and sends it to the server.

[1704] Step 4:

[1705] The server receives and analyzes the data. The server receives the JSON data sent from the device. After receiving the data, the server cleanses the data, filling in missing values ​​and processing outliers. For example, it removes unnatural age data (such as 0 or 150 years old) and obviously incorrect emotion tags.

[1706] Step 5:

[1707] The server uses an emotion engine to analyze user emotions based on text, audio, or video input. For example, it analyzes consumer reviews and feedback comments using natural language processing (NLP) techniques and assigns positive, negative, or neutral emotion tags.

[1708] Step 6:

[1709] The server generates a virtual consumer model based on the analyzed consumer behavior data and emotion data. Specifically, a machine learning algorithm is used to create a model that predicts the relationship between consumer purchasing behavior and emotions. The generated virtual consumer model includes the consumer's purchasing history, emotional state, interests, and future purchase predictions.

[1710] Step 7:

[1711] The server integrates multiple virtual consumer models. The server then integrates the multiple virtual consumer models that have been generated to create a virtual market. This virtual market mimics a real market environment and includes information such as supply and demand for each product or service, competitive information, and consumer sentiment data.

[1712] Step 8:

[1713] The user sets up a business scenario. The user inputs the business scenario into the device. For example, the scenario is set as "New product B will be introduced on June 1, 2023, at a price of 2,000 yen. Review sentiment at this time will be analyzed." The device then sends this scenario to the server.

[1714] Step 9:

[1715] The server receives and analyzes the scenario. The server analyzes the received scenario data and prepares for the simulation. Based on the set scenario, the server prepares to predict fluctuations in the virtual market.

[1716] Step 10:

[1717] The server runs the simulation. The server applies a scenario within the virtual market and starts the simulation. For example, it predicts sales estimates for the introduction of new product B, changes in market share with competitors' products, and emotional reactions based on consumer reviews at the time of introduction. The emotion engine analyzes consumer reviews and feedback to determine positive or negative emotions.

[1718] Step 11:

[1719] The server generates the simulation results. After the simulation is completed, the server aggregates the results and outputs them in JSON format. The results include sales forecasts, profit margins, market share fluctuation data, as well as sentiment analysis results.

[1720] Step 12:

[1721] The terminal receives and displays the simulation results. The terminal receives the simulation results sent from the server. The received results are saved as they are and displayed on the user interface.

[1722] Step 13:

[1723] The device visualizes the results. The device visualizes the received results in graphs and tables and provides them to the user. Sentiment analysis results are displayed as the percentage of positive responses, the percentage of negative responses, etc. The user analyzes the results and reviews their marketing strategies and business plans.

[1724] Example 2

[1725] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1726] Simply analyzing real-world consumer behavior data is not enough to fully understand consumer emotions and the motivations behind their purchasing behavior. For this reason, there is a need for methods to generate more sophisticated virtual consumer models and predict consumer behavior with high accuracy. In particular, the insufficient analysis of emotional data is a challenge.

[1727] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1728] In this invention, the server includes means for receiving and analyzing real consumer behavior data, means for generating a virtual consumer model based on the analyzed data, and means for integrating the generated virtual consumer model to build a virtual market. This enables comprehensive analysis of consumer behavior and emotions by integrating means for recognizing and analyzing consumer emotions with means for refining the virtual consumer model based on emotion data, making it possible to make more accurate predictions and formulate marketing strategies.

[1729] "Consumer behavior data" refers to information about what products and services consumers have purchased, such as their purchasing history, customer profiles, and product information.

[1730] An "analytical means" is a method or device for extracting, processing, and analyzing data to understand its patterns and characteristics.

[1731] A "virtual consumer model" is a consumer behavior prediction model that is virtually created based on consumer behavior data, and includes the consumer's purchasing history, emotional state, interests, and future purchasing predictions.

[1732] A "virtual market" is a virtual environment that mimics a real market environment and is constructed by integrating virtual consumer models.

[1733] A "business scenario" is a scenario used to set assumptions for a company's marketing strategy or business plan, such as the introduction of a new product, price changes, or advertising campaigns.

[1734] A "simulating means" is a method or device for predicting fluctuations and results in a virtual market based on a set business scenario.

[1735] "Simulation results" are forecast results obtained by running a simulation based on a scenario, and include sales forecasts, profit margins, market share fluctuation data, and sentiment analysis results.

[1736] A "means for recognizing and analyzing sentiment" is a method or device for identifying positive or negative sentiment from consumer reviews and feedback comments and analyzing the data.

[1737] "Emotional data" is data that represents the emotions and feedback that consumers have about products and services.

[1738] A "refinement means" is a method or device for processing and improving existing data or models in more detail and accuracy.

[1739] The system according to the present invention is characterized by the fact that it not only receives and analyzes real consumer behavior data, but also combines it with an emotion engine that recognizes user emotions to generate a more sophisticated virtual consumer model. The processing of the system program is explained in detail below.

[1740] Hardware and software used

[1741] In this system, three entities, the server, the terminal, and the user, work together. The specific hardware and software used include the following:

[1742] Server: A high-performance data analysis server (e.g., AWS EC2, Google Cloud Platform, Microsoft Azure).

[1743] Device: A device such as a computer or smartphone used by a user.

[1744] Software: Machine learning algorithms (e.g., TensorFlow, PyTorch), database management systems (e.g., MySQL, PostgreSQL), sentiment analysis engines (e.g., IBM Watson, Microsoft Text Analytics).

[1745] Specific examples of system processing

[1746] 1. Users collect consumer data

[1747] Users collect consumer purchasing history, customer profiles, and product information from companies' customer relationship management (CRM) systems and marketing databases, including consumer sentiment and feedback comments when they purchase a product.

[1748] 2. The device displays the data entry screen.

[1749] The terminal displays a form for the user to enter the collected consumer and emotional data, including consumer ID, purchased item, purchase date, payment amount, age, gender, region, and emotional tag (e.g., joy, anger, sadness, etc.).

[1750] 3. The user enters and submits data

[1751] The user enters the required information into the data input form on the terminal and clicks the submit button. The terminal converts the entered data into JSON format and sends it to the server.

[1752] 4. The server receives and analyzes the data

[1753] The server receives the JSON data sent from the device. After receiving the data, the server cleanses the data, filling in missing values ​​and processing outliers. It verifies that the emotion data is normal and removes any invalid data.

[1754] 5. The server generates a virtual consumer model

[1755] The server uses a machine learning algorithm to generate a virtual consumer model based on the analyzed consumer behavior data and emotional data, which includes the consumer's purchasing history, emotional state, interests, and future purchasing predictions.

[1756] 6. The server aggregates multiple virtual consumer models

[1757] The server integrates the generated virtual consumer models to create a virtual market, which includes supply and demand for goods and services, competitive information, and consumer sentiment data.

[1758] Examples and prompts

[1759] For example, a user can set up a business scenario in which "New Product B will be introduced on June 1, 2023, at a price of 2,000 yen, and consumer review sentiment will be analyzed." The device sends this scenario to the server, which then runs a simulation within the virtual market. The results of the simulation are calculated and displayed as graphs on the device, including sales after the introduction of New Product B, changes in market share compared to competing products, and sentiment analysis results based on consumer reviews.

[1760] Prompt Sentence Examples

[1761] For example, consider the following prompt:

[1762] "New Product B will be introduced on June 1, 2023, at a price of 2,000 yen. Analyze review sentiment at the time of this introduction to see whether there are more positive or negative reactions."

[1763] The above is a specific embodiment for carrying out the present invention. This system enables companies to carry out detailed analysis and predictions that take into account consumer behavior and emotions, enabling the formulation of flexible marketing strategies.

[1764] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1765] Step 1:

[1766] Users collect consumer data

[1767] What it does: The user collects consumer purchasing history, customer profiles, product information, and sentiment data from the company's CRM system and marketing database.

[1768] Input: Consumer behavior and sentiment data from customer relationship management systems and marketing databases.

[1769] Output: The collected consumer behavior and sentiment data is prepared in a format for data input.

[1770] Step 2:

[1771] The terminal displays a data entry screen.

[1772] Specific behavior: The device provides a user interface and displays a form for inputting collected data, including consumer ID, purchased item, purchase date, payment amount, age, gender, region, and emotion tag (e.g., joy, anger, sadness, etc.).

[1773] Input: Consumer behavior and sentiment data collected by users.

[1774] Output: A data entry form is displayed on the terminal screen.

[1775] Step 3:

[1776] The user enters and submits data

[1777] Specific operation: The user enters the required information into the data input form on the terminal and clicks the submit button. The terminal converts the entered data into JSON format and sends it to the server.

[1778] Input: Consumer behavior and sentiment data entered into a data entry form.

[1779] Output: The data is converted to JSON format and sent to the server.

[1780] Step 4:

[1781] The server receives and analyzes the data

[1782] Specific operation: The server receives JSON data sent from the device. After receiving it, it performs data cleansing, fills in missing values, and processes outliers. It also checks whether the emotion data is normal and removes invalid data.

[1783] Input: JSON formatted consumer behavior and sentiment data sent from the device.

[1784] Output: Cleansed and accurate consumer behavior and sentiment data.

[1785] Step 5:

[1786] The server generates a virtual consumer model.

[1787] Specific operation: The server uses a machine learning algorithm to generate a virtual consumer model based on the analyzed consumer behavior data and emotion data. The generated model includes the consumer's purchase history, emotional state, interests, and future purchase predictions.

[1788] Input: Cleansed consumer behavior and sentiment data.

[1789] Output: A hypothetical consumer model.

[1790] Step 6:

[1791] The server aggregates multiple virtual consumer models.

[1792] Specific operation: The server integrates the generated multiple virtual consumer models to build a virtual market, which includes supply and demand for goods and services, competitive information, and consumer sentiment data.

[1793] Input: Multiple hypothetical consumer models.

[1794] Output: An integrated virtual marketplace.

[1795] Step 7:

[1796] The user sets up a business scenario

[1797] Specific operation: The user sets up a business scenario on the device. For example, they set up a scenario such as "Introduce new product B on June 1, 2023 at a price of 2,000 yen. Analyze review sentiment at the time of introduction." They then enter detailed scenario information and send it to the server.

[1798] Input: Business scenario (e.g. introduction of new product B, price, introduction date).

[1799] Output: Business scenario data sent to the server.

[1800] Step 8:

[1801] The server receives and analyzes the scenario.

[1802] Specific operation: The server analyzes the received scenario data and prepares for the simulation. Based on the set scenario, it prepares to predict fluctuations in the virtual market.

[1803] Input: Submitted business scenario data.

[1804] Output: Simulation ready.

[1805] Step 9:

[1806] The server runs the simulation

[1807] How it works: The server applies a scenario within the virtual market and starts a simulation. For example, it predicts sales from the introduction of new product B, changes in market share of competing products, and emotional reactions to the introduction. The emotion engine analyzes consumer reviews and feedback to determine positive or negative emotions.

[1808] Input: Simulation preparation data, review and feedback data to be analyzed for sentiment.

[1809] Output: Simulation results (sales forecast, market share change of competing products, predicted emotional response).

[1810] Step 10:

[1811] The server generates the simulation results

[1812] How it works: After the simulation is completed, the server aggregates the results and outputs them in JSON format, including sales forecasts, profit margins, market share fluctuation data, and sentiment analysis results.

[1813] Input: Various data after the simulation is run.

[1814] Output: Simulation results organized in JSON format.

[1815] Step 11:

[1816] The terminal receives and displays the simulation results.

[1817] Specific operation: The terminal receives the simulation results sent from the server, saves them, and displays them on the user interface.

[1818] Input: Simulation results sent from the server.

[1819] Output: Simulation results displayed on the terminal screen.

[1820] Step 12:

[1821] The terminal visualizes the results

[1822] Specific operation: The device visualizes the received results in graphs and tables and provides them to the user. Sentiment analysis results are displayed as the percentage of positive reactions, the percentage of negative reactions, etc. The user analyzes the results and revises their marketing strategy and business plan.

[1823] Input: Received simulation results.

[1824] Output: A visual summary in graphical and tabular form.

[1825] (Application example 2)

[1826] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1827] Modern purchasing behavior analysis does not adequately collect and analyze consumer emotional data, making it difficult to accurately grasp consumers' true needs and reactions. As a result, the accuracy of marketing strategies and product recommendations is limited, making it difficult to improve consumer satisfaction and maximize sales. There is a need to solve this problem, generate more accurate consumer models, and provide a system that can make optimal product recommendations.

[1828] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1829] In this invention, the server includes means for receiving and analyzing real consumer behavior data, means for generating a virtual consumer model based on the analyzed data, means for integrating the generated virtual consumer models to build a virtual market, means for collecting gaze and facial expression data and analyzing consumer emotions, means for recommending optimal products to users based on the analyzed emotion data, and means for outputting simulation results. This makes it possible to generate a sophisticated virtual consumer model based on consumer emotions, and to recommend optimal products and formulate marketing strategies.

[1830] "Actual consumer behavior data" refers to data such as the history, behavioral information, and payment information of consumers when they actually purchase products, and indicates the specific behavior of individual consumers.

[1831] "Means for analysis" refers to a device or program for analyzing collected data and extracting useful information.

[1832] A "virtual consumer model" is a model that is generated based on real consumer behavior data and mimics changes in consumer purchasing behavior and emotions.

[1833] The "means for integrating and constructing a virtual market" is a device or program for aggregating multiple virtual consumer models and generating a virtual market that mimics a real market environment.

[1834] A "means for simulating business scenarios" is a device or program for testing business plans such as the introduction of new products, price changes, and the implementation of information provision systems in a virtual market and predicting their effects.

[1835] "Gaze and facial expression data" refers to data that shows the eye movements and facial expressions of consumers when they view products, and is information used to analyze consumers' emotions and interests.

[1836] "Means for analyzing consumer emotions" refers to a device or program for recognizing and analyzing the emotional state of consumers based on gaze and facial expression data.

[1837] The "means for recommending optimal products to users" refers to a device or program that selects and suggests products that are likely to interest users based on analyzed emotional data and purchasing behavior data.

[1838] The "means for outputting the simulation results" is a device or program for displaying or reporting the results of the business scenario simulation.

[1839] This invention is a system that builds a virtual market by collecting and analyzing real-world consumer behavior data and consumer emotion data, and then generating and integrating a virtual consumer model based on that data. It also simulates business scenarios within that virtual market and outputs the simulation results. This system also includes functions to collect gaze and facial expression data, analyze consumer emotions, and recommend optimal products to users based on the analyzed emotion data.

[1840] 1. Data collection and transmission

[1841] The user wears the smart glasses and browses the products. The smart glasses' built-in camera and sensors collect the user's gaze and facial expression data in real time.

[1842] The collected data is converted into JSON format and sent to the server via the terminal.

[1843] 2. Data Reception and Analysis

[1844] The server receives the JSON data sent from the device and performs data cleansing on the received data, including filling in missing values ​​and processing outliers.

[1845] The server analyzes consumer emotions from gaze and facial expression data using software such as OpenCV, dlib, and the GazeTracking library.

[1846] 3. Creating a Virtual Consumer Model

[1847] The server generates a virtual consumer model based on the analyzed consumer behavior and emotion data, often using machine learning algorithms.

[1848] The generated virtual consumer model includes purchasing history, emotional state, interests, and future purchase predictions.

[1849] 4. Building a virtual market

[1850] The server integrates the generated virtual consumer models to create a virtual market, which includes supply and demand for each product or service, competitive information, and consumer sentiment data.

[1851] 5. Business scenario simulation

[1852] Users set up business scenarios such as new product introductions, price changes, and advertising campaigns on their terminals.

[1853] The server predicts and runs simulations of virtual market fluctuations based on set scenarios, resulting in forecasts of sales, profit margins, market share fluctuations, and consumer emotional responses.

[1854] 6. Output of simulation results

[1855] The server outputs the simulation results in JSON format, and the terminal receives and visualizes them.

[1856] The user reviews their marketing strategy and business plan based on the output results.

[1857] Specific examples

[1858] If a user wears smart glasses and smiles when looking at Product A, the application will analyze their facial expression and gaze and recommend "Recommended Product A." Below is a specific example of a prompt sentence to input to the generative AI model.

[1859] Example prompt sentence:

[1860] When a user wears smart glasses and smiles when looking at product A, the application should analyze their facial expressions and gaze and recommend the most suitable product in real time.

[1861] The above is a specific embodiment for carrying out the invention. This system allows for a precise understanding of consumer emotions, making it possible to recommend optimal products and formulate marketing strategies.

[1862] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1863] Step 1:

[1864] A user puts on the smart glasses and browses the products.

[1865] (input)

[1866] User gaze and facial expression data.

[1867] (process)

[1868] The smart glasses' built-in cameras and sensors collect the user's gaze and facial expression data.

[1869] (output)

[1870] Real-time gaze and facial expression data is captured.

[1871] Step 2:

[1872] The data collected by the device is converted into JSON format and sent to the server.

[1873] (input)

[1874] Gaze and facial expression data.

[1875] (process)

[1876] The terminal converts the data into JSON format and sends it to the server over the network.

[1877] (output)

[1878] Gaze and facial expression data in JSON format is sent to the server.

[1879] Step 3:

[1880] The server performs data cleansing on the data received.

[1881] (input)

[1882] Gaze and expression data in JSON format.

[1883] (process)

[1884] The server cleanses the data by imputing missing values ​​and handling outliers.

[1885] (output)

[1886] Cleansed gaze and facial expression data.

[1887] Step 4:

[1888] The server analyzes consumer emotions from gaze and facial expression data.

[1889] (input)

[1890] Cleansed gaze and facial expression data.

[1891] (process)

[1892] Emotions are analyzed using software such as OpenCV, dlib, and GazeTracking. Specifically, changes in gaze direction and facial expressions are detected to determine the emotional state.

[1893] (output)

[1894] Consumer sentiment data.

[1895] Step 5:

[1896] The server generates a virtual consumer model based on the analyzed data.

[1897] (input)

[1898] Analyzed consumer sentiment data and consumption behavior data.

[1899] (process)

[1900] Use machine learning algorithms to create models that predict the relationship between consumer purchasing behavior and emotions.

[1901] (output)

[1902] Virtual consumer model.

[1903] Step 6:

[1904] The server integrates the generated virtual consumer models and constructs a virtual market.

[1905] (input)

[1906] Multiple virtual consumer models.

[1907] (process)

[1908] The server integrates multiple virtual consumer models to create a virtual market, which includes supply and demand for each product or service, competitive information, and consumer sentiment data.

[1909] (output)

[1910] Hypothetical market data.

[1911] Step 7:

[1912] The user sets up a business scenario on the terminal.

[1913] (input)

[1914] Scenario data for new product introductions, price changes, advertising campaigns, etc.

[1915] (process)

[1916] The user uses the interface on the terminal to input detailed information about the scenario and transmits it to the server.

[1917] (output)

[1918] The scenario data is sent to the server.

[1919] Step 8:

[1920] The server analyzes the received scenario data and executes a simulation.

[1921] (input)

[1922] Scenario and hypothetical market data.

[1923] (process)

[1924] Based on a set scenario, a simulation is run to predict fluctuations in a virtual market.

[1925] (output)

[1926] Simulation result data.

[1927] Step 9:

[1928] The server outputs the simulation results in JSON format, and the terminal receives and visualizes them.

[1929] (input)

[1930] Simulation result data.

[1931] (process)

[1932] The server outputs the simulation results in JSON format, and the terminal receives them and visualizes them in graphs or tables.

[1933] (output)

[1934] Visualized simulation results.

[1935] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1936] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1937] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1938] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1939] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right...

Claims

1. A means for receiving and analyzing actual consumer behavior data; A means for generating a virtual consumer model based on the analyzed data; A means for integrating the generated virtual consumer model and constructing a virtual market; a means for simulating business scenarios within the constructed virtual market; The system includes a means for outputting simulation results.

2. 2. The system according to claim 1, wherein the actual consumer behavior data is data obtained from a customer relationship management system.

3. 2. The system of claim 1, wherein the business scenario includes one of a new product introduction, a price change, and an advertising campaign.

Images