System

Abstract

To provide a system.SOLUTION: A system includes: means for using data for mimicking a user's behavior pattern and preference to generate a pseudo user; means for inputting a specific prompt sentence into a generative AI model to execute a simulation for verifying whether or not a new business service is accepted by the pseudo user; verification means for inputting a simulation result into the generative AI model to verify the validity of a new business service; means for recognizing an emotional state of the pseudo-user using an emotion engine to adjust a behavior pattern of the pseudo-user based on the recognized emotional state; means for simulating the behavior pattern of the adjusted pseudo user to collect the behavior data of the pseudo user; means for analyzing the collected behavior data by using a clustering algorithm; and means for outputting the result of the analysis as a report.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] One of the challenges when considering new business services is verifying the effectiveness of the service when considering ideas. Large-scale surveys require budget and cost, and it is difficult to ensure that people understand the service correctly. [Means for solving the problem]

[0005] The system of the present disclosure includes: a means for generating a pseudo user using data for imitating a user's behavioral patterns and preferences; a means for executing a simulation to verify whether a new business service will be accepted by the pseudo user by inputting a specific prompt sentence into a generation AI model; a verification means for confirming the effectiveness of the new business service by inputting the results of the simulation into the generation AI model; a means for recognizing the emotional state of the pseudo user using an emotion engine and adjusting the behavioral pattern of the pseudo user based on the recognized emotional state; a means for collecting behavioral data of the pseudo user by simulating the adjusted behavioral pattern of the pseudo user; a means for analyzing the collected behavioral data using a clustering algorithm; and a means for outputting the results of the analysis as a report. [Brief explanation of the drawings]

[0006] [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. 2 is a sequence diagram showing a flow of processing in the data processing system according to the first embodiment of the first form example. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1 of Embodiment 1. [Figure 13] FIG. 10 is a sequence diagram showing a processing flow of a data processing system in a second embodiment of the second form example. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 of Embodiment Example 2. [Figure 15] FIG. 10 is a sequence diagram showing the flow of processing in a data processing system according to a third embodiment of the third embodiment. [Figure 16] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 3 of Embodiment 3. [Figure 17] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the first embodiment of the first form example when an emotion engine is combined. [Figure 18] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1 of Form Example 1 when an emotion engine is combined. [Figure 19] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the second embodiment of the second form example when an emotion engine is combined. [Figure 20] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 of Form Example 2 when an emotion engine is combined. [Figure 21] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the third embodiment of the third form example when an emotion engine is combined. [Figure 22] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 3 of Form Example 3 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0007] 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.

[0008] First, the terms used in the following description will be explained.

[0009] 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, the 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), an APU (Accelerated Processing Unit), or a TPU (TENSOR PROCESSING UNIT (registered trademark)).

[0010] 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.

[0011] 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.

[0012] 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.

[0013] 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."

[0014] [First embodiment]

[0015] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0016] 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.

[0017] 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).

[0018] 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.

[0019] 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.

[0020] 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 can be perceived by the user 20 (for example, voice and / or text).

[0021] The display 40A displays visual information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs sound in accordance with instructions from the processor 46.

[0022] The camera 42 is a small digital camera equipped with an optical system including a lens, an aperture, a shutter, and the like, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0023] 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.

[0024] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0025] 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.

[0026] 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.

[0027] 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.

[0028] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0029] "Example 1"

[0030] One embodiment of the present invention is a method for generating a large number of pseudo-users using a generative AI. Specifically, the generative AI learns data to imitate user behavior patterns and preferences, and generates pseudo-users based on that data. For example, past user behavior data and user profile data can be used as learning data.

[0031] "Example 2"

[0032] Furthermore, there is a means for executing a simulation to verify whether a new business service will be accepted by pseudo-users. Specifically, the new business service is provided to the generated pseudo-users, and their reactions are simulated. For example, if the new business service is an online shopping service, it is possible to simulate which products the pseudo-users will select and what purchasing behavior they will exhibit.

[0033] "Example 3"

[0034] There are also methods to verify the effectiveness of new business services based on the results of simulations. Specifically, the results of the simulations can be used to analyze the acceptability of new business services and areas for improvement. For example, based on the reactions of simulated users, it is possible to identify areas for improvement such as pricing and delivery methods for new business services, and selection of target users.

[0035] The processing flow of each embodiment will be described below.

[0036] "Example 1"

[0037] Step 1: The generative AI learns data to mimic user behavior patterns and preferences. This learning data is based on past user behavior data and user profile data.

[0038] Step 2: Based on the learned data, the generative AI generates pseudo-users, who are then used to verify the acceptability of the new business service.

[0039] "Example 2"

[0040] Step 1: Provide a new business service to the generated pseudo-users. This new business service may be, for example, an online shopping service.

[0041] Step 2: Simulate the reaction of the pseudo-user. Specifically, simulate which products the pseudo-user will select and what purchasing behavior they will exhibit.

[0042] "Example 3"

[0043] Step 1: Analyze the acceptability of the new business service and areas for improvement based on the results of the simulation.

[0044] Step 2: Based on the analysis results, identify areas for improvement such as pricing and delivery methods for the new business service, and selection of target users.

[0045] Example 1

[0046] Next, a description will be given of Example 1 of Form 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."

[0047] Conventional methods for testing new business services involve using actual user data, which poses challenges such as privacy issues and the high cost of data collection. Furthermore, testing based on actual user data takes time, and it is often difficult to obtain prompt feedback. Furthermore, it is difficult to accurately mimic user behavior patterns and preferences, resulting in low simulation accuracy.

[0048] 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.

[0049] In this invention, the server includes: means for generating a large number of pseudo-users using a generative AI model; means for collecting and preprocessing past user behavior data and user profile data; means for training the generative AI model based on the preprocessed data; means for generating pseudo-users by inputting prompt sentences to the trained generative AI model; means for running a simulation to verify whether a new business service will be accepted by the pseudo-users; means for verifying the effectiveness of the new business service based on the results of the simulation; and means for saving the generated pseudo-user data and using it in other systems or applications. This enables rapid and highly accurate verification of new business services while avoiding privacy issues.

[0050] A "generative AI model" is an artificial intelligence model that learns user behavior patterns and preferences and generates pseudo-users based on them.

[0051] A "pseudo-user" is a virtual user created using a generative AI model that mimics the behavioral patterns and preferences of a real user.

[0052] "User behavior data" is data related to past behavior of a user, and includes, for example, purchase history and browsing history.

[0053] "User profile data" refers to data relating to a user's attributes and preferences, including, for example, age, gender, and interests.

[0054] "Preprocessing" refers to the process of converting collected data into a format suitable for a generative AI model, and includes data cleaning, normalization, and feature extraction.

[0055] A "prompt sentence" is text that is input into a generative AI model and contains instructions and conditions for generating a pseudo-user.

[0056] The "simulation" is a virtual experiment using the generated pseudo-users to verify whether the new business service will be accepted. The server may include means for executing a simulation to verify whether the new business service will be accepted by the pseudo-users by inputting a specific prompt sentence into the generative AI model.

[0057] "New business services" refer to new products or services offered by companies or organizations.

[0058] "Effectiveness" refers to the ability or effectiveness of the new business service to achieve the target results. The server may include a verification means for verifying the effectiveness of the new business service by inputting the results of the simulation into the generative AI model.

[0059] "Saving" means recording the data of the generated pseudo-user in a database or the like so that it can be used later.

[0060] "Usage" means utilizing the saved pseudo-user data in other systems or applications.

[0061] This invention relates to a system for generating a large number of pseudo-users using a generative AI model and verifying whether a new business service will be accepted by the pseudo-users. A specific embodiment of this system will be described below.

[0062] The server first collects past user behavior data and user profile data from a database. This data includes information such as the actions users have taken and their preferences. The collected data is preprocessed using Python's Pandas library, among other tools. Preprocessing includes data cleaning, normalization, and feature extraction.

[0063] The server then trains a generative AI model based on the preprocessed data. An example of a generative AI model that can be used is OpenAI's GPT-4. The trained model has the ability to learn user behavior patterns and preferences and generate pseudo-users based on them.

[0064] Once training is complete, the server inputs a prompt to the generative AI model. The prompt provides the generative AI model with instructions and conditions for generating a pseudo-user. For example, the following prompts can be used:

[0065] "User A frequently shops online on weekday evenings. He tends to buy mainly electronic devices and books. His purchase history includes the latest smartphones and best-selling novels. Please generate a pseudo-user that mimics User A's behavioral patterns."

[0066] When this prompt is entered, the generative AI model generates a pseudo-user that mimics the behavioral patterns and preferences of User A. The generated pseudo-user is expected to behave in a similar way to User A.

[0067] Using the generated pseudo-users, the server runs a simulation to verify whether the new business service will be accepted by the pseudo-users. Based on the results of the simulation, the effectiveness of the new business service is confirmed, and areas for improvement or strengthening are identified as necessary.

[0068] Finally, the server stores the generated pseudo-user data in a database and makes it available to other systems and applications, enabling rapid and highly accurate verification of new business services while avoiding privacy issues.

[0069] The above is a specific embodiment for carrying out the present invention.

[0070] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0071] Step 1: Data collection

[0072] The server retrieves past user behavior data and user profile data from the database. As input, it uses database connection information and SQL queries. As output, it obtains the retrieved raw data. Specifically, the server executes the SQL queries to extract user behavior history and profile information.

[0073] Step 2: Data Preprocessing

[0074] The server preprocesses the acquired raw data and converts it into a format suitable for generative AI models. The acquired raw data is used as input, and preprocessed data is obtained as output. Specifically, the server cleans the data (filling in missing values ​​and removing outliers), normalizes it, and extracts features.

[0075] Step 3: Train the model

[0076] The server trains a generative AI model based on the preprocessed data. As input, it uses the preprocessed data and a generative AI model (e.g., GPT-4). As output, it obtains a trained generative AI model. Specifically, the server inputs the preprocessed data into the model and optimizes the model's parameters.

[0077] Step 4: Enter the prompt statement

[0078] The server inputs a prompt sentence into the trained generative AI model. The prompt sentence and the trained generative AI model are used as input. The generated pseudo-user is obtained as output. Specifically, the server inputs the prompt sentence into the model, and the model generates a pseudo-user.

[0079] Step 5: Run the simulation

[0080] The server uses the generated pseudo-users to run a simulation to verify whether the new business service will be accepted. The generated pseudo-users and the scenario of the new business service are used as input. The simulation results are obtained as output. Specifically, the server simulates the behavior of the pseudo-users and evaluates the acceptance of the service.

[0081] Step 6: Saving and using the results

[0082] The server stores the generated pseudo-user data and simulation results, making them available to other systems and applications. The generated pseudo-user data and simulation results are used as input, and the stored data is obtained as output. Specifically, the server stores the data in a database and provides the data to other systems through an API.

[0083] The above are the specific processing steps of the program of this system.

[0084] (Application example 1)

[0085] Next, a description will be given of Application Example 1 of Embodiment 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."

[0086] Conventional methods for testing new business services require the use of actual user data, which is time-consuming and costly, and also poses problems from the perspective of protecting user privacy. It is also difficult to accurately grasp user behavioral patterns and preferences, making it difficult to provide personalized services. Furthermore, product recommendation systems are required to make highly accurate recommendations based on user preferences, but current systems are unable to adequately meet this demand.

[0087] 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.

[0088] In this invention, the server includes means for generating a large number of pseudo users using generative AI, means for running a simulation to verify whether a new business service will be accepted by the pseudo users, means for verifying the effectiveness of the new business service based on the results of the simulation, means for recommending products based on the behavioral data of the pseudo users, and means for providing a personalized shopping experience based on user preferences. This enables efficient and accurate verification of new business services and personalized product recommendations while protecting user privacy.

[0089] "Generative AI" is an artificial intelligence technology that learns user behavior patterns and preferences and generates pseudo-users based on them.

[0090] A "pseudo-user" is a virtual user created by generative AI that mimics the behavioral patterns and preferences of a real user.

[0091] A "simulation" is a virtual experiment using pseudo-users to verify the extent to which a new business service will be accepted.

[0092] "New business services" are new products or services offered by a company, which provide new value to users.

[0093] "Effectiveness" refers to the ability of a new business service to be accepted by users and to achieve expected results.

[0094] "Behavioral Data" is historical information about the actions and choices a user makes on a website or application.

[0095] "Product recommendation" refers to proposing appropriate products to users based on their behavioral data and preferences.

[0096] A "personalized shopping experience" is a shopping experience that provides optimal products and services based on the user's individual preferences and behavioral patterns.

[0097] A system for implementing this invention includes means for generating a large number of pseudo-users using generative AI, means for running a simulation to verify whether a new business service will be accepted by the pseudo-users, means for confirming the effectiveness of the new business service based on the results of the simulation, means for recommending products based on behavioral data of the pseudo-users, and means for providing a personalized shopping experience based on user preferences.

[0098] System Program

[0099] The server generates pseudo-users using a generative AI model. The generative AI model learns past user behavior data and user profile data, and generates pseudo-users based on that data. The generated pseudo-users mimic the behavioral patterns and preferences of actual users.

[0100] Next, the server executes a simulation to verify whether the new business service will be accepted by the pseudo-users. This simulation makes it possible to confirm the effectiveness of the new business service.

[0101] Furthermore, the server recommends products based on the behavioral data of the simulated user, thereby providing a personalized shopping experience based on the user's preferences.

[0102] Hardware and software used

[0103] Hardware: Servers, smartphones

[0104] Software: Python, generative AI models (e.g., GPT-3 (registered trademark))

[0105] Data processing and calculation

[0106] 1. Data collection: The server collects past user behavior data and user profile data.

[0107] 2. Data training: The server trains the generative AI model on this data.

[0108] 3. Pseudo-user generation: The server generates pseudo-users using the trained model.

[0109] 4. Simulation execution: The server executes a simulation of the new business service using a pseudo user.

[0110] 5. Product recommendation: The server recommends products based on the behavioral data of the pseudo-user.

[0111] 6. Personalized Offering: The server provides a personalized shopping experience based on the user's preferences.

[0112] Specific examples

[0113] For example, if a user is looking for "smartphone accessories," the server will refer to the pseudo-user's behavioral data and recommend related products (cases, chargers, earphones, etc.).

[0114] Prompt Sentence Examples

[0115] Generate pseudo-users based on user behavior data and recommend products based on the user's preferences. For example, recommend related products based on the data of users who have previously purchased "smartphone accessories."

[0116] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0117] Step 1:

[0118] The server collects historical user behavior and user profile data, including the user's actions and selections on websites and applications. The input is the user's behavior and profile data, and the output is the data stored on the server. Specifically, the server retrieves the required data by querying the database.

[0119] Step 2:

[0120] The server trains the collected data into a generative AI model. The inputs are the user behavior data and profile data collected in step 1, and the output is a trained generative AI model. Specifically, the server preprocesses the data and feeds it to the generative AI model.

[0121] Step 3:

[0122] The server generates a pseudo-user using a trained generative AI model. The input is the trained generative AI model, and the output is the generated pseudo-user data. Specifically, the server inputs prompts into the generative AI model to generate the pseudo-user's profile and behavioral patterns.

[0123] Step 4:

[0124] The server runs a simulation of the new business service using the generated pseudo-user. The inputs are the pseudo-user data and the specifications of the new business service, and the output is the simulation results. Specifically, the server simulates how the pseudo-user uses the new business service and records the results.

[0125] Step 5:

[0126] The server checks the effectiveness of the new business service based on the simulation results. The input is the simulation results, and the output is an evaluation of the effectiveness of the new business service. Specifically, the server analyzes the simulation results and identifies the degree of acceptance of the service and areas for improvement.

[0127] Step 6:

[0128] The server recommends products based on the behavioral data of the pseudo-user. The input is the behavioral data of the pseudo-user, and the output is a list of recommended products. Specifically, the server analyzes the pseudo-user's preferences and behavioral patterns to select the most suitable products.

[0129] Step 7:

[0130] The server provides a personalized shopping experience based on the user's preferences. The inputs are a list of recommended products and the user's preference data, and the output is a personalized shopping experience. Specific operations include displaying products that match the user's preferences and providing an interface that encourages purchases.

[0131] Example 2

[0132] Next, a description will be given of Example 2 of Form 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."

[0133] Before launching a new business service to the market, it is important to verify whether the service will be accepted by users. However, testing with actual users is time-consuming and costly, so an efficient method is required. In addition, it is difficult to accurately imitate user behavior patterns and preferences, which can result in low simulation accuracy. To solve these issues, a simulation system using pseudo-users is needed.

[0134] 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.

[0135] In this invention, the server includes means for generating a large number of pseudo users using generative artificial intelligence, means for running a simulation to verify whether a new business service will be accepted by the pseudo users, means for verifying the effectiveness of the new business service based on the results of the simulation, means for providing the new business service to the generated pseudo users and simulating their reactions, and means for displaying the results of the simulation to users. This makes it possible to efficiently and accurately verify the acceptability of a new business service before it is released to the market.

[0136] "Generative AI" is a system that generates data and information using artificial intelligence technology.

[0137] A "pseudo user" is a virtual user created by imitating the behavioral patterns and preferences of an actual user.

[0138] "New business services" refer to products or services that are planned to be newly introduced to the market.

[0139] A "simulation" is a virtual experiment using pseudo-users to verify the acceptability of a new business service.

[0140] "Simulation results" refers to data and information obtained after running a simulation.

[0141] "User" refers to a person or organization that uses this system to verify new business services.

[0142] "Generating means" refers to a method or device for creating a pseudo-user using generative artificial intelligence.

[0143] The "means for providing" refers to a method or device for presenting the new business service to the pseudo-user.

[0144] The "display means" refers to a method or device for visually showing the simulation results to the user.

[0145] This invention is a system that uses generative artificial intelligence to generate pseudo-users and simulates whether a new business service will be accepted by the pseudo-users. A specific embodiment of this system will be described below.

[0146] The server first generates a pseudo-user using a generative AI model (e.g., a general natural language processing model). Examples of generative AI models used include OpenAI's GPT-4. The generated pseudo-users have attributes such as age, gender, and interests.

[0147] Next, the server provides the new business service to the simulated user. If this service is an online shopping service, the server presents a product list to the simulated user. The simulated user simulates which product to select and what purchasing behavior to take through the generative AI model.

[0148] The terminal provides an interface that allows the user to check the simulation results, and the user can find areas for improvement in the new business service based on the simulation results.

[0149] As a specific example, if the new business service is an "online shopping service," the following prompt sentence would be input into the generative AI model:

[0150] Example prompt sentence:

[0151] Pseudo-user A is a 30-year-old male who is interested in technology products. Please present him with the following product list and simulate which product he would choose.

[0152] 1. Smartphone

[0153] 2. Laptop

[0154] 3. Smartwatch

[0155] 4. Wireless earphones

[0156] The server inputs this prompt into the generative AI model and simulates the purchasing behavior of pseudo-user A. For example, if pseudo-user A selects "smartphone," the reason for the selection and the degree of interest in other products are also output as simulation results.

[0157] Users can check the simulation results on their devices and identify areas for improvement in their new business services. For example, if the adoption rate of smartwatches is low, they can take measures such as revising the product description or pricing.

[0158] This system makes it possible to efficiently and accurately verify the acceptability of new business services before they are brought to market.

[0159] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0160] Step 1:

[0161] The server generates a pseudo-user using a generative AI model. As input, it inputs a prompt statement specifying the pseudo-user's attributes (e.g., age, gender, interests) into the generative AI model. The generative AI model generates a pseudo-user based on this prompt statement and obtains the pseudo-user's data as output. Specifically, the server inputs the following prompt statement into the generative AI model:

[0162] Generate a simulated user who is a 30-year-old male and interested in technology products.

[0163] The generative AI model generates a pseudo-user based on this prompt sentence.

[0164] Step 2:

[0165] The server provides a new business service to the generated pseudo-user. As input, it uses the pseudo-user data and new business service information (e.g., product list). The server presents the product list to the pseudo-user and obtains the pseudo-user's reaction data as output. Specifically, the server presents the following product list to the pseudo-user.

[0166] 1. Smartphone

[0167] 2. Laptop

[0168] 3. Smartwatch

[0169] 4. Wireless earphones

[0170] Step 3:

[0171] The server uses a generative AI model to simulate the pseudo-user's reaction. The pseudo-user's data and product list are used as input. The generative AI model simulates which products the pseudo-user will select and what purchasing behavior they will take, and obtains the simulation results as output. Specifically, the server inputs the following prompt sentence into the generative AI model:

[0172] Pseudo-user A is a 30-year-old male who is interested in technology products. Please present him with the following product list and simulate which product he would choose.

[0173] 1. Smartphone

[0174] 2. Laptop

[0175] 3. Smartwatch

[0176] 4. Wireless earphones

[0177] The generative AI model simulates the purchasing behavior of pseudo-user A based on this prompt sentence.

[0178] Step 4:

[0179] The terminal provides an interface that allows the user to check the simulation results. The simulation results are used as input. The terminal visually displays the simulation results and provides a state that allows the user to check the results as output. Specifically, the terminal displays the simulation results as follows:

[0180] Pseudo-user A chose "Smartphone." Reason for choosing: Because he is interested in the latest technology.

[0181] Based on these results, users can consider ways to improve their product listings and service content.

[0182] (Application example 2)

[0183] Next, a description will be given of Application Example 2 of Form 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."

[0184] It is an important task for companies to verify in advance whether a new business service will be accepted by the market. However, testing using actual users is costly and time-consuming. It is also difficult to accurately grasp user behavior patterns and preferences. This has led to a demand for a method to efficiently confirm the effectiveness of new business services and identify areas for improvement and strengthening.

[0185] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for generating a large number of pseudo users using generative AI, means for running a simulation to verify whether a new business service will be accepted by the pseudo users, means for confirming the effectiveness of the new business service based on the results of the simulation, means for simulating the products selected and purchasing behavior of the pseudo users, and means for analyzing the simulation results and identifying which products were most selected. This makes it possible to efficiently verify the market acceptance of a new business service and identify areas for improvement and enhancement.

[0186] "Generative AI" is a system that generates data and information using artificial intelligence technology.

[0187] A "pseudo user" is a virtual user created by imitating the behavioral patterns and preferences of an actual user.

[0188] "Simulation" is a method of imitating and virtually reproducing real-world events and processes.

[0189] "New business services" refer to new products or services that a company offers.

[0190] "Effectiveness" is an indicator of how effective something is in achieving a specific purpose or goal.

[0191] "Purchasing behavior" refers to the series of actions and decision-making process that consumers take from selecting a product to purchasing it.

[0192] Analysis is the process of examining data or information in detail to find meaning and patterns.

[0193] "Areas for improvement" are areas in the current system or process that need to be changed or modified to make it better.

[0194] "Enhancements" are areas of the current system or process that need further strengthening or expansion.

[0195] A system for implementing this invention includes means for generating a large number of pseudo-users using generative AI, means for running a simulation to verify whether a new business service will be accepted by the pseudo-users, means for confirming the effectiveness of the new business service based on the results of the simulation, means for simulating the products selected and purchasing behavior of the pseudo-users, and means for analyzing the results of the simulation and identifying which products were most selected.

[0196] The server uses generative AI to generate pseudo-users. These pseudo-users mimic the behavioral patterns and preferences of real users, enabling a realistic simulation. The generated pseudo-users are used to simulate reactions to new business services.

[0197] The simulation is run on a server, reproducing which products a simulated user would select and what purchasing behavior they would exhibit. The results of the simulation are collected and analyzed by the server. Based on the analysis results, the effectiveness of the new business service is confirmed and areas for improvement and enhancement are identified.

[0198] The implementation of this system uses the following hardware and software:

[0199] Hardware: Servers, smartphones

[0200] Software: Python, NumPy, generative AI models

[0201] As a concrete example, the following prompt sentence is input into the generative AI model:

[0202] Example prompt sentence:

[0203] To simulate adoption of a new online shopping service, create a simulated user and simulate their choice of the following products:

[0204] Product A: Price 100 yen, quality 80

[0205] Product B: Price 150 yen, quality 90

[0206] Product C: Price 200 yen, quality 85

[0207] The pseudo-user preferences are as follows:

[0208] User 1: Price-sensitive (-0.5), Quality-sensitive (1.0)

[0209] User 2: Price-sensitive (-0.3), Quality-sensitive (0.8)

[0210] Please let me know the simulation results.

[0211] By inputting this prompt into a generative AI model, it is possible to simulate the purchasing behavior of pseudo-users and verify their acceptance of new services.

[0212] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0213] Step 1:

[0214] The server uses generative AI to generate pseudo-users. Data on user behavior patterns and preferences is used as input. Based on this data, the generative AI generates a large number of pseudo-users that mimic the behavior of actual users. The output is a list of the generated pseudo-users.

[0215] Step 2:

[0216] The server runs a simulation to provide a new business service to the generated pseudo-users. As input, it uses a list of pseudo-users and detailed information about the new business service. The simulation reproduces which products the pseudo-users select and what purchasing behavior they exhibit. The output is data on the products selected and purchasing behavior of each pseudo-user.

[0217] Step 3:

[0218] The server collects and analyzes the results of the simulation. It uses the simulation result data as input. It uses data analysis tools to identify which products were most popular and what purchasing behaviors were most common. The output is a report of the analysis results.

[0219] Step 4:

[0220] The server checks the effectiveness of the new business service based on the analysis results. It uses the analysis result report as input. It evaluates whether the new business service will be accepted by the market and identifies areas for improvement or enhancement, if necessary. The output is an evaluation report of the new business service.

[0221] Step 5:

[0222] The server identifies areas for improvement and enhancement of the new business service based on the evaluation report and reflects them in the next simulation. The evaluation report is used as input. A new simulation reflecting the improvements and enhancements is prepared and verified again using simulated users. The output is the simulation results of the improved new business service.

[0223] Example 3

[0224] Next, a description will be given of a third embodiment of the third embodiment. 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."

[0225] There is a lack of effective methods to verify whether a new business service will be accepted by target users before it is launched on the market. It is also difficult to identify specific areas for improvement or strengthening based on simulation results. This makes it difficult to obtain appropriate feedback to increase the success rate of the business.

[0226] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[0227] In this invention, the server includes means for generating a large number of pseudo users using generative AI, means for running a simulation to verify whether a new business service will be accepted by the pseudo users, means for collecting simulation result data, means for inputting the collected data into a generative AI model and analyzing the data, means for feeding back the analysis results output from the generative AI model to users, and means for verifying the effectiveness of the new business service based on the simulation results. This makes it possible to effectively verify the acceptability of a new business service before it is released to the market and identify specific areas for improvement and enhancement.

[0228] "Generative AI" is an artificial intelligence technology that generates new data and models based on large amounts of data.

[0229] A "pseudo-user" is a virtual user created to mimic the behavior and preferences of a real user.

[0230] "Simulation" is a technology that imitates real-world systems and processes and virtually reproduces their behavior.

[0231] "Simulation result data" refers to various data and information obtained when a simulation is executed.

[0232] A "generative AI model" is a model that uses generative AI to perform specific tasks or generate data.

[0233] "Data analysis" is the process of extracting useful information and insights from collected data.

[0234] "Feedback" refers to returning the results or evaluation of a system or process to the user.

[0235] "New business services" are services or products that are planned to be newly introduced to the market.

[0236] "Effectiveness" is an indicator of how effective something is in achieving a specific purpose or goal.

[0237] "Areas for improvement" are parts or elements of the current system or process that need improvement.

[0238] "Improvement points" are parts or elements of the current system or process that need to be further strengthened.

[0239] The present invention provides a system for effectively verifying the acceptability of a new business service before it is launched into the market and identifying specific areas for improvement or strengthening. Specific embodiments of this system will be described below.

[0240] Hardware and software used

[0241] Hardware: Servers, terminals

[0242] Software: Generative AI models (e.g., GPT-4), simulation software (e.g., AnyLogic)

[0243] System configuration

[0244] 1. The server uses generative AI to generate a large number of pseudo-users. Pseudo-users are virtual users created to mimic the behavior and preferences of real users.

[0245] 2. The server runs a simulation to verify whether the new business service will be accepted by the pseudo-users. Using simulation software (e.g., AnyLogic), the behavioral patterns and reactions of the pseudo-users are modeled.

[0246] 3. The server collects the simulation results data, which are stored in a database and used for subsequent analysis.

[0247] 4. The server inputs the collected data into a generative AI model (e.g., GPT-4) and analyzes the data. The generative AI model uses the simulation results to identify areas for improvement, such as pricing for the new business service, delivery method, and selection of target users.

[0248] 5. The server sends the analysis results output from the generative AI model to the terminal.

[0249] 6. The terminal displays the analysis results to the user, who can then check areas for improvement in the new business service based on the displayed analysis results and make any necessary corrections.

[0250] Specific examples

[0251] For example, consider the case of launching an "online learning platform" as a new business service.

[0252] 1. The server launches AnyLogic and selects the simulation model of the "Online Learning Platform."

[0253] 2. The server inputs the pricing as "1,000 yen per month," the delivery method as "mobile app," and the target users as "working adults in their 20s and 30s."

[0254] 3. The server sets scenarios such as "study on weekday evenings" and "study all at once on weekends" as the behavioral patterns of the pseudo-users.

[0255] 4. The server executes the configured simulation model and simulates how a pseudo-user uses the platform.

[0256] 5. The server records the user's response to each scenario (e.g., study time, frequency of use, satisfaction, etc.).

[0257] 6. The server collects the simulation result data (e.g., study time, frequency of use, satisfaction level, etc.) and stores it in a database.

[0258] 7. The server converts the collected data into CSV format and formats it so that it can be input into the generative AI model.

[0259] 8. The server inputs the formatted data into GPT-4.

[0260] 9. From the simulation results, GPT-4 identifies areas for improvement, such as "the optimal pricing would be 1,000 yen per month," "the preferred delivery method is a mobile app," and "the optimal target users are working adults in their 20s and 30s."

[0261] 10. GPT-4 outputs the analysis results in text format.

[0262] 11. The server sends the analysis results output by GPT-4 to the terminal.

[0263] 12. The device displays analysis results to the user, such as "The optimal price setting is 1,000 yen per month," "The preferred delivery method is a mobile app," and "The optimal target users are working adults in their 20s and 30s."

[0264] 13. The user adjusts the pricing and delivery methods of the online learning platform based on the displayed analysis results.

[0265] Prompt Sentence Examples

[0266] "If you're launching an online learning platform as a new business service, use the simulation results to identify areas for improvement in pricing, delivery, and target users."

[0267] In this way, the system can confirm the effectiveness of the new business service based on the simulation results and identify specific areas for improvement and strengthening. The flow of the identification process in the third embodiment will be described with reference to FIG.

[0268] Step 1:

[0269] The server uses generative AI to generate a large number of pseudo-users.

[0270] Input: Data to mimic user behavior patterns and preferences

[0271] Data processing: Generative AI generates pseudo-user profiles based on input data.

[0272] Output: Pseudo user profile data

[0273] Specific operation: The server starts the generative AI model and inputs the user's behavioral patterns and preference data. The generative AI model generates a pseudo-user profile based on this data and stores it in a database.

[0274] Step 2:

[0275] The server executes a simulation to verify whether the new business service will be accepted by the pseudo-users.

[0276] Input: Pseudo user profile data, parameters of new business services (e.g., pricing, delivery method, target user attributes)

[0277] Data calculation: Simulation software evaluates the likelihood of new business services being accepted based on the behavioral patterns of simulated users.

[0278] Output: Simulation result data (e.g., study time, frequency of use, satisfaction, etc.)

[0279] Specific operation: The server starts the simulation software and inputs the profile data of the pseudo-user and the parameters of the new business service. The simulation software runs the simulation based on this data and generates the result data.

[0280] Step 3:

[0281] The server collects the result data of the simulation.

[0282] Input: Simulation result data

[0283] Data processing: The resulting data is stored in a database and formatted so that it can be input into the generative AI model.

[0284] Output: Formatted data

[0285] Specific operation: The server retrieves the result data from the simulation software, stores it in a database, and then converts it into CSV format, which is a format that can be input into the generative AI model.

[0286] Step 4:

[0287] The server inputs the collected data into a generative AI model and analyzes the data.

[0288] Input: Formatted data

[0289] Data calculation: Generative AI models use simulation results to identify areas for improvement in new business services, such as pricing, delivery methods, and target user selection.

[0290] Output: Analysis results (e.g., areas for improvement and enhancement)

[0291] How it works: The server inputs the formatted data into the generative AI model, which analyzes the data and identifies areas for improvement and strengthening. The analysis results are output in text format.

[0292] Step 5:

[0293] The server transmits the analysis results output from the generative AI model to the terminal.

[0294] Input: Analysis results

[0295] Data processing: Converting the analysis results into a format that is easy for users to understand.

[0296] Output: Feedback data

[0297] Specific operation: The server receives the analysis results output by the generative AI model, converts them into a format that is easy for the user to understand, and then sends the feedback data to the device.

[0298] Step 6:

[0299] The terminal displays the analysis results to the user.

[0300] Input: Feedback data

[0301] Data processing: Converting feedback data into a display format.

[0302] Output: Displayed analysis results

[0303] Specific operation: The terminal receives the feedback data from the server and displays it to the user. The user can then check the areas for improvement in the new business service based on the displayed analysis results and make any necessary corrections.

[0304] (Application example 3)

[0305] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[0306] When developing a new business service, it is extremely important to evaluate in advance whether the service will be accepted by the market. However, it is difficult to predict actual market reactions, and there is a lack of concrete methods for optimizing elements such as pricing, delivery methods, and target user selection. For this reason, there is a need for a system that can perform simulations in advance, confirm the effectiveness of new business services based on the reactions of simulated users, and identify areas for improvement.

[0307] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means. In this invention, the server includes means for generating a large number of pseudo users using generative AI, means for running a simulation to verify whether a new business service will be accepted by the pseudo users, means for confirming the effectiveness of the new business service based on the results of the simulation, means for identifying areas for improvement such as pricing, provision method, and selection of target users based on the results of the simulation, and means for proposing optimal pricing and provision method. This makes it possible to predict and optimize market acceptance of the new business service.

[0308] "Generative AI" is a system that uses artificial intelligence technology to generate new data from large amounts of data.

[0309] A "pseudo user" is a virtual user created by imitating the behavior and preferences of an actual user.

[0310] "Simulation" is a method of virtually recreating a real-world situation and analyzing the results.

[0311] "New business services" refer to products and services that are newly introduced to the market.

[0312] "Effectiveness" is an indicator of how effective something is for a specific purpose.

[0313] "Pricing" is the process of determining the selling price of a product or service.

[0314] "Delivery method" refers to the means or format by which a product or service is delivered to a user.

[0315] "Target users" are the user group that is expected to use a particular product or service.

[0316] "Improvements" are specific changes that will resolve current problems and improve the situation.

[0317] A "proposal" is the act of suggesting a solution or improvement to a specific problem.

[0318] The system for implementing this invention consists of three main elements: a server, a terminal, and a user. The following explains how each element functions.

[0319] server

[0320] The server uses generative AI to generate a large number of pseudo-users. These pseudo-users are generated based on data that mimics the behavioral patterns and preferences of actual users. These pseudo-users are used in simulations to verify whether new business services will be accepted.

[0321] The results of the simulation are analyzed on the server and used to confirm the effectiveness of the new business service. Specifically, the simulation results are used to identify areas for improvement such as pricing, delivery methods, and target user selection, and optimal pricing and delivery methods are proposed.

[0322] Terminal

[0323] The terminal provides an interface for users to access the system. Mobile devices such as smartphones and tablets are mainly used. The application on the terminal displays the simulation results and proposals sent from the server and helps users make decisions based on them.

[0324] User

[0325] Users can access the system via their devices and check the simulation results and proposals for new business services. Based on this information, users can identify areas for improvement in services and formulate optimal strategies before actually launching them on the market.

[0326] Hardware and software used

[0327] Hardware: Servers (with high-performance computing power), smartphones, tablets

[0328] Software: Python, Pandas (data processing library), NumPy (numerical calculation library), Scikit-learn (machine learning library)

[0329] Specific examples

[0330] For example, consider the case of simulating pricing for a new product, a "smartwatch," and identifying the optimal price range. The server uses generative AI to generate pseudo-users and run the simulation. Based on the simulation results, it identifies areas for improvement in pricing and delivery methods and proposes the optimal price range.

[0331] Prompt Sentence Examples

[0332] "Please simulate pricing for a new product, a smartwatch, and identify the optimal price point. Also, please analyze the acceptability of a subscription model versus a one-time purchase model."

[0333] In this way, a system is realized in which servers, terminals, and users work together to predict and optimize market acceptance of new business services.

[0334] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[0335] Step 1:

[0336] The server uses generative AI to generate a large number of pseudo-users. As input, it uses data to imitate user behavior patterns and preferences. As data processing, it creates pseudo-user profiles based on this data. The output is a list of the generated pseudo-users.

[0337] Step 2:

[0338] The server uses the generated pseudo users to run a simulation of the new business service. As input, it uses a list of pseudo users and information about the new business service. As data calculation, it simulates the reactions of the pseudo users and determines whether each pseudo user will accept the service. The output is data of the simulation results.

[0339] Step 3:

[0340] The server analyzes the simulation results and confirms the effectiveness of the new business service. The simulation result data is used as input. The data is processed by statistically analyzing the simulation results and extracting the service acceptance rate and response patterns. The output is the evaluation result of the effectiveness.

[0341] Step 4:

[0342] The server identifies areas for improvement based on the simulation results, such as pricing, delivery methods, and target user selection. It uses the simulation results and effectiveness evaluation results as input. It analyzes the impact of each element as data calculation and extracts areas for improvement for optimization. The output is a list of areas for improvement.

[0343] Step 5:

[0344] The server proposes optimal pricing and delivery methods based on the identified improvements. As input, it uses a list of improvements. As data processing, it generates specific proposals based on the improvements. The output is a proposal for optimal pricing and delivery methods.

[0345] Step 6:

[0346] The terminal displays the simulation results and suggestions sent from the server. It receives data from the server as input. It visually displays the results and suggestions through a user interface. The output is a display that can be confirmed by the user.

[0347] Step 7:

[0348] The user accesses the system through a terminal and checks the simulation results and proposals. The information displayed on the terminal is used as input. The specific action of the user is to make a decision based on the displayed information. The output is the user's decision.

[0349] 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.

[0350] "Example 1"

[0351] In one embodiment of the present invention, the generative AI generates a pseudo-user using an emotion engine that recognizes the user's emotions. The emotion engine understands the user's emotional state and adjusts the pseudo-user's behavioral patterns based on the user's emotional state. For example, if the user expresses joy, the emotion engine causes the pseudo-user to imitate the joyful behavioral patterns. This allows the pseudo-user to more accurately reflect the actual user's emotional state. The server may include means for recognizing the pseudo-user's emotional state using the emotion engine and adjusting the pseudo-user's behavioral patterns based on the recognized emotional state.

[0352] "Example 2"

[0353] In another embodiment of the present invention, the results of the simulation are used to analyze how the new business service affects the user's emotional state, for example, whether the new business service induces a feeling of pleasure or discomfort in the user, and the analysis results are used to identify areas for improvement or enhancement of the new business service.

[0354] "Example 3"

[0355] In one embodiment of the present invention, the generative AI generates a pseudo-user using an emotion engine that recognizes the user's emotions. The emotion engine understands the user's emotional state and adjusts the pseudo-user's behavioral patterns based on that understanding. For example, if the user expresses joy, the emotion engine causes the pseudo-user to mimic the joyful behavioral patterns. This allows the pseudo-user to more accurately reflect the actual user's emotional state.

[0356] The processing flow of each embodiment will be described below.

[0357] "Example 1"

[0358] Step 1: The generative AI activates an emotion engine that recognizes the user's emotions.

[0359] Step 2: The emotion engine understands the user's emotional state.

[0360] Step 3: The emotion engine adjusts the behavioral patterns of the simulated user based on the understood emotional state. For example, if the user shows an emotion of joy, the emotion engine makes the simulated user imitate the behavioral patterns of joy.

[0361] "Example 2"

[0362] Step 1: When the new business service is launched, the simulation begins.

[0363] Step 2: Based on the simulation results, analyze how the new business service affects the user's emotional state.

[0364] Step 3: The analysis results are used to identify areas for improvement or enhancement of the new business service. For example, we analyze whether the new business service evokes feelings of delight or discomfort in users.

[0365] Example 1

[0366] Next, a description will be given of Example 1 of Form 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."

[0367] Conventional methods for evaluating new business services have difficulty accurately imitating the behavior and emotions of actual users, making it difficult to accurately evaluate the effectiveness of the service. Furthermore, there was a lack of means to generate pseudo-users that reflected users' behavioral patterns and preferences, making the simulation results unreliable. Furthermore, because the behavioral simulation of pseudo-users did not take into account emotional states, it was not possible to accurately identify areas for improvement or strengthening of the service.

[0368] 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.

[0369] In this invention, the server includes means for generating a large number of pseudo-users using generative AI, means for running a simulation to verify whether a new business service will be accepted by the pseudo-users, means for verifying the effectiveness of the new business service based on the results of the simulation, means for collecting past user behavior data and user profile data, means for preprocessing the collected data, means for training a generative AI model using the preprocessed data, means for recognizing the emotional state of a user using an emotion engine and adjusting the behavior pattern of the pseudo-user, and means for performing behavioral simulations using the generated pseudo-users. This makes it possible to generate pseudo-users that accurately reflect the behavior patterns and emotional state of users, and to evaluate the effectiveness of the new business service with high accuracy.

[0370] "Generative AI" is AI that has the ability to learn data and generate new data and information based on that data.

[0371] A "pseudo user" is a virtual user created by imitating the behavioral patterns and preferences of an actual user.

[0372] A "start-up service" is a newly developed product or service that needs to be evaluated for effectiveness before being introduced to the market.

[0373] "Simulation" is a process of virtually testing how a new business service will be accepted by pseudo users.

[0374] "Effectiveness" is an indicator that shows the likelihood that a new business service will be successful in the market and the degree to which it will be accepted by users.

[0375] "User behavior data" is a record of a user's past behavior, including purchase history and browsing history.

[0376] "User profile data" refers to user attribute information, including age, gender, region, and interests.

[0377] "Preprocessing" is the process of converting collected data into a form suitable for analysis and model training, and includes cleaning and normalizing the data.

[0378] A "generative artificial intelligence model" is a part of generative artificial intelligence, and is a model for generating new data and information based on training data.

[0379] An "emotion engine" is a technology that recognizes a user's emotional state and adjusts behavioral patterns based on that emotion.

[0380] "Behavioral simulation" is a process in which the behavior of users is reproduced in a virtual environment using the generated pseudo-users, and new business services are evaluated.

[0381] The present invention relates to a system for generating a large number of pseudo-users using generative artificial intelligence and evaluating the effectiveness of new business services. Specific embodiments of this system will be described below.

[0382] The server first collects historical user behavior data and user profile data. This data collection involves running queries from a database or retrieving data from an external API. For example, the server runs an SQL query to retrieve purchase history and browsing history from a database, and calls an external API to collect social media user profile data.

[0383] The server then preprocesses the collected data, which includes data cleaning, normalization, and feature engineering. Specifically, the server uses Python's Pandas library to manipulate data frames, impute missing values, remove outliers, normalize numerical data, and one-hot encode categorical data.

[0384] Using the preprocessed data, the server trains a generative artificial intelligence model. A deep learning framework (e.g., TENSORFLOW (registered trademark), PyTorch) is used for this training. The server trains the neural network using TensorFlow's Model.fit method, inputting the training data to train the model. The training process aims to minimize the loss function and improve accuracy.

[0385] Furthermore, the server recognizes the user's emotional state using an emotion engine (e.g., IBM Watson® Tone Analyzer) that extracts emotions from the user's text data and calculates an emotion score, such as from the user's reviews and comments.

[0386] The server generates a pseudo-user by combining the output of the generative AI model with the emotion score of the emotion engine. The generated pseudo-user mimics the behavioral patterns and preferences of real users and has behavioral patterns that reflect their emotional state.

[0387] Finally, the server performs a behavioral simulation using the generated pseudo-user. This verifies whether the new business service will be accepted by the pseudo-user, and confirms the effectiveness of the new business service based on the results of the simulation. Specifically, it simulates what products the pseudo-user will view and purchase on an online shopping site. For example, the following prompt sentence is input into the generative AI model. The server may include a means for executing a simulation to verify whether the new business service will be accepted by the pseudo-user by inputting a specific prompt sentence into the generative AI model, and a confirmation means for confirming the effectiveness of the new business service by inputting the results of the simulation into the generative AI model.

[0388] Example prompt: "Predict what products a woman in her 30s, living in Tokyo, interested in fashion, would view based on her recent purchase history."

[0389] Based on this prompt, the generative AI model generates a pseudo-user's behavioral pattern and imitates user behavior on an online shopping site.

[0390] The above is a specific embodiment for carrying out the present invention.

[0391] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0392] Step 1:

[0393] The server collects historical user behavior data and user profile data.

[0394] Input: Data from a database or external API

[0395] What it does: The server executes SQL queries to retrieve purchase and browsing history from a database, and calls external APIs to collect social media user profile data.

[0396] Output: Collected user behavior and profile data

[0397] Step 2:

[0398] The server pre-processes the collected data.

[0399] Input: Collected user behavior and profile data

[0400] What it does: The server uses Python's Pandas library to manipulate data frames, impute missing values, remove outliers, normalize numerical data, and one-hot encode categorical data.

[0401] Output: Preprocessed data

[0402] Step 3:

[0403] The server uses the preprocessed data to train a generative AI model.

[0404] Input: Preprocessed data

[0405] Specific operation: The server trains the neural network using TensorFlow's Model.fit method, inputting training data to train the model. During the training process, the goal is to minimize the loss function and improve accuracy.

[0406] Output: A trained generative AI model

[0407] Step 4:

[0408] The server uses an emotion engine to recognize the user's emotional state.

[0409] Input: User text data (reviews, comments, etc.)

[0410] Specific operation: The server uses an emotion engine (e.g., emotion analysis API) to extract emotions from the user's text data and calculate an emotion score.

[0411] Output: Sentiment score

[0412] Step 5:

[0413] The server generates pseudo-users by combining the output of the generative AI model and the emotion engine.

[0414] Input: Trained generative AI model, sentiment score

[0415] Specific operation: Based on the output of the generative AI model, the server creates a pseudo-user profile that reflects the emotion score of the emotion engine.

[0416] Output: Pseudo user profile

[0417] Step 6:

[0418] The server performs a behavioral simulation using the generated pseudo-user.

[0419] Input: Pseudo user profile

[0420] Specific operation: The server simulates what products a pseudo-user would browse and purchase on an online shopping site. Specifically, it applies a product recommendation algorithm based on the pseudo-user's profile and evaluates the recommendation results.

[0421] Output: Simulation results

[0422] The above is the flow of processing of the program of this system.

[0423] (Application example 1)

[0424] Next, a description will be given of Application Example 1 of Embodiment 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."

[0425] Conventional methods for evaluating new business services have the drawback of taking time and cost to obtain feedback from actual users. It is also difficult to accurately identify areas for improvement and strengthening a service, as it is difficult to evaluate the user's emotional state. Furthermore, there is a lack of means to quickly generate product ratings and reviews, leaving users with a lack of reference information when considering a purchase.

[0426] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means. In this invention, the server includes means for generating a large number of pseudo users using a generative AI, means for running a simulation to verify whether a new business service will be accepted by the pseudo users, means for confirming the effectiveness of the new business service based on the results of the simulation, means for recognizing the emotional state of the pseudo users using an emotion engine and adjusting the behavioral patterns of the pseudo users based on the emotional state, means for generating product ratings and reviews based on the behavioral patterns of the pseudo users, and means for storing the generated reviews in a database. This enables new business services to be evaluated quickly and at low cost, and enables accurate evaluations that take into account the emotional state of users and rapid generation of product ratings and reviews.

[0427] "Generative AI" is an artificial intelligence technology that learns user behavior patterns and preferences and generates pseudo-users based on them.

[0428] A "pseudo-user" is a virtual user created by generative AI that mimics the behavioral patterns and emotional states of a real user.

[0429] A "simulation" is a virtual experiment using pseudo-users to verify the extent to which a new business service will be accepted.

[0430] An "emotion engine" is a technology that recognizes the user's emotional state and adjusts the behavioral patterns of the pseudo-user based on that.

[0431] A "behavioral pattern" is a series of actions and reactions that indicate how a user will behave in a particular situation.

[0432] "Preferences" are the likes and preferences that a user has for a particular product or service.

[0433] "Evaluation" refers to the opinions and impressions that pseudo-users give about new business services and products.

[0434] A "review" is a detailed evaluation or opinion of a product or service provided by a simulated user.

[0435] "Database" means an information system for storing and managing generated reviews and ratings.

[0436] The system for implementing this invention uses multiple pieces of hardware and software, including a server, a database server, and an emotion recognition API. Specifically, the server generates a pseudo-user using a generative AI model, recognizes the pseudo-user's emotional state using an emotion engine, and adjusts behavioral patterns based on the recognition.

[0437] Hardware and software used

[0438] Hardware: Servers, database servers

[0439] Software: Python, TensorFlow, emotion recognition API (e.g., Microsoft® Azure® Emotion API), database management system (e.g., MySQL®)

[0440] Program processing

[0441] 1. Data Collection:

[0442] The server retrieves past user behavior data and user profile data from the database, for example, using the following SQL query:

[0443] sql

[0444] SELECT FROM user_behavior_data;

[0445] 2. Data Preprocessing:

[0446] The server preprocesses the acquired data and converts it into a format that the generative AI model can learn from, specifically by normalizing the data and filling in missing values.

[0447] 3. Generative AI model training:

[0448] The server uses TensorFlow to build a generative AI model and trains it using preprocessed data. For example, the model is trained as follows:

[0449] python

[0450] model.fit(preprocessed_data)

[0451] 4. Create a pseudo user:

[0452] The server generates pseudo-users using a trained generative AI model. For example, the server generates pseudo-users as follows:

[0453] python

[0454] pseudo_users = model.generate(num_users=1000)

[0455] 5. Applying the Emotion Engine:

[0456] The server uses an emotion recognition API to recognize the emotional state of the pseudo-user and adjusts behavioral patterns based on that. For example, the server recognizes the emotional state as follows:

[0457] python

[0458] emotion_state = emotion_api.recognize(pseudo_user_data)

[0459] 6. Review Generation:

[0460] The server generates product ratings and reviews based on the adjusted pseudo-user behavior patterns. For example, the server generates reviews as follows:

[0461] python

[0462] reviews = generate_reviews(pseudo_users)

[0463] 7. Save to database:

[0464] The server stores the generated reviews in a database. For example, the reviews are stored as follows:

[0465] sql

[0466] INSERT INTO product_reviews (user_id, product_id, review) VALUES (...)

[0467] Examples and prompts

[0468] As a concrete example, the following prompt sentence is input into the generative AI model:

[0469] User behavior data: { "user_id": 123, "actions": ["view_product", "add_to_cart", "purchase"], "emotions": ["happy", "satisfied"]}

[0470] User profile data: { "age": 30, "gender": "female", "interests": ["fashion", "electronics"]}

[0471] Based on this prompt, the generative AI model generates a pseudo-user and adjusts its behavioral patterns using an emotion engine. Finally, the reviews generated by the pseudo-users are stored in a database for reference by real users.

[0472] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0473] Step 1:

[0474] The server retrieves past user behavior data and user profile data from the database. The input is the user behavior data and user profile data stored in the database. Specifically, it executes an SQL query to retrieve the data. The output is the retrieved user behavior data and user profile data.

[0475] Step 2:

[0476] The server preprocesses the acquired data and converts it into a format that can be used by the generative AI model. The input is the user behavior data and user profile data acquired in step 1. Specific operations include normalizing the data, filling in missing values, and encoding categorical data. The output is the preprocessed data.

[0477] Step 3:

[0478] The server uses TensorFlow to build a generative AI model and trains it using preprocessed data. The input is the data preprocessed in step 2. Specific operations include defining, compiling, and fitting the model. The output is a trained generative AI model.

[0479] Step 4:

[0480] The server generates pseudo-users using the trained generative AI model. The generative AI model trained in step 3 is used as input. Specifically, it inputs a prompt for generation to the model and generates pseudo-user data. The output is the generated pseudo-user data.

[0481] Step 5:

[0482] The server uses an emotion recognition API to recognize the emotional state of the pseudo-user and adjusts the behavioral pattern based on the recognition. The pseudo-user data generated in step 4 is used as input. Specifically, the server sends the pseudo-user data to the emotion recognition API, acquires the emotional state, and adjusts the behavioral pattern based on the acquired emotional state. The output is pseudo-user data with the adjusted emotional state. The server may include means for recognizing the emotional state of the pseudo-user using an emotion engine and adjusting the behavioral pattern of the pseudo-user based on the recognized emotional state.

[0483] Step 6:

[0484] The server generates product ratings and reviews based on the adjusted pseudo-user behavior patterns. The pseudo-user data adjusted in step 5 is used as input. Specifically, the server analyzes the pseudo-user behavior patterns and generates ratings and reviews. The output is the generated ratings and reviews.

[0485] Step 7:

[0486] The server stores the generated reviews in a database. It uses as input the ratings and reviews generated in step 6. It executes an SQL query to insert the reviews into the database. The output is the reviews stored in the database.

[0487] Example 2

[0488] Next, a description will be given of Example 2 of Form 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."

[0489] Before launching a new business service into the market, it is important to verify whether the service will be accepted by users. However, testing with actual users is costly and time-consuming. It is also difficult to accurately grasp users' emotional state and identify areas for improvement in the service. To solve these issues, an efficient and low-cost method is needed to verify the acceptability of new business services and analyze users' emotional state.

[0490] 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.

[0491] In this invention, the server includes means for generating a large number of pseudo users using generative artificial intelligence, means for running simulations to verify whether a new business service will be accepted by the pseudo users, means for simulating the behavior of the pseudo users and collecting the results, means for analyzing the emotional states of the pseudo users based on the results of the simulation, and means for identifying areas for improvement or strengthening the new business service based on the analysis results. This makes it possible to efficiently and low-costly verify the acceptability of the new business service, accurately grasp the emotional states of users, and identify areas for improvement in the service.

[0492] "Generative AI" refers to systems or models that generate data using artificial intelligence technology.

[0493] A "pseudo user" refers to a virtual user that is generated by imitating the behavior and attributes of an actual user.

[0494] "Simulation" refers to a technique for virtually recreating a real-world situation and observing behavior and reactions under specific conditions.

[0495] "Emotional state" refers to the psychological response or emotion a user has to a particular situation or stimulus.

[0496] "Areas for improvement" refers to parts of the current system or service that need to be corrected or changed in order to improve performance or functionality.

[0497] "Enhancements" refer to areas in current systems and services that require additions or improvements to further improve existing functions and performance.

[0498] This invention is a system that uses generative artificial intelligence to generate a large number of pseudo-users, runs simulations to verify whether a new business service will be accepted by the pseudo-users, and confirms the effectiveness of the new business service based on the results.

[0499] The server first generates pseudo-users using generative artificial intelligence (e.g., GPT-4). The pseudo-users have attributes such as age, gender, interests, and purchasing history, and are generated based on data that mimics the behavioral patterns and preferences of actual users. The data of the generated pseudo-users is stored in a database.

[0500] Next, the server provides the new business service to the pseudo-user. If the new business service is an online shopping service, it simulates which products the pseudo-user will select and what purchasing behavior they will take. For example, it records the pseudo-user's actions from adding a specific product to their cart until they complete the purchase.

[0501] The server collects the results of the simulation and stores them in a database, which includes information such as the user's behavioral history, selected products, and purchase completion rate.

[0502] The server then analyzes the simulated user's emotional state based on the simulation results. This analysis uses emotion analysis tools (such as IBM Watson's Natural Language Understanding) to analyze the simulated user's behavioral data and identify emotions such as joy, displeasure, and excitement.

[0503] Finally, the server uses the analysis results to identify areas for improvement and enhancement of the new business service, and provides this information to the service's development team to help improve the service.

[0504] Examples:

[0505] If the new business service is an online shopping service, the following simulation is performed.

[0506] 1. The server uses a generative AI model (e.g., GPT-4) to generate a pseudo-user who is a woman in her 30s.

[0507] 2. The server simulates an online shopping service that provides fashion items to this pseudo-user.

[0508] 3. Record the actions of a simulated user as they add a specific dress to their cart and complete the purchase.

[0509] 4. The server collects behavioral data of the simulated user and uses a sentiment analysis tool (e.g., IBM Watson's Natural Language Understanding) to analyze the emotional state of the simulated user when selecting a dress. The server may include means for collecting behavioral data of the simulated user by simulating the adjusted behavioral pattern of the simulated user. The server may include means for collecting behavioral data of the simulated user by simulating the adjusted behavioral pattern of the simulated user.

[0510] 5. Based on the analysis results, the server identifies areas for improvement in the online shopping service (for example, providing more detailed product descriptions or adding user reviews).

[0511] Example prompt sentence:

[0512] "Generate a pseudo-user, a woman in her 30s, and simulate an online shopping service that offers fashion items. Record which products the pseudo-user selects and what purchasing behavior they exhibit, and perform sentiment analysis based on that behavioral data."

[0513] In this way, the system can verify whether a new business service is acceptable to pseudo-users and identify areas for improvement in the service.

[0514] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0515] Step 1:

[0516] The server generates pseudo-users using generative artificial intelligence (e.g., GPT-4).

[0517] Input: Prompt text "Please generate a pseudo-user for a woman in her 30s"

[0518] Data processing: Generative AI generates pseudo-user attributes (age, gender, interests, purchasing history, etc.) based on the prompt text.

[0519] Output: Generated pseudo user data

[0520] Specific operation: The server inputs a prompt sentence into the generative AI model, obtains the generated pseudo-user data, and stores it in a database.

[0521] Step 2:

[0522] The server provides a new business service to the pseudo-user.

[0523] Input: Data of the generated pseudo-users, details of the new business service (e.g., online shopping service)

[0524] Data processing: The server provides the new business service to the pseudo-user and displays the service interface to the pseudo-user.

[0525] Output: Interface for pseudo-users to use the service

[0526] Specific operation: The server starts an online shopping service that provides fashion items to the pseudo-user.

[0527] Step 3:

[0528] The server simulates the actions of the pseudo-users and collects the results.

[0529] Input: Pseudo user data, interface for new business services

[0530] Data processing: The server simulates the actions of a pseudo-user (selecting a product, adding to cart, completing purchase, etc.) and records the action data.

[0531] Output: Pseudo user behavior data

[0532] Specific operation: The server simulates the actions of a pseudo-user from adding a specific dress to the cart to completing the purchase, and saves the action history in a database.

[0533] Step 4:

[0534] The server analyzes the emotional state of the simulated user based on the results of the simulation.

[0535] Input: Pseudo user behavior data

[0536] Data processing: Using sentiment analysis tools (e.g., IBM Watson's Natural Language Understanding), simulated user behavior data is analyzed to identify emotional states.

[0537] Output: Emotional state data of the simulated user

[0538] Specific operation: The server inputs the behavioral data of the simulated user into an emotion analysis tool to identify emotions such as joy, displeasure, and excitement.

[0539] Step 5:

[0540] The server identifies areas for improvement and enhancement of the new business service based on the analysis results.

[0541] Input: Emotional state data of a simulated user

[0542] Data processing: The server analyzes the emotional state data to identify areas for improvement and enhancement of the service.

[0543] Output: A list of improvements and enhancements

[0544] Specific behavior: Based on the results of the sentiment analysis, the server identifies improvements to the online shopping service, such as providing more detailed product descriptions or adding user reviews, and reports them to the development team.

[0545] (Application example 2)

[0546] Next, a description will be given of Application Example 2 of Form 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."

[0547] It is an important task for companies to verify in advance whether a new business service will be accepted by the market. However, testing with actual users is costly and time-consuming. It is also difficult to accurately grasp the user's emotional state and identify areas for improvement or strengthening of the service. To solve these problems, an efficient and low-cost method is needed to simulate the acceptance of new business services and analyze the user's emotional state.

[0548] 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.

[0549] In this invention, the server includes means for generating a large number of pseudo users using generative AI, means for running a simulation to verify whether a new business service will be accepted by the pseudo users, means for confirming the effectiveness of the new business service based on the results of the simulation, means for analyzing the emotional state of the pseudo users, and means for identifying areas for improvement or enhancement of the new business service based on the analysis results. This makes it possible to efficiently and low-costly simulate whether a new business service will be accepted in the market, accurately grasp the emotional state of users, and identify areas for improvement or enhancement of the service.

[0550] "Generative AI" is a system that uses artificial intelligence technology to generate new data from large amounts of data.

[0551] A "pseudo-user" is a virtual user created to mimic the behavior and preferences of a real user.

[0552] "Simulation" is a method of predicting results under specific conditions by imitating and virtually executing real-world situations.

[0553] "New business services" refer to new products or services that a company offers.

[0554] "Effectiveness" is an indicator of how effective something is for a specific purpose.

[0555] An "emotional state" is the emotional state that a user feels in response to a particular situation or stimulus.

[0556] "Areas for improvement" are parts of the current system or service that need to be corrected or changed to make it better.

[0557] "Areas of improvement" are areas of the current system or service that can be further strengthened to increase effectiveness.

[0558] This invention is a system that uses generative AI to generate a large number of pseudo-users, simulates whether a new business service will be accepted by the pseudo-users, confirms the effectiveness of the new business service based on the results, and further analyzes the emotional state of the pseudo-users. This system is implemented as follows.

[0559] The server first uses generative AI to generate a large number of pseudo-users. These pseudo-users are generated based on data that mimics the behavioral patterns and preferences of actual users. Next, the server provides the new business service to the pseudo-users and simulates their reactions. This simulation makes it possible to predict what actions the pseudo-users will take, which products they will select, and so on.

[0560] Based on the simulation results, the server confirms the effectiveness of the new business service. Furthermore, it collects feedback from the simulated users and analyzes their emotional state using a sentiment analysis tool (e.g., TextBlob). This analysis allows us to understand how the simulated users feel about the new business service.

[0561] As a concrete example, consider the case of simulating a new business service for an online shopping website. The server uses generative AI to generate 100 pseudo-users and provides each pseudo-user with a product list (e.g., "Product A," "Product B," and "Product C"). The pseudo-users randomly select a product and send their selection results and feedback to the server. The server performs sentiment analysis based on the feedback and classifies the pseudo-user's emotional state into "happy," "neutral," "unhappy," etc.

[0562] This system makes it possible to efficiently and inexpensively simulate whether a new business service will be accepted in the market, accurately grasp the emotional state of users, and identify areas for improvement and strengthening of the service.

[0563] An example of a prompt sentence might be:

[0564] "For a new business service, an online shopping site, please simulate which products pseudo-users select and what emotional state they experience. The product list is "Product A," "Product B," and "Product C." Analyze the emotional state of each pseudo-user based on their feedback and output the results."

[0565] In this way, the server can simulate the acceptance of a new business service and identify areas for improvement or enhancement of the service by analyzing the emotional state of the user.

[0566] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0567] Step 1:

[0568] The server uses generative AI to generate a large number of pseudo-users. As input, it uses data to mimic user behavior patterns and preferences. The generated pseudo-users are virtual users that mimic the behavior of real users. As output, it obtains a list of the generated pseudo-users.

[0569] Step 2:

[0570] The server provides the new business service to the pseudo-users and simulates their reactions. As input, it uses the list of generated pseudo-users and information about the new business service. It simulates how the pseudo-users will behave in response to the provided service. As output, it obtains the behavioral data of each pseudo-user.

[0571] Step 3:

[0572] The server checks the effectiveness of the new business service based on the results of the simulation. As input, it uses the behavioral data of the pseudo-users. The server analyzes which products each pseudo-user selected and what purchasing behavior they exhibited. As output, it obtains a report on the effectiveness of the new business service.

[0573] Step 4:

[0574] The server collects the feedback of the pseudo-users and analyzes their emotional states using a sentiment analysis tool. As input, it uses the feedback data of the pseudo-users. The sentiment analysis tool (e.g., TextBlob) analyzes the text of the feedback and classifies the emotional states into "happy," "neutral," "unhappy," etc. As output, it obtains the emotional state data of each pseudo-user.

[0575] Step 5:

[0576] The server identifies areas for improvement and enhancement of the new business service based on the analysis results. As input, it uses emotional state data and behavioral data. The server integrates these data to identify which parts of the new business service bring pleasure to the user and which parts cause discomfort. As output, it obtains a report on areas for improvement and enhancement.

[0577] In this way, the server can simulate the acceptance of a new business service and identify areas for improvement or enhancement of the service by analyzing the emotional state of the user.

[0578] Example 3

[0579] Next, a description will be given of a third embodiment of the third embodiment. 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."

[0580] To assess market acceptance of new business services in advance, it is important to obtain feedback from actual users. However, testing with actual users is time-consuming and costly, so an efficient method is required. Furthermore, unless realistic simulations that take into account the emotional state of users are performed, it is difficult to accurately predict the actual market reaction.

[0581] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[0582] In this invention, the server includes means for generating a large number of pseudo-users using generative artificial intelligence, means for running a simulation to verify whether a new business service will be accepted by the pseudo-users, means for verifying the effectiveness of the new business service based on the results of the simulation, means for recognizing the emotional state of the pseudo-users using an emotion engine and adjusting their behavioral patterns, and means for storing the simulation results in a database and analyzing the collected data, thereby enabling efficient and accurate evaluation of the market acceptance of the new business service.

[0583] "Generative AI" is an AI technology that generates text and data using natural language processing and machine learning.

[0584] A "pseudo-user" is a virtual user created to mimic the behavioral patterns and emotional state of a real user.

[0585] "Simulation" is a process of virtually verifying market acceptance of a new business service using pseudo-users.

[0586] An "emotion engine" is a software technology that recognizes a user's emotional state and adjusts behavioral patterns based on that state.

[0587] A "database" is a system for systematically storing and managing collected simulation results and other data.

[0588] "Data analysis" is the process of analyzing collected data using statistical methods and machine learning to extract useful information and patterns.

[0589] "New business services" are products or services that a company or organization intends to bring to market.

[0590] The present invention is a system for efficiently and accurately evaluating the market acceptance of new business services. A specific embodiment of this system will be described below.

[0591] Hardware and software used

[0592] Hardware: High-performance servers (e.g., high-performance computers)

[0593] Software: Generative AI (e.g., natural language processing models), emotion engines (e.g., emotion recognition software)

[0594] System Overview

[0595] This system uses generative artificial intelligence to generate a large number of pseudo-users and runs simulations to verify whether new business services will be accepted by the pseudo-users. Furthermore, it checks the effectiveness of the new business services based on the results of the simulations and identifies areas for improvement or strengthening.

[0596] Creating a pseudo user

[0597] The server uses generative artificial intelligence to generate a pseudo-user based on the prompt text. The generated pseudo-user mimics the behavioral patterns and emotional states of a real user, enabling a more realistic simulation.

[0598] Using the Emotion Engine

[0599] The server uses the emotion engine to recognize the emotional state of the pseudo-user and adjust its behavioral patterns. For example, if the pseudo-user shows the emotion of "joy," the emotion engine adjusts its behavioral patterns to be positive.

[0600] Running the simulation

[0601] The server uses the generated pseudo-users to simulate the reactions of new business services. The simulation results are stored in a database and later analyzed in detail.

[0602] Data collection and analysis

[0603] The server collects the simulation results and stores them in a database. The collected data is analyzed using data analysis tools to identify the likelihood of new business services being accepted and areas for improvement.

[0604] Specific examples

[0605] Simulation example

[0606] New Business Service: Online Learning Platform

[0607] Pseudo-user reactions: Satisfaction with pricing, feedback on delivery methods, opinions on target user selection

[0608] Example prompts for generative AI models

[0609] We are planning to provide an online learning platform as a new business service. Please simulate the reactions of pseudo-users under the following conditions.

[0610] Pricing: ¥5,000 per month

[0611] Delivery method: Video lectures and live sessions

[0612] Target users: University students and working adults

[0613] This system allows the user to check the effectiveness of a new business service in advance and identify areas that need improvement. The flow of the identification process in the third embodiment will be described with reference to FIG.

[0614] Step 1:

[0615] Creating a pseudo user

[0616] Input: Prompt (e.g., "We are planning to provide an online learning platform as a new business service. Please simulate the reactions of simulated users under the following conditions: Pricing: 5,000 yen per month, Delivery method: video lectures and live sessions, Target users: university students and working adults.")

[0617] Specific operation: The server uses generative AI to generate a pseudo-user based on the prompt. The generative AI analyzes the prompt and generates the pseudo-user's profile and behavioral patterns.

[0618] Output: Pseudo-user data (e.g., a pseudo-user profile of a 30-year-old male who works in an IT company and whose hobby is reading)

[0619] Step 2:

[0620] Recognizing emotional states and adjusting behavioral patterns

[0621] Input: Pseudo user data

[0622] Specific operation: The server uses the emotion engine to recognize the emotional state of the pseudo-user. The emotion engine analyzes the pseudo-user's profile and behavioral patterns to identify the emotional state. For example, if the pseudo-user shows the emotion of "joy," the emotion engine adjusts its behavioral patterns to be more positive.

[0623] Output: Adjusted simulated user behavior patterns

[0624] Step 3:

[0625] Running the simulation

[0626] Input: Adjusted simulated user behavior patterns, details of new business services (e.g., pricing, delivery method)

[0627] Specific operation: The server uses the generated pseudo-user to simulate the reaction to the new business service. In the simulation, the details of the new business service are presented to the pseudo-user and their reaction is recorded.

[0628] Output: Simulation results (e.g., simulated user feedback and behavioral data)

[0629] Step 4:

[0630] Data collection and storage

[0631] Input: Simulation results

[0632] Specific operation: The server collects the simulation results and stores them in a database. The collected data includes feedback and behavioral data of simulated users.

[0633] Output: Saved simulation data

[0634] Step 5:

[0635] Analyzing the data

[0636] Input: Saved simulation data

[0637] Specific operations: The server uses a data analysis tool to statistically analyze the collected data. The data analysis tool analyzes the feedback and behavioral data of the simulated users to identify the acceptability of the new business service and areas for improvement.

[0638] Output: Analysis results (e.g., improvements such as "change pricing to 3,000 yen per month" or "change delivery method from video lectures to live sessions")

[0639] In this way, the server can efficiently and accurately assess market acceptance of new business services.

[0640] (Application example 3)

[0641] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[0642] To assess market acceptance of new business services in advance, it is important to predict actual user reactions and identify areas for improvement. However, conventional methods have had issues with the difficulty of accurately imitating users' emotional states and behavioral patterns, resulting in low reliability of simulation results. Furthermore, there has been a lack of means to effectively analyze collected data and identify specific areas for improvement.

[0643] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[0644] In this invention, the server includes: means for generating a large number of pseudo users using generative AI; means for running a simulation to verify whether a new business service will be accepted by the pseudo users; means for verifying the effectiveness of the new business service based on the results of the simulation; means for understanding the user's emotional state using an emotion engine and adjusting the pseudo user's behavioral patterns based on the understanding; means for simulating the pseudo user's behavioral patterns and collecting behavioral data; means for analyzing the collected behavioral data using a clustering algorithm; and means for outputting the analysis results as a report. This enables highly accurate simulations that reflect the user's emotional state and identifies specific areas for improvement. The server may also include means for analyzing the collected behavioral data using a clustering algorithm.

[0645] "Generative AI" is a system that generates data and information using artificial intelligence technology.

[0646] A "pseudo-user" is a virtual user created to mimic the behavior and emotions of a real user.

[0647] "Simulation" is a method of conducting experiments and verification by imitating real-world situations.

[0648] An "emotion engine" is a technology that analyzes a user's emotional state and adjusts behavioral patterns based on that.

[0649] A "behavioral pattern" is a series of actions that a user tends to take under certain circumstances.

[0650] A "clustering algorithm" is a computational method for grouping data based on similarity.

[0651] A "report" is a document that summarizes the results of analysis and research.

[0652] "New business services" are products or services that are newly introduced to the market.

[0653] "Effectiveness" is an indicator of how effective something is in achieving a specific purpose or goal.

[0654] "Data collection" is the process of gathering the necessary information or data.

[0655] A system for implementing the present invention includes a generative AI, an emotion engine, a clustering algorithm, and a report generation function. A specific embodiment of the system is described below.

[0656] System Configuration

[0657] The server uses generative AI to generate a large number of pseudo-users. The pseudo-users are generated based on data that mimics the user's behavioral patterns and preferences. The emotion engine analyzes the user's emotional state and adjusts the pseudo-user's behavioral patterns accordingly. This allows the pseudo-users to more accurately reflect the emotional state of the actual user.

[0658] simulation

[0659] The server runs a simulation to verify whether the new business service will be accepted by pseudo-users. It simulates the behavioral patterns of the pseudo-users and collects behavioral data. The collected behavioral data is analyzed using a clustering algorithm. Specifically, it uses a clustering algorithm such as KMeans to group the behavioral patterns of the pseudo-users.

[0660] Data Analysis and Report Generation

[0661] The server confirms the effectiveness of the new business service based on the results of the simulation. Furthermore, the server outputs the analysis results as a report. The report includes areas for improvement and strengthening of the new business service, which can be used to improve business strategies. The server may include means for analyzing the collected behavioral data using a clustering algorithm and outputting the results of the analysis as a report.

[0662] Hardware and software used

[0663] Generative AI: Uses machine learning frameworks such as TensorFlow and PyTorch.

[0664] Emotion engine: Uses an emotion analysis engine that uses natural language processing technology.

[0665] Clustering algorithms: Use a data analysis library such as scikit-learn.

[0666] Report generation: Use data visualization libraries such as Pandas and Matplotlib.

[0667] Specific examples

[0668] For example, when selling a new fashion item on an online shopping site, this system is used to simulate the reactions of pseudo-users. The emotion engine analyzes the user's emotional state and generates behavioral patterns for the pseudo-user to "purchase," "view," and "share" the item. From the simulation results, areas for improvement in pricing and promotion methods are identified and output as a report.

[0669] Prompt Sentence Examples

[0670] Simulate the behavioral patterns of simulated users to evaluate pricing and promotion methods for new fashion items. Analyze users' emotional states using an emotion engine to generate simulated users. Identify areas for improvement from the simulation results and generate reports.

[0671] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[0672] Step 1:

[0673] The server uses generative AI to generate a large number of pseudo-users. As input, it uses data to mimic user behavior patterns and preferences. The generative AI generates pseudo-users based on this data and outputs a list of pseudo-users. Specifically, it uses a machine learning framework (such as TensorFlow or PyTorch) to learn user behavior patterns and generate pseudo-users based on them.

[0674] Step 2:

[0675] The server uses an emotion engine to analyze the emotional state of the pseudo-user. It uses a list of pseudo-users as input. The emotion engine analyzes the emotional state of the pseudo-user using natural language processing technology and obtains a list of pseudo-users with assigned emotional states as output. Specifically, the emotion analysis engine extracts the emotional state from text data and behavioral data.

[0676] Step 3:

[0677] The server adjusts the behavioral patterns of the pseudo-users based on their emotional states. As input, it uses a list of pseudo-users with assigned emotional states. The emotion engine adjusts the behavioral patterns of the pseudo-users according to their emotional states, and as output, it obtains a list of pseudo-users with adjusted behavioral patterns. Specifically, it applies an algorithm that changes behavioral patterns based on their emotional states.

[0678] Step 4:

[0679] The server runs a simulation to verify whether the new business service will be accepted by the pseudo users. As input, it uses a list of pseudo users with adjusted behavioral patterns and information about the new business service. The simulation imitates the behavior of the pseudo users and collects behavioral data as output. Specifically, it simulates how the pseudo users will behave toward the new business service.

[0680] Step 5:

[0681] The server analyzes the collected behavioral data using a clustering algorithm. It uses the behavioral data obtained from the simulation as input. It groups the behavioral data using a clustering algorithm (e.g., KMeans) and obtains the clustering results as output. Specifically, it clusters the behavioral data using a data analysis library such as scikit-learn.

[0682] Step 6:

[0683] The server outputs the analysis results as a report. The clustering results are used as input. The report generation function identifies areas for improvement and strengthening of the new business service based on the analysis results, and generates a report as output. Specifically, it uses data visualization libraries such as Pandas and Matplotlib to visually represent the analysis results.

[0684] 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.

[0685] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ) 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.

[0686] Another example of generative AI is Gemini (registered trademark) (Internet search engine). <url: https: gemini.google.com ?hl="ja">) are mentioned.

[0687] 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.

[0688] [Second embodiment]

[0689] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0690] 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.

[0691] 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).

[0692] 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.

[0693] 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.

[0694] 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).

[0695] 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.

[0696] 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.

[0697] 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.

[0698] 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.

[0699] 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.

[0700] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0701] "Example 1"

[0702] One embodiment of the present invention is a method for generating a large number of pseudo-users using a generative AI. Specifically, the generative AI learns data to imitate user behavior patterns and preferences, and generates pseudo-users based on that data. For example, past user behavior data and user profile data can be used as learning data.

[0703] "Example 2"

[0704] Furthermore, there is a means for executing a simulation to verify whether a new business service will be accepted by pseudo-users. Specifically, the new business service is provided to the generated pseudo-users, and their reactions are simulated. For example, if the new business service is an online shopping service, it is possible to simulate which products the pseudo-users will select and what purchasing behavior they will exhibit.

[0705] "Example 3"

[0706] There are also methods to verify the effectiveness of new business services based on the results of simulations. Specifically, the results of the simulations can be used to analyze the acceptability of new business services and areas for improvement. For example, based on the reactions of simulated users, it is possible to identify areas for improvement such as pricing and delivery methods for new business services, and selection of target users.

[0707] The processing flow of each embodiment will be described below.

[0708] "Example 1"

[0709] Step 1: The generative AI learns data to mimic user behavior patterns and preferences. This learning data is based on past user behavior data and user profile data.

[0710] Step 2: Based on the learned data, the generative AI generates pseudo-users, who are then used to verify the acceptability of the new business service.

[0711] "Example 2"

[0712] Step 1: Provide a new business service to the generated pseudo-users. This new business service may be, for example, an online shopping service.

[0713] Step 2: Simulate the reaction of the pseudo-user. Specifically, simulate which products the pseudo-user will select and what purchasing behavior they will exhibit.

[0714] "Example 3"

[0715] Step 1: Analyze the acceptability of the new business service and areas for improvement based on the results of the simulation.

[0716] Step 2: Based on the analysis results, identify areas for improvement such as pricing and delivery methods for the new business service, and selection of target users.

[0717] Example 1

[0718] Next, a description will be given of Example 1 of Form 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."

[0719] Conventional methods for testing new business services involve using actual user data, which poses challenges such as privacy issues and the high cost of data collection. Furthermore, testing based on actual user data takes time, and it is often difficult to obtain prompt feedback. Furthermore, it is difficult to accurately mimic user behavior patterns and preferences, resulting in low simulation accuracy.

[0720] 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.

[0721] In this invention, the server includes: means for generating a large number of pseudo-users using a generative AI model; means for collecting and preprocessing past user behavior data and user profile data; means for training the generative AI model based on the preprocessed data; means for generating pseudo-users by inputting prompt sentences to the trained generative AI model; means for running a simulation to verify whether a new business service will be accepted by the pseudo-users; means for verifying the effectiveness of the new business service based on the results of the simulation; and means for saving the generated pseudo-user data and using it in other systems or applications. This enables rapid and highly accurate verification of new business services while avoiding privacy issues.

[0722] A "generative AI model" is an artificial intelligence model that learns user behavior patterns and preferences and generates pseudo-users based on them.

[0723] A "pseudo-user" is a virtual user created using a generative AI model that mimics the behavioral patterns and preferences of a real user.

[0724] "User behavior data" is data related to past behavior of a user, and includes, for example, purchase history and browsing history.

[0725] "User profile data" refers to data relating to a user's attributes and preferences, including, for example, age, gender, and interests.

[0726] "Preprocessing" refers to the process of converting collected data into a format suitable for a generative AI model, and includes data cleaning, normalization, and feature extraction.

[0727] A "prompt sentence" is text that is input into a generative AI model and contains instructions and conditions for generating a pseudo-user.

[0728] A "simulation" is a virtual experiment using generated pseudo-users to verify whether a new business service will be accepted.

[0729] "New business services" refer to new products or services offered by companies or organizations.

[0730] "Effectiveness" refers to the ability or effectiveness of a new business service to achieve its intended results.

[0731] "Saving" means recording the data of the generated pseudo-user in a database or the like so that it can be used later.

[0732] "Usage" means utilizing the saved pseudo-user data in other systems or applications.

[0733] This invention relates to a system for generating a large number of pseudo-users using a generative AI model and verifying whether a new business service will be accepted by the pseudo-users. A specific embodiment of this system will be described below.

[0734] The server first collects past user behavior data and user profile data from a database. This data includes information such as the actions users have taken and their preferences. The collected data is preprocessed using Python's Pandas library, among other tools. Preprocessing includes data cleaning, normalization, and feature extraction.

[0735] The server then trains a generative AI model based on the preprocessed data. The generative AI model can be, for example, OpenAI's GPT-4. The trained model has the ability to learn user behavior patterns and preferences and generate pseudo-users based on them.

[0736] Once training is complete, the server inputs a prompt to the generative AI model. The prompt provides the generative AI model with instructions and conditions for generating a pseudo-user. For example, the following prompts can be used:

[0737] "User A frequently shops online on weekday evenings. He tends to buy mainly electronic devices and books. His purchase history includes the latest smartphones and best-selling novels. Please generate a pseudo-user that mimics User A's behavioral patterns."

[0738] When this prompt is entered, the generative AI model generates a pseudo-user that mimics the behavioral patterns and preferences of User A. The generated pseudo-user is expected to behave in a similar way to User A.

[0739] Using the generated pseudo-users, the server runs a simulation to verify whether the new business service will be accepted by the pseudo-users. Based on the results of the simulation, the effectiveness of the new business service is confirmed, and areas for improvement or strengthening are identified as necessary.

[0740] Finally, the server stores the generated pseudo-user data in a database and makes it available to other systems and applications, enabling rapid and highly accurate verification of new business services while avoiding privacy issues.

[0741] The above is a specific embodiment for carrying out the present invention.

[0742] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0743] Step 1: Data collection

[0744] The server retrieves past user behavior data and user profile data from the database. As input, it uses database connection information and SQL queries. As output, it obtains the retrieved raw data. Specifically, the server executes the SQL queries to extract user behavior history and profile information.

[0745] Step 2: Data Preprocessing

[0746] The server preprocesses the acquired raw data and converts it into a format suitable for generative AI models. The acquired raw data is used as input, and preprocessed data is obtained as output. Specifically, the server cleans the data (filling in missing values ​​and removing outliers), normalizes it, and extracts features.

[0747] Step 3: Train the model

[0748] The server trains a generative AI model based on the preprocessed data. As input, it uses the preprocessed data and a generative AI model (e.g., GPT-4). As output, it obtains a trained generative AI model. Specifically, the server inputs the preprocessed data into the model and optimizes the model's parameters.

[0749] Step 4: Enter the prompt statement

[0750] The server inputs a prompt sentence into the trained generative AI model. The prompt sentence and the trained generative AI model are used as input. The generated pseudo-user is obtained as output. Specifically, the server inputs the prompt sentence into the model, and the model generates a pseudo-user.

[0751] Step 5: Run the simulation

[0752] The server uses the generated pseudo-users to run a simulation to verify whether the new business service will be accepted. The generated pseudo-users and the scenario of the new business service are used as input. The simulation results are obtained as output. Specifically, the server simulates the behavior of the pseudo-users and evaluates the acceptance of the service.

[0753] Step 6: Saving and using the results

[0754] The server stores the generated pseudo-user data and simulation results, making them available to other systems and applications. The generated pseudo-user data and simulation results are used as input, and the stored data is obtained as output. Specifically, the server stores the data in a database and provides the data to other systems through an API.

[0755] The above are the specific processing steps of the program of this system.

[0756] (Application example 1)

[0757] Next, a description will be given of Application Example 1 of Form 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."

[0758] Conventional methods for testing new business services require the use of actual user data, which is time-consuming and costly, and also poses problems from the perspective of protecting user privacy. It is also difficult to accurately grasp user behavioral patterns and preferences, making it difficult to provide personalized services. Furthermore, product recommendation systems are required to make highly accurate recommendations based on user preferences, but current systems are unable to adequately meet this demand.

[0759] 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.

[0760] In this invention, the server includes means for generating a large number of pseudo users using generative AI, means for running a simulation to verify whether a new business service will be accepted by the pseudo users, means for verifying the effectiveness of the new business service based on the results of the simulation, means for recommending products based on the behavioral data of the pseudo users, and means for providing a personalized shopping experience based on user preferences. This enables efficient and accurate verification of new business services and personalized product recommendations while protecting user privacy.

[0761] "Generative AI" is an artificial intelligence technology that learns user behavior patterns and preferences and generates pseudo-users based on them.

[0762] A "pseudo-user" is a virtual user created by generative AI that mimics the behavioral patterns and preferences of a real user.

[0763] A "simulation" is a virtual experiment using pseudo-users to verify the extent to which a new business service will be accepted.

[0764] "New business services" are new products or services offered by a company, which provide new value to users.

[0765] "Effectiveness" refers to the ability of a new business service to be accepted by users and to achieve expected results.

[0766] "Behavioral Data" is historical information about the actions and choices a user makes on a website or application.

[0767] "Product recommendation" refers to proposing appropriate products to users based on their behavioral data and preferences.

[0768] A "personalized shopping experience" is a shopping experience that provides optimal products and services based on the user's individual preferences and behavioral patterns.

[0769] A system for implementing this invention includes means for generating a large number of pseudo-users using generative AI, means for running a simulation to verify whether a new business service will be accepted by the pseudo-users, means for confirming the effectiveness of the new business service based on the results of the simulation, means for recommending products based on behavioral data of the pseudo-users, and means for providing a personalized shopping experience based on user preferences.

[0770] System Program

[0771] The server generates pseudo-users using a generative AI model. The generative AI model learns past user behavior data and user profile data, and generates pseudo-users based on that data. The generated pseudo-users mimic the behavioral patterns and preferences of actual users.

[0772] Next, the server executes a simulation to verify whether the new business service will be accepted by the pseudo-users. This simulation makes it possible to confirm the effectiveness of the new business service.

[0773] Furthermore, the server recommends products based on the behavioral data of the simulated user, thereby providing a personalized shopping experience based on the user's preferences.

[0774] Hardware and software used

[0775] Hardware: Servers, smartphones

[0776] Software: Python, generative AI models (e.g., GPT-3)

[0777] Data processing and calculation

[0778] 1. Data collection: The server collects past user behavior data and user profile data.

[0779] 2. Data training: The server trains the generative AI model on this data.

[0780] 3. Pseudo-user generation: The server generates pseudo-users using the trained model.

[0781] 4. Simulation execution: The server executes a simulation of the new business service using a pseudo user.

[0782] 5. Product recommendation: The server recommends products based on the behavioral data of the pseudo-user.

[0783] 6. Personalized Offering: The server provides a personalized shopping experience based on the user's preferences.

[0784] Specific examples

[0785] For example, if a user is looking for "smartphone accessories," the server will refer to the pseudo-user's behavioral data and recommend related products (cases, chargers, earphones, etc.).

[0786] Prompt Sentence Examples

[0787] Generate pseudo-users based on user behavior data and recommend products based on the user's preferences. For example, recommend related products based on the data of users who have previously purchased "smartphone accessories."

[0788] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0789] Step 1:

[0790] The server collects historical user behavior and user profile data, including the user's actions and selections on websites and applications. The input is the user's behavior and profile data, and the output is the data stored on the server. Specifically, the server retrieves the required data by querying the database.

[0791] Step 2:

[0792] The server trains the collected data into a generative AI model. The inputs are the user behavior data and profile data collected in step 1, and the output is a trained generative AI model. Specifically, the server preprocesses the data and feeds it to the generative AI model.

[0793] Step 3:

[0794] The server generates a pseudo-user using a trained generative AI model. The input is the trained generative AI model, and the output is the generated pseudo-user data. Specifically, the server inputs prompts into the generative AI model to generate the pseudo-user's profile and behavioral patterns.

[0795] Step 4:

[0796] The server runs a simulation of the new business service using the generated pseudo-user. The inputs are the pseudo-user data and the specifications of the new business service, and the output is the simulation results. Specifically, the server simulates how the pseudo-user uses the new business service and records the results.

[0797] Step 5:

[0798] The server checks the effectiveness of the new business service based on the simulation results. The input is the simulation results, and the output is an evaluation of the effectiveness of the new business service. Specifically, the server analyzes the simulation results and identifies the degree of acceptance of the service and areas for improvement.

[0799] Step 6:

[0800] The server recommends products based on the behavioral data of the pseudo-user. The input is the behavioral data of the pseudo-user, and the output is a list of recommended products. Specifically, the server analyzes the pseudo-user's preferences and behavioral patterns to select the most suitable products.

[0801] Step 7:

[0802] The server provides a personalized shopping experience based on the user's preferences. The inputs are a list of recommended products and the user's preference data, and the output is a personalized shopping experience. Specific operations include displaying products that match the user's preferences and providing an interface that encourages purchases.

[0803] Example 2

[0804] Next, a description will be given of Example 2 of Form 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."

[0805] Before launching a new business service to the market, it is important to verify whether the service will be accepted by users. However, testing with actual users is time-consuming and costly, so an efficient method is required. In addition, it is difficult to accurately imitate user behavior patterns and preferences, which can result in low simulation accuracy. To solve these issues, a simulation system using pseudo-users is needed.

[0806] 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.

[0807] In this invention, the server includes means for generating a large number of pseudo users using generative artificial intelligence, means for running a simulation to verify whether a new business service will be accepted by the pseudo users, means for verifying the effectiveness of the new business service based on the results of the simulation, means for providing the new business service to the generated pseudo users and simulating their reactions, and means for displaying the results of the simulation to users. This makes it possible to efficiently and accurately verify the acceptability of a new business service before it is released to the market.

[0808] "Generative AI" is a system that generates data and information using artificial intelligence technology.

[0809] A "pseudo user" is a virtual user created by imitating the behavioral patterns and preferences of an actual user.

[0810] "New business services" refer to products or services that are planned to be newly introduced to the market.

[0811] A "simulation" is a virtual experiment using pseudo-users to verify the acceptability of a new business service.

[0812] "Simulation results" refers to data and information obtained after running a simulation.

[0813] "User" refers to a person or organization that uses this system to verify new business services.

[0814] "Generating means" refers to a method or device for creating a pseudo-user using generative artificial intelligence.

[0815] The "means for providing" refers to a method or device for presenting the new business service to the pseudo-user.

[0816] The "display means" refers to a method or device for visually showing the simulation results to the user.

[0817] This invention is a system that uses generative artificial intelligence to generate pseudo-users and simulates whether a new business service will be accepted by the pseudo-users. A specific embodiment of this system will be described below.

[0818] The server first generates a pseudo-user using a generative AI model (e.g., a general natural language processing model). Examples of generative AI models used include OpenAI's GPT-4. The generated pseudo-users have attributes such as age, gender, and interests.

[0819] Next, the server provides the new business service to the simulated user. If this service is an online shopping service, the server presents a product list to the simulated user. The simulated user simulates which product to select and what purchasing behavior to take through the generative AI model.

[0820] The terminal provides an interface that allows the user to check the simulation results, and the user can find areas for improvement in the new business service based on the simulation results.

[0821] As a specific example, if the new business service is an "online shopping service," the following prompt sentence would be input into the generative AI model:

[0822] Example prompt sentence:

[0823] Pseudo-user A is a 30-year-old male who is interested in technology products. Please present him with the following product list and simulate which product he would choose.

[0824] 1. Smartphone

[0825] 2. Laptop

[0826] 3. Smartwatch

[0827] 4. Wireless earphones

[0828] The server inputs this prompt into the generative AI model and simulates the purchasing behavior of pseudo-user A. For example, if pseudo-user A selects "smartphone," the reason for the selection and the degree of interest in other products are also output as simulation results.

[0829] Users can check the simulation results on their devices and identify areas for improvement in their new business services. For example, if the adoption rate of smartwatches is low, they can take measures such as revising the product description or pricing.

[0830] This system makes it possible to efficiently and accurately verify the acceptability of new business services before they are brought to market.

[0831] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0832] Step 1:

[0833] The server generates a pseudo-user using a generative AI model. As input, it inputs a prompt statement specifying the pseudo-user's attributes (e.g., age, gender, interests) into the generative AI model. The generative AI model generates a pseudo-user based on this prompt statement and obtains the pseudo-user's data as output. Specifically, the server inputs the following prompt statement into the generative AI model:

[0834] Generate a simulated user who is a 30-year-old male and interested in technology products.

[0835] The generative AI model generates a pseudo-user based on this prompt sentence.

[0836] Step 2:

[0837] The server provides a new business service to the generated pseudo-user. As input, it uses the pseudo-user data and new business service information (e.g., product list). The server presents the product list to the pseudo-user and obtains the pseudo-user's reaction data as output. Specifically, the server presents the following product list to the pseudo-user.

[0838] 1. Smartphone

[0839] 2. Laptop

[0840] 3. Smartwatch

[0841] 4. Wireless earphones

[0842] Step 3:

[0843] The server uses a generative AI model to simulate the pseudo-user's reaction. The pseudo-user's data and product list are used as input. The generative AI model simulates which products the pseudo-user will select and what purchasing behavior they will take, and obtains the simulation results as output. Specifically, the server inputs the following prompt sentence into the generative AI model:

[0844] Pseudo-user A is a 30-year-old male who is interested in technology products. Please present him with the following product list and simulate which product he would choose.

[0845] 1. Smartphone

[0846] 2. Laptop

[0847] 3. Smartwatch

[0848] 4. Wireless earphones

[0849] The generative AI model simulates the purchasing behavior of pseudo-user A based on this prompt sentence.

[0850] Step 4:

[0851] The terminal provides an interface that allows the user to check the simulation results. The simulation results are used as input. The terminal visually displays the simulation results and provides a state that allows the user to check the results as output. Specifically, the terminal displays the simulation results as follows:

[0852] Pseudo-user A chose "Smartphone." Reason for choosing: Because he is interested in the latest technology.

[0853] Based on these results, users can consider ways to improve their product listings and service content.

[0854] (Application example 2)

[0855] Next, a description will be given of Application Example 2 of Form 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."

[0856] It is an important task for companies to verify in advance whether a new business service will be accepted by the market. However, testing using actual users is costly and time-consuming. It is also difficult to accurately grasp user behavior patterns and preferences. This has led to a demand for a method to efficiently confirm the effectiveness of new business services and identify areas for improvement and strengthening.

[0857] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for generating a large number of pseudo users using generative AI, means for running a simulation to verify whether a new business service will be accepted by the pseudo users, means for confirming the effectiveness of the new business service based on the results of the simulation, means for simulating the products selected and purchasing behavior of the pseudo users, and means for analyzing the simulation results and identifying which products were most selected. This makes it possible to efficiently verify the market acceptance of a new business service and identify areas for improvement and enhancement.

[0858] "Generative AI" is a system that generates data and information using artificial intelligence technology.

[0859] A "pseudo user" is a virtual user created by imitating the behavioral patterns and preferences of an actual user.

[0860] "Simulation" is a method of imitating and virtually reproducing real-world events and processes.

[0861] "New business services" refer to new products or services that a company offers.

[0862] "Effectiveness" is an indicator of how effective something is in achieving a specific purpose or goal.

[0863] "Purchasing behavior" refers to the series of actions and decision-making process that consumers take from selecting a product to purchasing it.

[0864] Analysis is the process of examining data or information in detail to find meaning and patterns.

[0865] "Areas for improvement" are areas in the current system or process that need to be changed or modified to make it better.

[0866] "Enhancements" are areas of the current system or process that need further strengthening or expansion.

[0867] A system for implementing this invention includes means for generating a large number of pseudo-users using generative AI, means for running a simulation to verify whether a new business service will be accepted by the pseudo-users, means for confirming the effectiveness of the new business service based on the results of the simulation, means for simulating the products selected and purchasing behavior of the pseudo-users, and means for analyzing the results of the simulation and identifying which products were most selected.

[0868] The server uses generative AI to generate pseudo-users. These pseudo-users mimic the behavioral patterns and preferences of real users, enabling a realistic simulation. The generated pseudo-users are used to simulate reactions to new business services.

[0869] The simulation is run on a server, reproducing which products a simulated user would select and what purchasing behavior they would exhibit. The results of the simulation are collected and analyzed by the server. Based on the analysis results, the effectiveness of the new business service is confirmed and areas for improvement and enhancement are identified.

[0870] The implementation of this system uses the following hardware and software:

[0871] Hardware: Servers, smartphones

[0872] Software: Python, NumPy, generative AI models

[0873] As a concrete example, the following prompt sentence is input into the generative AI model:

[0874] Example prompt sentence:

[0875] To simulate adoption of a new online shopping service, create a simulated user and simulate their choice of the following products:

[0876] Product A: Price 100 yen, quality 80

[0877] Product B: Price 150 yen, quality 90

[0878] Product C: Price 200 yen, quality 85

[0879] The pseudo-user preferences are as follows:

[0880] User 1: Price-sensitive (-0.5), Quality-sensitive (1.0)

[0881] User 2: Price-sensitive (-0.3), Quality-sensitive (0.8)

[0882] Please let me know the simulation results.

[0883] By inputting this prompt into a generative AI model, it is possible to simulate the purchasing behavior of pseudo-users and verify their acceptance of new services.

[0884] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0885] Step 1:

[0886] The server uses generative AI to generate pseudo-users. Data on user behavior patterns and preferences is used as input. Based on this data, the generative AI generates a large number of pseudo-users that mimic the behavior of actual users. The output is a list of the generated pseudo-users.

[0887] Step 2:

[0888] The server runs a simulation to provide a new business service to the generated pseudo-users. As input, it uses a list of pseudo-users and detailed information about the new business service. The simulation reproduces which products the pseudo-users select and what purchasing behavior they exhibit. The output is data on the products selected and purchasing behavior of each pseudo-user.

[0889] Step 3:

[0890] The server collects and analyzes the results of the simulation. It uses the simulation result data as input. It uses data analysis tools to identify which products were most popular and what purchasing behaviors were most common. The output is a report of the analysis results.

[0891] Step 4:

[0892] The server checks the effectiveness of the new business service based on the analysis results. It uses the analysis result report as input. It evaluates whether the new business service will be accepted by the market and identifies areas for improvement or enhancement, if necessary. The output is an evaluation report of the new business service.

[0893] Step 5:

[0894] The server identifies areas for improvement and enhancement of the new business service based on the evaluation report and reflects them in the next simulation. The evaluation report is used as input. A new simulation reflecting the improvements and enhancements is prepared and verified again using simulated users. The output is the simulation results of the improved new business service.

[0895] Example 3

[0896] Next, a description will be given of Example 3 of Form Example 3. 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."

[0897] There is a lack of effective methods to verify whether a new business service will be accepted by target users before it is launched on the market. It is also difficult to identify specific areas for improvement or strengthening based on simulation results. This makes it difficult to obtain appropriate feedback to increase the success rate of the business.

[0898] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[0899] In this invention, the server includes means for generating a large number of pseudo users using generative AI, means for running a simulation to verify whether a new business service will be accepted by the pseudo users, means for collecting simulation result data, means for inputting the collected data into a generative AI model and analyzing the data, means for feeding back the analysis results output from the generative AI model to users, and means for verifying the effectiveness of the new business service based on the simulation results. This makes it possible to effectively verify the acceptability of a new business service before it is released to the market and identify specific areas for improvement and enhancement.

[0900] "Generative AI" is an artificial intelligence technology that generates new data and models based on large amounts of data.

[0901] A "pseudo-user" is a virtual user created to mimic the behavior and preferences of a real user.

[0902] "Simulation" is a technology that imitates real-world systems and processes and virtually reproduces their behavior.

[0903] "Simulation result data" refers to various data and information obtained when a simulation is executed.

[0904] A "generative AI model" is a model that uses generative AI to perform specific tasks or generate data.

[0905] "Data analysis" is the process of extracting useful information and insights from collected data.

[0906] "Feedback" refers to returning the results or evaluation of a system or process to the user.

[0907] "New business services" are services or products that are planned to be newly introduced to the market.

[0908] "Effectiveness" is an indicator of how effective something is in achieving a specific purpose or goal.

[0909] "Areas for improvement" are parts or elements of the current system or process that need improvement.

[0910] "Improvement points" are parts or elements of the current system or process that need to be further strengthened.

[0911] The present invention provides a system for effectively verifying the acceptability of a new business service before it is launched into the market and identifying specific areas for improvement or strengthening. Specific embodiments of this system will be described below.

[0912] Hardware and software used

[0913] Hardware: Servers, terminals

[0914] Software: Generative AI models (e.g., GPT-4), simulation software (e.g., AnyLogic)

[0915] System configuration

[0916] 1. The server uses generative AI to generate a large number of pseudo-users. Pseudo-users are virtual users created to mimic the behavior and preferences of real users.

[0917] 2. The server runs a simulation to verify whether the new business service will be accepted by the pseudo-users. Using simulation software (e.g., AnyLogic), the behavioral patterns and reactions of the pseudo-users are modeled.

[0918] 3. The server collects the simulation results data, which are stored in a database and used for subsequent analysis.

[0919] 4. The server inputs the collected data into a generative AI model (e.g., GPT-4) and analyzes the data. The generative AI model uses the simulation results to identify areas for improvement, such as pricing for the new business service, delivery method, and selection of target users.

[0920] 5. The server sends the analysis results output from the generative AI model to the terminal.

[0921] 6. The terminal displays the analysis results to the user, who can then check areas for improvement in the new business service based on the displayed analysis results and make any necessary corrections.

[0922] Specific examples

[0923] For example, consider the case of launching an "online learning platform" as a new business service.

[0924] 1. The server launches AnyLogic and selects the simulation model of the "Online Learning Platform."

[0925] 2. The server inputs the pricing as "1,000 yen per month," the delivery method as "mobile app," and the target users as "working adults in their 20s and 30s."

[0926] 3. The server sets scenarios such as "study on weekday evenings" and "study all at once on weekends" as the behavioral patterns of the pseudo-users.

[0927] 4. The server executes the configured simulation model and simulates how a pseudo-user uses the platform.

[0928] 5. The server records the user's response to each scenario (e.g., study time, frequency of use, satisfaction, etc.).

[0929] 6. The server collects the simulation result data (e.g., study time, frequency of use, satisfaction level, etc.) and stores it in a database.

[0930] 7. The server converts the collected data into CSV format and formats it so that it can be input into the generative AI model.

[0931] 8. The server inputs the formatted data into GPT-4.

[0932] 9. From the simulation results, GPT-4 identifies areas for improvement, such as "the optimal pricing would be 1,000 yen per month," "the preferred delivery method is a mobile app," and "the optimal target users are working adults in their 20s and 30s."

[0933] 10. GPT-4 outputs the analysis results in text format.

[0934] 11. The server sends the analysis results output by GPT-4 to the terminal.

[0935] 12. The device displays analysis results to the user, such as "The optimal price setting is 1,000 yen per month," "The preferred delivery method is a mobile app," and "The optimal target users are working adults in their 20s and 30s."

[0936] 13. The user adjusts the pricing and delivery methods of the online learning platform based on the displayed analysis results.

[0937] Prompt Sentence Examples

[0938] "If you're launching an online learning platform as a new business service, use the simulation results to identify areas for improvement in pricing, delivery, and target users."

[0939] In this way, the system can confirm the effectiveness of the new business service based on the simulation results and identify specific areas for improvement and strengthening. The flow of the identification process in the third embodiment will be described with reference to FIG.

[0940] Step 1:

[0941] The server uses generative AI to generate a large number of pseudo-users.

[0942] Input: Data to mimic user behavior patterns and preferences

[0943] Data processing: Generative AI generates pseudo-user profiles based on input data.

[0944] Output: Pseudo user profile data

[0945] Specific operation: The server starts the generative AI model and inputs the user's behavioral patterns and preference data. The generative AI model generates a pseudo-user profile based on this data and stores it in a database.

[0946] Step 2:

[0947] The server executes a simulation to verify whether the new business service will be accepted by the pseudo-users.

[0948] Input: Pseudo user profile data, parameters of new business services (e.g., pricing, delivery method, target user attributes)

[0949] Data calculation: Simulation software evaluates the likelihood of new business services being accepted based on the behavioral patterns of simulated users.

[0950] Output: Simulation result data (e.g., study time, frequency of use, satisfaction, etc.)

[0951] Specific operation: The server starts the simulation software and inputs the profile data of the pseudo-user and the parameters of the new business service. The simulation software runs the simulation based on this data and generates the result data.

[0952] Step 3:

[0953] The server collects the result data of the simulation.

[0954] Input: Simulation result data

[0955] Data processing: The resulting data is stored in a database and formatted so that it can be input into the generative AI model.

[0956] Output: Formatted data

[0957] Specific operation: The server retrieves the result data from the simulation software, stores it in a database, and then converts it into CSV format, which is a format that can be input into the generative AI model.

[0958] Step 4:

[0959] The server inputs the collected data into a generative AI model and analyzes the data.

[0960] Input: Formatted data

[0961] Data calculation: Generative AI models use simulation results to identify areas for improvement in new business services, such as pricing, delivery methods, and target user selection.

[0962] Output: Analysis results (e.g., areas for improvement and enhancement)

[0963] How it works: The server inputs the formatted data into the generative AI model, which analyzes the data and identifies areas for improvement and strengthening. The analysis results are output in text format.

[0964] Step 5:

[0965] The server transmits the analysis results output from the generative AI model to the terminal.

[0966] Input: Analysis results

[0967] Data processing: Converting the analysis results into a format that is easy for users to understand.

[0968] Output: Feedback data

[0969] Specific operation: The server receives the analysis results output by the generative AI model, converts them into a format that is easy for the user to understand, and then sends the feedback data to the device.

[0970] Step 6:

[0971] The terminal displays the analysis results to the user.

[0972] Input: Feedback data

[0973] Data processing: Converting feedback data into a display format.

[0974] Output: Displayed analysis results

[0975] Specific operation: The terminal receives the feedback data from the server and displays it to the user. The user can then check the areas for improvement in the new business service based on the displayed analysis results and make any necessary corrections.

[0976] (Application example 3)

[0977] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[0978] When developing a new business service, it is extremely important to evaluate in advance whether the service will be accepted by the market. However, it is difficult to predict actual market reactions, and there is a lack of concrete methods for optimizing elements such as pricing, delivery methods, and target user selection. For this reason, there is a need for a system that can perform simulations in advance, confirm the effectiveness of new business services based on the reactions of simulated users, and identify areas for improvement.

[0979] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means. In this invention, the server includes means for generating a large number of pseudo users using generative AI, means for running a simulation to verify whether a new business service will be accepted by the pseudo users, means for confirming the effectiveness of the new business service based on the results of the simulation, means for identifying areas for improvement such as pricing, provision method, and selection of target users based on the results of the simulation, and means for proposing optimal pricing and provision method. This makes it possible to predict and optimize market acceptance of the new business service.

[0980] "Generative AI" is a system that uses artificial intelligence technology to generate new data from large amounts of data.

[0981] A "pseudo user" is a virtual user created by imitating the behavior and preferences of an actual user.

[0982] "Simulation" is a method of virtually recreating a real-world situation and analyzing the results.

[0983] "New business services" refer to products and services that are newly introduced to the market.

[0984] "Effectiveness" is an indicator of how effective something is for a specific purpose.

[0985] "Pricing" is the process of determining the selling price of a product or service.

[0986] "Delivery method" refers to the means or format by which a product or service is delivered to a user.

[0987] "Target users" are the user group that is expected to use a particular product or service.

[0988] "Improvements" are specific changes that will resolve current problems and improve the situation.

[0989] A "proposal" is the act of suggesting a solution or improvement to a specific problem.

[0990] The system for implementing this invention consists of three main elements: a server, a terminal, and a user. The following explains how each element functions.

[0991] server

[0992] The server uses generative AI to generate a large number of pseudo-users. These pseudo-users are generated based on data that mimics the behavioral patterns and preferences of actual users. These pseudo-users are used in simulations to verify whether new business services will be accepted.

[0993] The results of the simulation are analyzed on the server and used to confirm the effectiveness of the new business service. Specifically, the simulation results are used to identify areas for improvement such as pricing, delivery methods, and target user selection, and optimal pricing and delivery methods are proposed.

[0994] Terminal

[0995] The terminal provides an interface for users to access the system. Mobile devices such as smartphones and tablets are mainly used. The application on the terminal displays the simulation results and proposals sent from the server and helps users make decisions based on them.

[0996] User

[0997] Users can access the system via their devices and check the simulation results and proposals for new business services. Based on this information, users can identify areas for improvement in services and formulate optimal strategies before actually launching them on the market.

[0998] Hardware and software used

[0999] Hardware: Servers (with high-performance computing power), smartphones, tablets

[1000] Software: Python, Pandas (data processing library), NumPy (numerical calculation library), Scikit-learn (machine learning library)

[1001] Specific examples

[1002] For example, consider the case of simulating pricing for a new product, a "smartwatch," and identifying the optimal price range. The server uses generative AI to generate pseudo-users and run the simulation. Based on the simulation results, it identifies areas for improvement in pricing and delivery methods and proposes the optimal price range.

[1003] Prompt Sentence Examples

[1004] "Please simulate pricing for a new product, a smartwatch, and identify the optimal price point. Also, please analyze the acceptability of a subscription model versus a one-time purchase model."

[1005] In this way, a system is realized in which servers, terminals, and users work together to predict and optimize market acceptance of new business services.

[1006] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[1007] Step 1:

[1008] The server uses generative AI to generate a large number of pseudo-users. As input, it uses data to imitate user behavior patterns and preferences. As data processing, it creates pseudo-user profiles based on this data. The output is a list of the generated pseudo-users.

[1009] Step 2:

[1010] The server uses the generated pseudo users to run a simulation of the new business service. As input, it uses a list of pseudo users and information about the new business service. As data calculation, it simulates the reactions of the pseudo users and determines whether each pseudo user will accept the service. The output is data of the simulation results.

[1011] Step 3:

[1012] The server analyzes the simulation results and confirms the effectiveness of the new business service. The simulation result data is used as input. The data is processed by statistically analyzing the simulation results and extracting the service acceptance rate and response patterns. The output is the evaluation result of the effectiveness.

[1013] Step 4:

[1014] The server identifies areas for improvement based on the simulation results, such as pricing, delivery methods, and target user selection. It uses the simulation results and effectiveness evaluation results as input. It analyzes the impact of each element as data calculation and extracts areas for improvement for optimization. The output is a list of areas for improvement.

[1015] Step 5:

[1016] The server proposes optimal pricing and delivery methods based on the identified improvements. As input, it uses a list of improvements. As data processing, it generates specific proposals based on the improvements. The output is a proposal for optimal pricing and delivery methods.

[1017] Step 6:

[1018] The terminal displays the simulation results and suggestions sent from the server. It receives data from the server as input. It visually displays the results and suggestions through a user interface. The output is a display that can be confirmed by the user.

[1019] Step 7:

[1020] The user accesses the system through a terminal and checks the simulation results and proposals. The information displayed on the terminal is used as input. The specific action of the user is to make a decision based on the displayed information. The output is the user's decision.

[1021] 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.

[1022] "Example 1"

[1023] In one embodiment of the present invention, the generative AI generates a pseudo-user using an emotion engine that recognizes the user's emotions. The emotion engine understands the user's emotional state and adjusts the pseudo-user's behavioral patterns based on that understanding. For example, if the user expresses joy, the emotion engine causes the pseudo-user to mimic the joyful behavioral patterns. This allows the pseudo-user to more accurately reflect the actual user's emotional state.

[1024] "Example 2"

[1025] In another embodiment of the present invention, the results of the simulation are used to analyze how the new business service affects the user's emotional state, for example, whether the new business service induces a feeling of pleasure or discomfort in the user, and the analysis results are used to identify areas for improvement or enhancement of the new business service.

[1026] "Example 3"

[1027] In one embodiment of the present invention, the generative AI generates a pseudo-user using an emotion engine that recognizes the user's emotions. The emotion engine understands the user's emotional state and adjusts the pseudo-user's behavioral patterns based on that understanding. For example, if the user expresses joy, the emotion engine causes the pseudo-user to mimic the joyful behavioral patterns. This allows the pseudo-user to more accurately reflect the actual user's emotional state.

[1028] The processing flow of each embodiment will be described below.

[1029] "Example 1"

[1030] Step 1: The generative AI activates an emotion engine that recognizes the user's emotions.

[1031] Step 2: The emotion engine understands the user's emotional state.

[1032] Step 3: The emotion engine adjusts the behavioral patterns of the simulated user based on the understood emotional state. For example, if the user shows an emotion of joy, the emotion engine makes the simulated user imitate the behavioral patterns of joy.

[1033] "Example 2"

[1034] Step 1: When the new business service is launched, the simulation begins.

[1035] Step 2: Based on the simulation results, analyze how the new business service affects the user's emotional state.

[1036] Step 3: The analysis results are used to identify areas for improvement or enhancement of the new business service. For example, we analyze whether the new business service evokes feelings of delight or discomfort in users.

[1037] Example 1

[1038] Next, a description will be given of Example 1 of Form 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."

[1039] Conventional methods for evaluating new business services have difficulty accurately imitating the behavior and emotions of actual users, making it difficult to accurately evaluate the effectiveness of the service. Furthermore, there was a lack of means to generate pseudo-users that reflected users' behavioral patterns and preferences, making the simulation results unreliable. Furthermore, because the behavioral simulation of pseudo-users did not take into account emotional states, it was not possible to accurately identify areas for improvement or strengthening of the service.

[1040] 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.

[1041] In this invention, the server includes means for generating a large number of pseudo-users using generative AI, means for running a simulation to verify whether a new business service will be accepted by the pseudo-users, means for verifying the effectiveness of the new business service based on the results of the simulation, means for collecting past user behavior data and user profile data, means for preprocessing the collected data, means for training a generative AI model using the preprocessed data, means for recognizing the emotional state of a user using an emotion engine and adjusting the behavior pattern of the pseudo-user, and means for performing behavioral simulations using the generated pseudo-users. This makes it possible to generate pseudo-users that accurately reflect the behavior patterns and emotional state of users, and to evaluate the effectiveness of the new business service with high accuracy.

[1042] "Generative AI" is AI that has the ability to learn data and generate new data and information based on that data.

[1043] A "pseudo user" is a virtual user created by imitating the behavioral patterns and preferences of an actual user.

[1044] A "start-up service" is a newly developed product or service that needs to be evaluated for effectiveness before being introduced to the market.

[1045] "Simulation" is a process of virtually testing how a new business service will be accepted by pseudo users.

[1046] "Effectiveness" is an indicator that shows the likelihood that a new business service will be successful in the market and the degree to which it will be accepted by users.

[1047] "User behavior data" is a record of a user's past behavior, including purchase history and browsing history.

[1048] "User profile data" refers to user attribute information, including age, gender, region, and interests.

[1049] "Preprocessing" is the process of converting collected data into a form suitable for analysis and model training, and includes cleaning and normalizing the data.

[1050] A "generative artificial intelligence model" is a part of generative artificial intelligence, and is a model for generating new data and information based on training data.

[1051] An "emotion engine" is a technology that recognizes a user's emotional state and adjusts behavioral patterns based on that emotion.

[1052] "Behavioral simulation" is a process in which the behavior of users is reproduced in a virtual environment using the generated pseudo-users, and new business services are evaluated.

[1053] The present invention relates to a system for generating a large number of pseudo-users using generative artificial intelligence and evaluating the effectiveness of new business services. Specific embodiments of this system will be described below.

[1054] The server first collects historical user behavior data and user profile data. This data collection involves running queries from a database or retrieving data from an external API. For example, the server runs an SQL query to retrieve purchase history and browsing history from a database, and calls an external API to collect social media user profile data.

[1055] The server then preprocesses the collected data, which includes data cleaning, normalization, and feature engineering. Specifically, the server uses Python's Pandas library to manipulate data frames, impute missing values, remove outliers, normalize numerical data, and one-hot encode categorical data.

[1056] Using the preprocessed data, the server trains a generative artificial intelligence model. This training uses a deep learning framework (e.g., TensorFlow, PyTorch). The server trains the neural network using TensorFlow's Model.fit method, inputting the training data to train the model. The training process aims to minimize the loss function and improve accuracy.

[1057] Furthermore, the server recognizes the user's emotional state using an emotion engine (e.g., IBM Watson Tone Analyzer) that extracts emotions from the user's text data and calculates an emotion score, e.g., from the user's reviews and comments.

[1058] The server generates a pseudo-user by combining the output of the generative AI model with the emotion score of the emotion engine. The generated pseudo-user mimics the behavioral patterns and preferences of real users and has behavioral patterns that reflect their emotional state.

[1059] Finally, the server performs a behavioral simulation using the generated pseudo-user. This verifies whether the new business service will be accepted by the pseudo-user and confirms its effectiveness based on the results of the simulation. Specifically, it simulates what products the pseudo-user will view and purchase on an online shopping site. For example, the following prompt sentence is input into the generative AI model:

[1060] Example prompt: "Predict what products a woman in her 30s, living in Tokyo, interested in fashion, would view based on her recent purchase history."

[1061] Based on this prompt, the generative AI model generates a pseudo-user's behavioral pattern and imitates user behavior on an online shopping site.

[1062] The above is a specific embodiment for carrying out the present invention.

[1063] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1064] Step 1:

[1065] The server collects historical user behavior data and user profile data.

[1066] Input: Data from a database or external API

[1067] What it does: The server executes SQL queries to retrieve purchase and browsing history from a database, and calls external APIs to collect social media user profile data.

[1068] Output: Collected user behavior and profile data

[1069] Step 2:

[1070] The server pre-processes the collected data.

[1071] Input: Collected user behavior and profile data

[1072] What it does: The server uses Python's Pandas library to manipulate data frames, impute missing values, remove outliers, normalize numerical data, and one-hot encode categorical data.

[1073] Output: Preprocessed data

[1074] Step 3:

[1075] The server uses the preprocessed data to train a generative AI model.

[1076] Input: Preprocessed data

[1077] Specific operation: The server trains the neural network using TensorFlow's Model.fit method, inputting training data to train the model. During the training process, the goal is to minimize the loss function and improve accuracy.

[1078] Output: A trained generative AI model

[1079] Step 4:

[1080] The server uses an emotion engine to recognize the user's emotional state.

[1081] Input: User text data (reviews, comments, etc.)

[1082] Specific operation: The server uses an emotion engine (e.g., emotion analysis API) to extract emotions from the user's text data and calculate an emotion score.

[1083] Output: Sentiment score

[1084] Step 5:

[1085] The server generates pseudo-users by combining the output of the generative AI model and the emotion engine.

[1086] Input: Trained generative AI model, sentiment score

[1087] Specific operation: Based on the output of the generative AI model, the server creates a pseudo-user profile that reflects the emotion score of the emotion engine.

[1088] Output: Pseudo user profile

[1089] Step 6:

[1090] The server performs a behavioral simulation using the generated pseudo-user.

[1091] Input: Pseudo user profile

[1092] Specific operation: The server simulates what products a pseudo-user would browse and purchase on an online shopping site. Specifically, it applies a product recommendation algorithm based on the pseudo-user's profile and evaluates the recommendation results.

[1093] Output: Simulation results

[1094] The above is the flow of processing of the program of this system.

[1095] (Application example 1)

[1096] Next, a description will be given of Application Example 1 of Form 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."

[1097] Conventional methods for evaluating new business services have the drawback of taking time and cost to obtain feedback from actual users. It is also difficult to accurately identify areas for improvement and strengthening a service, as it is difficult to evaluate the user's emotional state. Furthermore, there is a lack of means to quickly generate product ratings and reviews, leaving users with a lack of reference information when considering a purchase.

[1098] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means. In this invention, the server includes: means for generating a large number of pseudo users using a generative AI; means for running a simulation to verify whether a new business service will be accepted by the pseudo users; means for verifying the effectiveness of the new business service based on the results of the simulation; means for recognizing the emotional state of the pseudo users using an emotion engine and adjusting the behavioral patterns of the pseudo users based on the emotional state; means for generating product ratings and reviews based on the behavioral patterns of the pseudo users; and means for storing the generated reviews in a database. This enables evaluation of new business services quickly and at low cost, enabling accurate evaluations that take into account the emotional state of users, and rapid generation of product ratings and reviews. The server may also include means for generating product ratings and product reviews based on the behavioral patterns of the pseudo users, and means for storing the generated reviews in a database.

[1099] "Generative AI" is an artificial intelligence technology that learns user behavior patterns and preferences and generates pseudo-users based on them.

[1100] A "pseudo-user" is a virtual user created by generative AI that mimics the behavioral patterns and emotional states of a real user.

[1101] A "simulation" is a virtual experiment using pseudo-users to verify the extent to which a new business service will be accepted.

[1102] An "emotion engine" is a technology that recognizes the user's emotional state and adjusts the behavioral patterns of the pseudo-user based on that.

[1103] A "behavioral pattern" is a series of actions and reactions that indicate how a user will behave in a particular situation.

[1104] "Preferences" are the likes and preferences that a user has for a particular product or service.

[1105] "Evaluation" refers to the opinions and impressions that pseudo-users give about new business services and products.

[1106] A "review" is a detailed evaluation or opinion of a product or service by a pseudo-user. The server may include a means for generating product evaluations and reviews based on the behavioral patterns of the pseudo-user, and a means for storing the generated reviews in a database.

[1107] "Database" means an information system for storing and managing generated reviews and ratings.

[1108] The system for implementing this invention uses multiple pieces of hardware and software, including a server, a database server, and an emotion recognition API. Specifically, the server generates a pseudo-user using a generative AI model, recognizes the pseudo-user's emotional state using an emotion engine, and adjusts behavioral patterns based on the recognition.

[1109] Hardware and software used

[1110] Hardware: Servers, database servers

[1111] Software: Python, TensorFlow, emotion recognition API (e.g., Microsoft Azure Emotion API), database management system (e.g., MySQL)

[1112] Program processing

[1113] 1. Data Collection:

[1114] The server retrieves past user behavior data and user profile data from the database, for example, using the following SQL query:

[1115] sql

[1116] SELECT FROM user_behavior_data;

[1117] 2. Data Preprocessing:

[1118] The server preprocesses the acquired data and converts it into a format that the generative AI model can learn from, specifically by normalizing the data and filling in missing values.

[1119] 3. Generative AI model training:

[1120] The server uses TensorFlow to build a generative AI model and trains it using preprocessed data. For example, the model is trained as follows:

[1121] python

[1122] model.fit(preprocessed_data)

[1123] 4. Create a pseudo user:

[1124] The server generates pseudo-users using a trained generative AI model. For example, the server generates pseudo-users as follows:

[1125] python

[1126] pseudo_users = model.generate(num_users=1000)

[1127] 5. Applying the Emotion Engine:

[1128] The server uses an emotion recognition API to recognize the emotional state of the pseudo-user and adjusts behavioral patterns based on that. For example, the server recognizes the emotional state as follows:

[1129] python

[1130] emotion_state = emotion_api.recognize(pseudo_user_data)

[1131] 6. Review Generation:

[1132] The server generates product ratings and reviews based on the adjusted pseudo-user behavior patterns. For example, the server generates reviews as follows:

[1133] python

[1134] reviews = generate_reviews(pseudo_users)

[1135] 7. Save to database:

[1136] The server stores the generated reviews in a database. For example, the reviews are stored as follows:

[1137] sql

[1138] INSERT INTO product_reviews (user_id, product_id, review) VALUES (...)

[1139] Examples and prompts

[1140] As a concrete example, the following prompt sentence is input into the generative AI model:

[1141] User behavior data: { "user_id": 123, "actions": ["view_product", "add_to_cart", "purchase"], "emotions": ["happy", "satisfied"]}

[1142] User profile data: { "age": 30, "gender": "female", "interests": ["fashion", "electronics"]}

[1143] Based on this prompt, the generative AI model generates a pseudo-user and adjusts its behavioral patterns using an emotion engine. Finally, the reviews generated by the pseudo-users are stored in a database for reference by real users.

[1144] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1145] Step 1:

[1146] The server retrieves past user behavior data and user profile data from the database. The input is the user behavior data and user profile data stored in the database. Specifically, it executes an SQL query to retrieve the data. The output is the retrieved user behavior data and user profile data.

[1147] Step 2:

[1148] The server preprocesses the acquired data and converts it into a format that can be used by the generative AI model. The input is the user behavior data and user profile data acquired in step 1. Specific operations include normalizing the data, filling in missing values, and encoding categorical data. The output is the preprocessed data.

[1149] Step 3:

[1150] The server uses TensorFlow to build a generative AI model and trains it using preprocessed data. The input is the data preprocessed in step 2. Specific operations include defining, compiling, and fitting the model. The output is a trained generative AI model.

[1151] Step 4:

[1152] The server generates pseudo-users using the trained generative AI model. The generative AI model trained in step 3 is used as input. Specifically, it inputs a prompt for generation to the model and generates pseudo-user data. The output is the generated pseudo-user data.

[1153] Step 5:

[1154] The server uses the emotion recognition API to recognize the emotional state of the pseudo-user and adjusts the behavioral patterns based on that. The pseudo-user data generated in step 4 is used as input. Specifically, the server sends the pseudo-user data to the emotion recognition API, acquires the emotional state, and adjusts the behavioral patterns based on that. The output is pseudo-user data with the adjusted emotional state.

[1155] Step 6:

[1156] The server generates product ratings and reviews based on the adjusted pseudo-user behavior patterns. The pseudo-user data adjusted in step 5 is used as input. Specifically, the server analyzes the pseudo-user behavior patterns and generates ratings and reviews. The output is the generated ratings and reviews.

[1157] Step 7:

[1158] The server stores the generated reviews in a database. It uses as input the ratings and reviews generated in step 6. It executes an SQL query to insert the reviews into the database. The output is the reviews stored in the database.

[1159] Example 2

[1160] Next, a description will be given of Example 2 of Form 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."

[1161] Before launching a new business service into the market, it is important to verify whether the service will be accepted by users. However, testing with actual users is costly and time-consuming. It is also difficult to accurately grasp users' emotional state and identify areas for improvement in the service. To solve these issues, an efficient and low-cost method is needed to verify the acceptability of new business services and analyze users' emotional state.

[1162] 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.

[1163] In this invention, the server includes: a means for generating a large number of pseudo users using generative AI; a means for running a simulation to verify whether a new business service will be accepted by the pseudo users; a means for simulating the behavior of the pseudo users and collecting the results; a means for analyzing the emotional state of the pseudo users based on the simulation results; and a means for identifying areas for improvement or enhancement of the new business service based on the analysis results. This makes it possible to efficiently and inexpensively verify the acceptability of the new business service, accurately grasp the emotional state of users, and identify areas for improvement of the service. The server may further include a means for identifying areas for improvement or enhancement of the new business service by inputting a specific prompt sentence into the generative AI model.

[1164] "Generative AI" refers to systems or models that generate data using artificial intelligence technology.

[1165] A "pseudo user" refers to a virtual user that is generated by imitating the behavior and attributes of an actual user.

[1166] "Simulation" refers to a technique for virtually recreating a real-world situation and observing behavior and reactions under specific conditions.

[1167] "Emotional state" refers to the psychological response or emotion a user has to a particular situation or stimulus.

[1168] "Areas for improvement" refers to parts of the current system or service that need to be corrected or changed in order to improve performance or functionality.

[1169] "Enhancements" refer to parts of current systems or services that need to be added or improved to further improve existing functions or performance. The server may further include means for identifying improvements and enhancements to the new business service by inputting specific prompt sentences into the generative AI model.

[1170] This invention is a system that uses generative artificial intelligence to generate a large number of pseudo-users, runs simulations to verify whether a new business service will be accepted by the pseudo-users, and confirms the effectiveness of the new business service based on the results.

[1171] The server first generates pseudo-users using generative artificial intelligence (e.g., GPT-4). The pseudo-users have attributes such as age, gender, interests, and purchasing history, and are generated based on data that mimics the behavioral patterns and preferences of actual users. The data of the generated pseudo-users is stored in a database.

[1172] Next, the server provides the new business service to the pseudo-user. If the new business service is an online shopping service, it simulates which products the pseudo-user will select and what purchasing behavior they will take. For example, it records the pseudo-user's actions from adding a specific product to their cart until they complete the purchase.

[1173] The server collects the results of the simulation and stores them in a database, which includes information such as the user's behavioral history, selected products, and purchase completion rate.

[1174] The server then analyzes the simulated user's emotional state based on the simulation results. This analysis uses emotion analysis tools (such as IBM Watson's Natural Language Understanding) to analyze the simulated user's behavioral data and identify emotions such as joy, displeasure, and excitement.

[1175] Finally, the server identifies improvements and enhancements to the new business service based on the analysis results. This information is provided to the service development team to help improve the service. The server may further include means for identifying improvements and enhancements to the new business service by inputting a specific prompt sentence into the generative AI model.

[1176] Examples:

[1177] If the new business service is an online shopping service, the following simulation is performed.

[1178] 1. The server uses a generative AI model (e.g., GPT-4) to generate a pseudo-user who is a woman in her 30s.

[1179] 2. The server simulates an online shopping service that provides fashion items to this pseudo-user.

[1180] 3. Record the actions of a simulated user as they add a specific dress to their cart and complete the purchase.

[1181] 4. The server collects the behavioral data of the simulated user and uses an emotion analysis tool (e.g., IBM Watson's Natural Language Understanding) to analyze the emotional state of the simulated user when selecting a dress.

[1182] 5. Based on the analysis results, the server identifies areas for improvement in the online shopping service (for example, providing more detailed product descriptions or adding user reviews).

[1183] Example prompt sentence:

[1184] "Generate a pseudo-user, a woman in her 30s, and simulate an online shopping service that offers fashion items. Record which products the pseudo-user selects and what purchasing behavior they exhibit, and perform sentiment analysis based on that behavioral data."

[1185] In this way, the system can verify whether a new business service is acceptable to pseudo-users and identify areas for improvement in the service.

[1186] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1187] Step 1:

[1188] The server generates pseudo-users using generative artificial intelligence (e.g., GPT-4).

[1189] Input: Prompt text "Please generate a pseudo-user for a woman in her 30s"

[1190] Data processing: Generative AI generates pseudo-user attributes (age, gender, interests, purchasing history, etc.) based on the prompt text.

[1191] Output: Generated pseudo user data

[1192] Specific operation: The server inputs a prompt sentence into the generative AI model, obtains the generated pseudo-user data, and stores it in a database.

[1193] Step 2:

[1194] The server provides a new business service to the pseudo-user.

[1195] Input: Data of the generated pseudo-users, details of the new business service (e.g., online shopping service)

[1196] Data processing: The server provides the new business service to the pseudo-user and displays the service interface to the pseudo-user.

[1197] Output: Interface for pseudo-users to use the service

[1198] Specific operation: The server starts an online shopping service that provides fashion items to the pseudo-user.

[1199] Step 3:

[1200] The server simulates the actions of the pseudo-users and collects the results.

[1201] Input: Pseudo user data, interface for new business services

[1202] Data processing: The server simulates the actions of a pseudo-user (selecting a product, adding to cart, completing purchase, etc.) and records the action data.

[1203] Output: Pseudo user behavior data

[1204] Specific operation: The server simulates the actions of a pseudo-user from adding a specific dress to the cart to completing the purchase, and saves the action history in a database.

[1205] Step 4:

[1206] The server analyzes the emotional state of the simulated user based on the results of the simulation.

[1207] Input: Pseudo user behavior data

[1208] Data processing: Using sentiment analysis tools (e.g., IBM Watson's Natural Language Understanding), simulated user behavior data is analyzed to identify emotional states.

[1209] Output: Emotional state data of the simulated user

[1210] Specific operation: The server inputs the behavioral data of the simulated user into an emotion analysis tool to identify emotions such as joy, displeasure, and excitement.

[1211] Step 5:

[1212] The server identifies areas for improvement and enhancement of the new business service based on the analysis results.

[1213] Input: Emotional state data of a simulated user

[1214] Data processing: The server analyzes the emotional state data to identify areas for improvement and enhancement of the service.

[1215] Output: A list of improvements and enhancements

[1216] Specific behavior: Based on the results of the sentiment analysis, the server identifies improvements to the online shopping service, such as providing more detailed product descriptions or adding user reviews, and reports them to the development team.

[1217] (Application example 2)

[1218] Next, a description will be given of Application Example 2 of Form 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."

[1219] It is an important task for companies to verify in advance whether a new business service will be accepted by the market. However, testing with actual users is costly and time-consuming. It is also difficult to accurately grasp the user's emotional state and identify areas for improvement or strengthening of the service. To solve these problems, an efficient and low-cost method is needed to simulate the acceptance of new business services and analyze the user's emotional state.

[1220] 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.

[1221] In this invention, the server includes means for generating a large number of pseudo users using generative AI, means for running a simulation to verify whether a new business service will be accepted by the pseudo users, means for confirming the effectiveness of the new business service based on the results of the simulation, means for analyzing the emotional state of the pseudo users, and means for identifying areas for improvement or enhancement of the new business service based on the analysis results. This makes it possible to efficiently and low-costly simulate whether a new business service will be accepted in the market, accurately grasp the emotional state of users, and identify areas for improvement or enhancement of the service.

[1222] "Generative AI" is a system that uses artificial intelligence technology to generate new data from large amounts of data.

[1223] A "pseudo-user" is a virtual user created to mimic the behavior and preferences of a real user.

[1224] "Simulation" is a method of predicting results under specific conditions by imitating and virtually executing real-world situations.

[1225] "New business services" refer to new products or services that a company offers.

[1226] "Effectiveness" is an indicator of how effective something is for a specific purpose.

[1227] An "emotional state" is the emotional state that a user feels in response to a particular situation or stimulus.

[1228] "Areas for improvement" are parts of the current system or service that need to be corrected or changed to make it better.

[1229] "Areas of improvement" are areas of the current system or service that can be further strengthened to increase effectiveness.

[1230] This invention is a system that uses generative AI to generate a large number of pseudo-users, simulates whether a new business service will be accepted by the pseudo-users, confirms the effectiveness of the new business service based on the results, and further analyzes the emotional state of the pseudo-users. This system is implemented as follows.

[1231] The server first uses generative AI to generate a large number of pseudo-users. These pseudo-users are generated based on data that mimics the behavioral patterns and preferences of actual users. Next, the server provides the new business service to the pseudo-users and simulates their reactions. This simulation makes it possible to predict what actions the pseudo-users will take, which products they will select, and so on.

[1232] Based on the simulation results, the server confirms the effectiveness of the new business service. Furthermore, it collects feedback from the simulated users and analyzes their emotional state using a sentiment analysis tool (e.g., TextBlob). This analysis allows us to understand how the simulated users feel about the new business service.

[1233] As a concrete example, consider the case of simulating a new business service for an online shopping website. The server uses generative AI to generate 100 pseudo-users and provides each pseudo-user with a product list (e.g., "Product A," "Product B," and "Product C"). The pseudo-users randomly select a product and send their selection results and feedback to the server. The server performs sentiment analysis based on the feedback and classifies the pseudo-user's emotional state into "happy," "neutral," "unhappy," etc.

[1234] This system makes it possible to efficiently and inexpensively simulate whether a new business service will be accepted in the market, accurately grasp the emotional state of users, and identify areas for improvement and strengthening of the service.

[1235] An example of a prompt sentence might be:

[1236] "For a new business service, an online shopping site, please simulate which products pseudo-users select and what emotional state they experience. The product list is "Product A," "Product B," and "Product C." Analyze the emotional state of each pseudo-user based on their feedback and output the results."

[1237] In this way, the server can simulate the acceptance of a new business service and identify areas for improvement or enhancement of the service by analyzing the emotional state of the user.

[1238] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1239] Step 1:

[1240] The server uses generative AI to generate a large number of pseudo-users. As input, it uses data to mimic user behavior patterns and preferences. The generated pseudo-users are virtual users that mimic the behavior of real users. As output, it obtains a list of the generated pseudo-users.

[1241] Step 2:

[1242] The server provides the new business service to the pseudo-users and simulates their reactions. As input, it uses the list of generated pseudo-users and information about the new business service. It simulates how the pseudo-users will behave in response to the provided service. As output, it obtains the behavioral data of each pseudo-user.

[1243] Step 3:

[1244] The server checks the effectiveness of the new business service based on the results of the simulation. As input, it uses the behavioral data of the pseudo-users. The server analyzes which products each pseudo-user selected and what purchasing behavior they exhibited. As output, it obtains a report on the effectiveness of the new business service.

[1245] Step 4:

[1246] The server collects the feedback of the pseudo-users and analyzes their emotional states using a sentiment analysis tool. As input, it uses the feedback data of the pseudo-users. The sentiment analysis tool (e.g., TextBlob) analyzes the text of the feedback and classifies the emotional states into "happy," "neutral," "unhappy," etc. As output, it obtains the emotional state data of each pseudo-user.

[1247] Step 5:

[1248] The server identifies areas for improvement and enhancement of the new business service based on the analysis results. As input, it uses emotional state data and behavioral data. The server integrates these data to identify which parts of the new business service bring pleasure to the user and which parts cause discomfort. As output, it obtains a report on areas for improvement and enhancement.

[1249] In this way, the server can simulate the acceptance of a new business service and identify areas for improvement or enhancement of the service by analyzing the emotional state of the user.

[1250] Example 3

[1251] Next, a description will be given of Example 3 of Form Example 3. 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."

[1252] To assess market acceptance of new business services in advance, it is important to obtain feedback from actual users. However, testing with actual users is time-consuming and costly, so an efficient method is required. Furthermore, unless realistic simulations that take into account the emotional state of users are performed, it is difficult to accurately predict the actual market reaction.

[1253] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[1254] In this invention, the server includes means for generating a large number of pseudo-users using generative artificial intelligence, means for running a simulation to verify whether a new business service will be accepted by the pseudo-users, means for verifying the effectiveness of the new business service based on the results of the simulation, means for recognizing the emotional state of the pseudo-users using an emotion engine and adjusting their behavioral patterns, and means for storing the simulation results in a database and analyzing the collected data, thereby enabling efficient and accurate evaluation of the market acceptance of the new business service.

[1255] "Generative AI" is an AI technology that generates text and data using natural language processing and machine learning.

[1256] A "pseudo-user" is a virtual user created to mimic the behavioral patterns and emotional state of a real user.

[1257] "Simulation" is a process of virtually verifying market acceptance of a new business service using pseudo-users.

[1258] An "emotion engine" is a software technology that recognizes a user's emotional state and adjusts behavioral patterns based on that state.

[1259] A "database" is a system for systematically storing and managing collected simulation results and other data.

[1260] "Data analysis" is the process of analyzing collected data using statistical methods and machine learning to extract useful information and patterns.

[1261] "New business services" are products or services that a company or organization intends to bring to market.

[1262] The present invention is a system for efficiently and accurately evaluating the market acceptance of new business services. A specific embodiment of this system will be described below.

[1263] Hardware and software used

[1264] Hardware: High-performance servers (e.g., high-performance computers)

[1265] Software: Generative AI (e.g., natural language processing models), emotion engines (e.g., emotion recognition software)

[1266] System Overview

[1267] This system uses generative artificial intelligence to generate a large number of pseudo-users and runs simulations to verify whether new business services will be accepted by the pseudo-users. Furthermore, it checks the effectiveness of the new business services based on the results of the simulations and identifies areas for improvement or strengthening.

[1268] Creating a pseudo user

[1269] The server uses generative artificial intelligence to generate a pseudo-user based on the prompt text. The generated pseudo-user mimics the behavioral patterns and emotional states of a real user, enabling a more realistic simulation.

[1270] Using the Emotion Engine

[1271] The server uses the emotion engine to recognize the emotional state of the pseudo-user and adjust its behavioral patterns. For example, if the pseudo-user shows the emotion of "joy," the emotion engine adjusts its behavioral patterns to be positive.

[1272] Running the simulation

[1273] The server uses the generated pseudo-users to simulate the reactions of new business services. The simulation results are stored in a database and later analyzed in detail.

[1274] Data collection and analysis

[1275] The server collects the simulation results and stores them in a database. The collected data is analyzed using data analysis tools to identify the likelihood of new business services being accepted and areas for improvement.

[1276] Specific examples

[1277] Simulation example

[1278] New Business Service: Online Learning Platform

[1279] Pseudo-user reactions: Satisfaction with pricing, feedback on delivery methods, opinions on target user selection

[1280] Example prompts for generative AI models

[1281] We are planning to provide an online learning platform as a new business service. Please simulate the reactions of pseudo-users under the following conditions.

[1282] Pricing: ¥5,000 per month

[1283] Delivery method: Video lectures and live sessions

[1284] Target users: University students and working adults

[1285] This system allows the user to check the effectiveness of a new business service in advance and identify areas that need improvement. The flow of the identification process in the third embodiment will be described with reference to FIG.

[1286] Step 1:

[1287] Creating a pseudo user

[1288] Input: Prompt (e.g., "We are planning to provide an online learning platform as a new business service. Please simulate the reactions of simulated users under the following conditions: Pricing: 5,000 yen per month, Delivery method: video lectures and live sessions, Target users: university students and working adults.")

[1289] Specific operation: The server uses generative AI to generate a pseudo-user based on the prompt. The generative AI analyzes the prompt and generates the pseudo-user's profile and behavioral patterns.

[1290] Output: Pseudo-user data (e.g., a pseudo-user profile of a 30-year-old male who works in an IT company and whose hobby is reading)

[1291] Step 2:

[1292] Recognizing emotional states and adjusting behavioral patterns

[1293] Input: Pseudo user data

[1294] Specific operation: The server uses the emotion engine to recognize the emotional state of the pseudo-user. The emotion engine analyzes the pseudo-user's profile and behavioral patterns to identify the emotional state. For example, if the pseudo-user shows the emotion of "joy," the emotion engine adjusts its behavioral patterns to be more positive.

[1295] Output: Adjusted simulated user behavior patterns

[1296] Step 3:

[1297] Running the simulation

[1298] Input: Adjusted simulated user behavior patterns, details of new business services (e.g., pricing, delivery method)

[1299] Specific operation: The server uses the generated pseudo-user to simulate the reaction to the new business service. In the simulation, the details of the new business service are presented to the pseudo-user and their reaction is recorded.

[1300] Output: Simulation results (e.g., simulated user feedback and behavioral data)

[1301] Step 4:

[1302] Data collection and storage

[1303] Input: Simulation results

[1304] Specific operation: The server collects the simulation results and stores them in a database. The collected data includes feedback and behavioral data of simulated users.

[1305] Output: Saved simulation data

[1306] Step 5:

[1307] Analyzing the data

[1308] Input: Saved simulation data

[1309] Specific operations: The server uses a data analysis tool to statistically analyze the collected data. The data analysis tool analyzes the feedback and behavioral data of the simulated users to identify the acceptability of the new business service and areas for improvement.

[1310] Output: Analysis results (e.g., improvements such as "change pricing to 3,000 yen per month" or "change delivery method from video lectures to live sessions")

[1311] In this way, the server can efficiently and accurately assess market acceptance of new business services.

[1312] (Application example 3)

[1313] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[1314] To assess market acceptance of new business services in advance, it is important to predict actual user reactions and identify areas for improvement. However, conventional methods have had issues with the difficulty of accurately imitating users' emotional states and behavioral patterns, resulting in low reliability of simulation results. Furthermore, there has been a lack of means to effectively analyze collected data and identify specific areas for improvement.

[1315] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[1316] In this invention, the server includes means for generating a large number of pseudo users using generative AI, means for running a simulation to verify whether a new business service will be accepted by the pseudo users, means for verifying the effectiveness of the new business service based on the results of the simulation, means for understanding the emotional state of the user using an emotion engine and adjusting the behavioral patterns of the pseudo users based on that understanding, means for simulating the behavioral patterns of the pseudo users and collecting behavioral data, means for analyzing the collected behavioral data using a clustering algorithm, and means for outputting the analysis results as a report. This enables highly accurate simulations that reflect the emotional state of the user and identifies specific areas for improvement.

[1317] "Generative AI" is a system that generates data and information using artificial intelligence technology.

[1318] A "pseudo-user" is a virtual user created to mimic the behavior and emotions of a real user.

[1319] "Simulation" is a method of conducting experiments and verification by imitating real-world situations.

[1320] An "emotion engine" is a technology that analyzes a user's emotional state and adjusts behavioral patterns based on that.

[1321] A "behavioral pattern" is a series of actions that a user tends to take under certain circumstances.

[1322] A "clustering algorithm" is a computational method for grouping data based on similarity.

[1323] A "report" is a document that summarizes the results of analysis and research.

[1324] "New business services" are products or services that are newly introduced to the market.

[1325] "Effectiveness" is an indicator of how effective something is in achieving a specific purpose or goal.

[1326] "Data collection" is the process of gathering the necessary information or data.

[1327] A system for implementing the present invention includes a generative AI, an emotion engine, a clustering algorithm, and a report generation function. A specific embodiment of the system is described below.

[1328] System Configuration

[1329] The server uses generative AI to generate a large number of pseudo-users. The pseudo-users are generated based on data that mimics the user's behavioral patterns and preferences. The emotion engine analyzes the user's emotional state and adjusts the pseudo-user's behavioral patterns accordingly. This allows the pseudo-users to more accurately reflect the emotional state of the actual user.

[1330] simulation

[1331] The server runs a simulation to verify whether the new business service will be accepted by pseudo-users. It simulates the behavioral patterns of the pseudo-users and collects behavioral data. The collected behavioral data is analyzed using a clustering algorithm. Specifically, it uses a clustering algorithm such as KMeans to group the behavioral patterns of the pseudo-users.

[1332] Data Analysis and Report Generation

[1333] The server checks the effectiveness of the new business service based on the simulation results. It also outputs the analysis results as a report. The report includes areas for improvement and strengthening of the new business service, which can be used to improve business strategies.

[1334] Hardware and software used

[1335] Generative AI: Uses machine learning frameworks such as TensorFlow and PyTorch.

[1336] Emotion engine: Uses an emotion analysis engine that uses natural language processing technology.

[1337] Clustering algorithms: Use a data analysis library such as scikit-learn.

[1338] Report generation: Use data visualization libraries such as Pandas and Matplotlib.

[1339] Specific examples

[1340] For example, when selling a new fashion item on an online shopping site, this system is used to simulate the reactions of pseudo-users. The emotion engine analyzes the user's emotional state and generates behavioral patterns for the pseudo-user to "purchase," "view," and "share" the item. From the simulation results, areas for improvement in pricing and promotion methods are identified and output as a report.

[1341] Prompt Sentence Examples

[1342] Simulate the behavioral patterns of simulated users to evaluate pricing and promotion methods for new fashion items. Analyze users' emotional states using an emotion engine to generate simulated users. Identify areas for improvement from the simulation results and generate reports.

[1343] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[1344] Step 1:

[1345] The server uses generative AI to generate a large number of pseudo-users. As input, it uses data to mimic user behavior patterns and preferences. The generative AI generates pseudo-users based on this data and outputs a list of pseudo-users. Specifically, it uses a machine learning framework (such as TensorFlow or PyTorch) to learn user behavior patterns and generate pseudo-users based on them.

[1346] Step 2:

[1347] The server uses an emotion engine to analyze the emotional state of the pseudo-user. It uses a list of pseudo-users as input. The emotion engine analyzes the emotional state of the pseudo-user using natural language processing technology and obtains a list of pseudo-users with assigned emotional states as output. Specifically, the emotion analysis engine extracts the emotional state from text data and behavioral data.

[1348] Step 3:

[1349] The server adjusts the behavioral patterns of the pseudo-users based on their emotional states. As input, it uses a list of pseudo-users with assigned emotional states. The emotion engine adjusts the behavioral patterns of the pseudo-users according to their emotional states, and as output, it obtains a list of pseudo-users with adjusted behavioral patterns. Specifically, it applies an algorithm that changes behavioral patterns based on their emotional states.

[1350] Step 4:

[1351] The server runs a simulation to verify whether the new business service will be accepted by the pseudo users. As input, it uses a list of pseudo users with adjusted behavioral patterns and information about the new business service. The simulation imitates the behavior of the pseudo users and collects behavioral data as output. Specifically, it simulates how the pseudo users will behave toward the new business service.

[1352] Step 5:

[1353] The server analyzes the collected behavioral data using a clustering algorithm. It uses the behavioral data obtained from the simulation as input. It groups the behavioral data using a clustering algorithm (e.g., KMeans) and obtains the clustering results as output. Specifically, it clusters the behavioral data using a data analysis library such as scikit-learn.

[1354] Step 6:

[1355] The server outputs the analysis results as a report. The clustering results are used as input. The report generation function identifies areas for improvement and strengthening of the new business service based on the analysis results, and generates a report as output. Specifically, it uses data visualization libraries such as Pandas and Matplotlib to visually represent the analysis results.

[1356] 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.

[1357] 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> ) 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.

[1358] Another example of generative AI is Gemini (internet search engine). <url: https: gemini.google.com ?hl="ja">) are mentioned.

[1359] 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.

[1360] [Third embodiment]

[1361] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1362] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[1363] 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).

[1364] 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.

[1365] 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.

[1366] 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).

[1367] 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.

[1368] 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.

[1369] 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.

[1370] 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.

[1371] 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.

[1372] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[1373] "Example 1"

[1374] One embodiment of the present invention is a method for generating a large number of pseudo users using a generative AI. Specifically, the generative AI generates data to imitate the behavioral patterns and preferences of users.

[1375] For example, past user behavior data and user profile data can be used as learning data.

[1376] "Example 2"

[1377] Furthermore, there is a means for executing a simulation to verify whether a new business service will be accepted by pseudo-users. Specifically, the new business service is provided to the generated pseudo-users, and their reactions are simulated. For example, if the new business service is an online shopping service, it is possible to simulate which products the pseudo-users will select and what purchasing behavior they will exhibit.

[1378] "Example 3"

[1379] There are also methods to verify the effectiveness of new business services based on the results of simulations. Specifically, the results of the simulations can be used to analyze the acceptability of new business services and areas for improvement. For example, based on the reactions of simulated users, it is possible to identify areas for improvement such as pricing and delivery methods for new business services, and selection of target users.

[1380] The processing flow of each embodiment will be described below.

[1381] "Example 1"

[1382] Step 1: The generative AI learns data to mimic user behavior patterns and preferences. This learning data is based on past user behavior data and user profile data.

[1383] Step 2: Based on the learned data, the generative AI generates pseudo-users, who are then used to verify the acceptability of the new business service.

[1384] "Example 2"

[1385] Step 1: Provide a new business service to the generated pseudo-users. This new business service may be, for example, an online shopping service.

[1386] Step 2: Simulate the reaction of the pseudo-user. Specifically, simulate which products the pseudo-user will select and what purchasing behavior they will exhibit.

[1387] "Example 3"

[1388] Step 1: Analyze the acceptability of the new business service and areas for improvement based on the results of the simulation.

[1389] Step 2: Based on the analysis results, identify areas for improvement such as pricing and delivery methods for the new business service, and selection of target users.

[1390] Example 1

[1391] Next, a description will be given of Example 1 of Form 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."

[1392] Conventional methods for testing new business services involve using actual user data, which poses challenges such as privacy issues and the high cost of data collection. Furthermore, testing based on actual user data takes time, and it is often difficult to obtain prompt feedback. Furthermore, it is difficult to accurately mimic user behavior patterns and preferences, resulting in low simulation accuracy.

[1393] 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.

[1394] In this invention, the server includes: means for generating a large number of pseudo-users using a generative AI model; means for collecting and preprocessing past user behavior data and user profile data; means for training the generative AI model based on the preprocessed data; means for generating pseudo-users by inputting prompt sentences to the trained generative AI model; means for running a simulation to verify whether a new business service will be accepted by the pseudo-users; means for verifying the effectiveness of the new business service based on the results of the simulation; and means for saving the generated pseudo-user data and using it in other systems or applications. This enables rapid and highly accurate verification of new business services while avoiding privacy issues.

[1395] A "generative AI model" is an artificial intelligence model that learns user behavior patterns and preferences and generates pseudo-users based on them.

[1396] A "pseudo-user" is a virtual user created using a generative AI model that mimics the behavioral patterns and preferences of a real user.

[1397] "User behavior data" is data related to past behavior of a user, and includes, for example, purchase history and browsing history.

[1398] "User profile data" refers to data relating to a user's attributes and preferences, including, for example, age, gender, and interests.

[1399] "Preprocessing" refers to the process of converting collected data into a format suitable for a generative AI model, and includes data cleaning, normalization, and feature extraction.

[1400] A "prompt sentence" is text that is input into a generative AI model and contains instructions and conditions for generating a pseudo-user.

[1401] A "simulation" is a virtual experiment using generated pseudo-users to verify whether a new business service will be accepted.

[1402] "New business services" refer to new products or services offered by companies or organizations.

[1403] "Effectiveness" refers to the ability or effectiveness of a new business service to achieve its intended results.

[1404] "Saving" means recording the data of the generated pseudo-user in a database or the like so that it can be used later.

[1405] "Usage" means utilizing the saved pseudo-user data in other systems or applications.

[1406] This invention relates to a system for generating a large number of pseudo-users using a generative AI model and verifying whether a new business service will be accepted by the pseudo-users. A specific embodiment of this system will be described below.

[1407] The server first collects past user behavior data and user profile data from a database. This data includes information such as the actions users have taken and their preferences. The collected data is preprocessed using Python's Pandas library, among other tools. Preprocessing includes data cleaning, normalization, and feature extraction.

[1408] The server then trains a generative AI model based on the preprocessed data. The generative AI model can be, for example, OpenAI's GPT-4. The trained model has the ability to learn user behavior patterns and preferences and generate pseudo-users based on them.

[1409] Once training is complete, the server inputs a prompt to the generative AI model. The prompt provides the generative AI model with instructions and conditions for generating a pseudo-user. For example, the following prompts can be used:

[1410] "User A frequently shops online on weekday evenings. He tends to buy mainly electronic devices and books. His purchase history includes the latest smartphones and best-selling novels. Please generate a pseudo-user that mimics User A's behavioral patterns."

[1411] When this prompt is entered, the generative AI model generates a pseudo-user that mimics the behavioral patterns and preferences of User A. The generated pseudo-user is expected to behave in a similar way to User A.

[1412] Using the generated pseudo-users, the server runs a simulation to verify whether the new business service will be accepted by the pseudo-users. Based on the results of the simulation, the effectiveness of the new business service is confirmed, and areas for improvement or strengthening are identified as necessary.

[1413] Finally, the server stores the generated pseudo-user data in a database and makes it available to other systems and applications, enabling rapid and highly accurate verification of new business services while avoiding privacy issues.

[1414] The above is a specific embodiment for carrying out the present invention.

[1415] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1416] Step 1: Data collection

[1417] The server retrieves past user behavior data and user profile data from the database. As input, it uses database connection information and SQL queries. As output, it obtains the retrieved raw data. Specifically, the server executes the SQL queries to extract user behavior history and profile information.

[1418] Step 2: Data Preprocessing

[1419] The server preprocesses the acquired raw data and converts it into a format suitable for generative AI models. The acquired raw data is used as input, and preprocessed data is obtained as output. Specifically, the server cleans the data (filling in missing values ​​and removing outliers), normalizes it, and extracts features.

[1420] Step 3: Train the model

[1421] The server trains a generative AI model based on the preprocessed data. As input, it uses the preprocessed data and a generative AI model (e.g., GPT-4). As output, it obtains a trained generative AI model. Specifically, the server inputs the preprocessed data into the model and optimizes the model's parameters.

[1422] Step 4: Enter the prompt statement

[1423] The server inputs a prompt sentence into the trained generative AI model. The prompt sentence and the trained generative AI model are used as input. The generated pseudo-user is obtained as output. Specifically, the server inputs the prompt sentence into the model, and the model generates a pseudo-user.

[1424] Step 5: Run the simulation

[1425] The server uses the generated pseudo-users to run a simulation to verify whether the new business service will be accepted. The generated pseudo-users and the scenario of the new business service are used as input. The simulation results are obtained as output. Specifically, the server simulates the behavior of the pseudo-users and evaluates the acceptance of the service.

[1426] Step 6: Saving and using the results

[1427] The server stores the generated pseudo-user data and simulation results, making them available to other systems and applications. The generated pseudo-user data and simulation results are used as input, and the stored data is obtained as output. Specifically, the server stores the data in a database and provides the data to other systems through an API.

[1428] The above are the specific processing steps of the program of this system.

[1429] (Application example 1)

[1430] Next, a description will be given of Application Example 1 of Form 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."

[1431] Conventional methods for testing new business services require the use of actual user data, which is time-consuming and costly, and also poses problems from the perspective of protecting user privacy. It is also difficult to accurately grasp user behavioral patterns and preferences, making it difficult to provide personalized services. Furthermore, product recommendation systems are required to make highly accurate recommendations based on user preferences, but current systems are unable to adequately meet this demand.

[1432] 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.

[1433] In this invention, the server includes means for generating a large number of pseudo users using generative AI, means for running a simulation to verify whether a new business service will be accepted by the pseudo users, means for verifying the effectiveness of the new business service based on the results of the simulation, means for recommending products based on the behavioral data of the pseudo users, and means for providing a personalized shopping experience based on user preferences. This enables efficient and accurate verification of new business services and personalized product recommendations while protecting user privacy.

[1434] "Generative AI" is an artificial intelligence technology that learns user behavior patterns and preferences and generates pseudo-users based on them.

[1435] A "pseudo-user" is a virtual user created by generative AI that mimics the behavioral patterns and preferences of a real user.

[1436] A "simulation" is a virtual experiment using pseudo-users to verify the extent to which a new business service will be accepted.

[1437] "New business services" are new products or services offered by a company, which provide new value to users.

[1438] "Effectiveness" refers to the ability of a new business service to be accepted by users and to achieve expected results.

[1439] "Behavioral Data" is historical information about the actions and choices a user makes on a website or application.

[1440] "Product recommendation" refers to proposing appropriate products to users based on their behavioral data and preferences.

[1441] A "personalized shopping experience" is a shopping experience that provides optimal products and services based on the user's individual preferences and behavioral patterns.

[1442] A system for implementing this invention includes means for generating a large number of pseudo-users using generative AI, means for running a simulation to verify whether a new business service will be accepted by the pseudo-users, means for confirming the effectiveness of the new business service based on the results of the simulation, means for recommending products based on behavioral data of the pseudo-users, and means for providing a personalized shopping experience based on user preferences.

[1443] System Program

[1444] The server generates pseudo-users using a generative AI model. The generative AI model learns past user behavior data and user profile data, and generates pseudo-users based on that data. The generated pseudo-users mimic the behavioral patterns and preferences of actual users.

[1445] Next, the server executes a simulation to verify whether the new business service will be accepted by the pseudo-users. This simulation makes it possible to confirm the effectiveness of the new business service.

[1446] Furthermore, the server recommends products based on the behavioral data of the simulated user, thereby providing a personalized shopping experience based on the user's preferences.

[1447] Hardware and software used

[1448] Hardware: Servers, smartphones

[1449] Software: Python, generative AI models (e.g., GPT-3)

[1450] Data processing and calculation

[1451] 1. Data collection: The server collects past user behavior data and user profile data.

[1452] 2. Data training: The server trains the generative AI model on this data.

[1453] 3. Pseudo-user generation: The server generates pseudo-users using the trained model.

[1454] 4. Simulation execution: The server executes a simulation of the new business service using a pseudo user.

[1455] 5. Product recommendation: The server recommends products based on the behavioral data of the pseudo-user.

[1456] 6. Personalized Offering: The server provides a personalized shopping experience based on the user's preferences.

[1457] Specific examples

[1458] For example, if a user is looking for "smartphone accessories," the server will refer to the pseudo-user's behavioral data and recommend related products (cases, chargers, earphones, etc.).

[1459] Prompt Sentence Examples

[1460] Generate pseudo-users based on user behavior data and recommend products based on the user's preferences. For example, recommend related products based on the data of users who have previously purchased "smartphone accessories."

[1461] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1462] Step 1:

[1463] The server collects historical user behavior and user profile data, including the user's actions and selections on websites and applications. The input is the user's behavior and profile data, and the output is the data stored on the server. Specifically, the server retrieves the required data by querying the database.

[1464] Step 2:

[1465] The server trains the collected data into a generative AI model. The inputs are the user behavior data and profile data collected in step 1, and the output is a trained generative AI model. Specifically, the server preprocesses the data and feeds it to the generative AI model.

[1466] Step 3:

[1467] The server generates a pseudo-user using a trained generative AI model. The input is the trained generative AI model, and the output is the generated pseudo-user data. Specifically, the server inputs prompts into the generative AI model to generate the pseudo-user's profile and behavioral patterns.

[1468] Step 4:

[1469] The server runs a simulation of the new business service using the generated pseudo-user. The inputs are the pseudo-user data and the specifications of the new business service, and the output is the simulation results. Specifically, the server simulates how the pseudo-user uses the new business service and records the results.

[1470] Step 5:

[1471] The server checks the effectiveness of the new business service based on the simulation results. The input is the simulation results, and the output is an evaluation of the effectiveness of the new business service. Specifically, the server analyzes the simulation results and identifies the degree of acceptance of the service and areas for improvement.

[1472] Step 6:

[1473] The server recommends products based on the behavioral data of the pseudo-user. The input is the behavioral data of the pseudo-user, and the output is a list of recommended products. Specifically, the server analyzes the pseudo-user's preferences and behavioral patterns to select the most suitable products.

[1474] Step 7:

[1475] The server provides a personalized shopping experience based on the user's preferences. The inputs are a list of recommended products and the user's preference data, and the output is a personalized shopping experience. Specific operations include displaying products that match the user's preferences and providing an interface that encourages purchases.

[1476] Example 2

[1477] Next, a description will be given of Example 2 of Form 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."

[1478] Before launching a new business service to the market, it is important to verify whether the service will be accepted by users. However, testing with actual users is time-consuming and costly, so an efficient method is required. In addition, it is difficult to accurately imitate user behavior patterns and preferences, which can result in low simulation accuracy. To solve these issues, a simulation system using pseudo-users is needed.

[1479] 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.

[1480] In this invention, the server includes means for generating a large number of pseudo users using generative artificial intelligence, means for running a simulation to verify whether a new business service will be accepted by the pseudo users, means for verifying the effectiveness of the new business service based on the results of the simulation, means for providing the new business service to the generated pseudo users and simulating their reactions, and means for displaying the results of the simulation to users. This makes it possible to efficiently and accurately verify the acceptability of a new business service before it is released to the market.

[1481] "Generative AI" is a system that generates data and information using artificial intelligence technology.

[1482] A "pseudo user" is a virtual user created by imitating the behavioral patterns and preferences of an actual user.

[1483] "New business services" refer to products or services that are planned to be newly introduced to the market.

[1484] A "simulation" is a virtual experiment using pseudo-users to verify the acceptability of a new business service.

[1485] "Simulation results" refers to data and information obtained after running a simulation.

[1486] "User" refers to a person or organization that uses this system to verify new business services.

[1487] "Generating means" refers to a method or device for creating a pseudo-user using generative artificial intelligence.

[1488] The "means for providing" refers to a method or device for presenting the new business service to the pseudo-user.

[1489] The "display means" refers to a method or device for visually showing the simulation results to the user.

[1490] This invention is a system that uses generative artificial intelligence to generate pseudo-users and simulates whether a new business service will be accepted by the pseudo-users. A specific embodiment of this system will be described below.

[1491] The server first generates a pseudo-user using a generative AI model (e.g., a general natural language processing model). Examples of generative AI models used include OpenAI's GPT-4. The generated pseudo-users have attributes such as age, gender, and interests.

[1492] Next, the server provides the new business service to the simulated user. If this service is an online shopping service, the server presents a product list to the simulated user. The simulated user simulates which product to select and what purchasing behavior to take through the generative AI model.

[1493] The terminal provides an interface that allows the user to check the simulation results, and the user can find areas for improvement in the new business service based on the simulation results.

[1494] As a specific example, if the new business service is an "online shopping service," the following prompt sentence would be input into the generative AI model:

[1495] Example prompt sentence:

[1496] Pseudo-user A is a 30-year-old male who is interested in technology products. Please present him with the following product list and simulate which product he would choose.

[1497] 1. Smartphone

[1498] 2. Laptop

[1499] 3. Smartwatch

[1500] 4. Wireless earphones

[1501] The server inputs this prompt into the generative AI model and simulates the purchasing behavior of pseudo-user A. For example, if pseudo-user A selects "smartphone," the reason for the selection and the degree of interest in other products are also output as simulation results.

[1502] Users can check the simulation results on their devices and identify areas for improvement in their new business services. For example, if the adoption rate of smartwatches is low, they can take measures such as revising the product description or pricing.

[1503] This system makes it possible to efficiently and accurately verify the acceptability of new business services before they are brought to market.

[1504] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1505] Step 1:

[1506] The server generates a pseudo-user using a generative AI model. As input, it inputs a prompt statement specifying the pseudo-user's attributes (e.g., age, gender, interests) into the generative AI model. The generative AI model generates a pseudo-user based on this prompt statement and obtains the pseudo-user's data as output. Specifically, the server inputs the following prompt statement into the generative AI model:

[1507] Generate a simulated user who is a 30-year-old male and interested in technology products.

[1508] The generative AI model generates a pseudo-user based on this prompt sentence.

[1509] Step 2:

[1510] The server provides a new business service to the generated pseudo-user. As input, it uses the pseudo-user data and new business service information (e.g., product list). The server presents the product list to the pseudo-user and obtains the pseudo-user's reaction data as output. Specifically, the server presents the following product list to the pseudo-user.

[1511] 1. Smartphone

[1512] 2. Laptop

[1513] 3. Smartwatch

[1514] 4. Wireless earphones

[1515] Step 3:

[1516] The server uses a generative AI model to simulate the pseudo-user's reaction. The pseudo-user's data and product list are used as input. The generative AI model simulates which products the pseudo-user will select and what purchasing behavior they will take, and obtains the simulation results as output. Specifically, the server inputs the following prompt sentence into the generative AI model:

[1517] Pseudo-user A is a 30-year-old male who is interested in technology products. Please present him with the following product list and simulate which product he would choose.

[1518] 1. Smartphone

[1519] 2. Laptop

[1520] 3. Smartwatch

[1521] 4. Wireless earphones

[1522] The generative AI model simulates the purchasing behavior of pseudo-user A based on this prompt sentence.

[1523] Step 4:

[1524] The terminal provides an interface that allows the user to check the simulation results. The simulation results are used as input. The terminal visually displays the simulation results and provides a state that allows the user to check the results as output. Specifically, the terminal displays the simulation results as follows:

[1525] Pseudo-user A chose "Smartphone." Reason for choosing: Because he is interested in the latest technology.

[1526] Based on these results, users can consider ways to improve their product listings and service content.

[1527] (Application example 2)

[1528] Next, a description will be given of Application Example 2 of Form 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."

[1529] It is an important task for companies to verify in advance whether a new business service will be accepted by the market. However, testing using actual users is costly and time-consuming. It is also difficult to accurately grasp user behavior patterns and preferences. This has led to a demand for a method to efficiently confirm the effectiveness of new business services and identify areas for improvement and strengthening.

[1530] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for generating a large number of pseudo users using generative AI, means for running a simulation to verify whether a new business service will be accepted by the pseudo users, means for confirming the effectiveness of the new business service based on the results of the simulation, means for simulating the products selected and purchasing behavior of the pseudo users, and means for analyzing the simulation results and identifying which products were most selected. This makes it possible to efficiently verify the market acceptance of a new business service and identify areas for improvement and enhancement.

[1531] "Generative AI" is a system that generates data and information using artificial intelligence technology.

[1532] A "pseudo user" is a virtual user created by imitating the behavioral patterns and preferences of an actual user.

[1533] "Simulation" is a method of imitating and virtually reproducing real-world events and processes.

[1534] "New business services" refer to new products or services that a company offers.

[1535] "Effectiveness" is an indicator of how effective something is in achieving a specific purpose or goal.

[1536] "Purchasing behavior" refers to the series of actions and decision-making process that consumers take from selecting a product to purchasing it.

[1537] Analysis is the process of examining data or information in detail to find meaning and patterns.

[1538] "Areas for improvement" are areas in the current system or process that need to be changed or modified to make it better.

[1539] "Enhancements" are areas of the current system or process that need further strengthening or expansion.

[1540] A system for implementing this invention includes means for generating a large number of pseudo-users using generative AI, means for running a simulation to verify whether a new business service will be accepted by the pseudo-users, means for confirming the effectiveness of the new business service based on the results of the simulation, means for simulating the products selected and purchasing behavior of the pseudo-users, and means for analyzing the results of the simulation and identifying which products were most selected.

[1541] The server uses generative AI to generate pseudo-users. These pseudo-users mimic the behavioral patterns and preferences of real users, enabling a realistic simulation. The generated pseudo-users are used to simulate reactions to new business services.

[1542] The simulation is run on a server, reproducing which products a simulated user would select and what purchasing behavior they would exhibit. The results of the simulation are collected and analyzed by the server. Based on the analysis results, the effectiveness of the new business service is confirmed and areas for improvement and enhancement are identified.

[1543] The implementation of this system uses the following hardware and software:

[1544] Hardware: Servers, smartphones

[1545] Software: Python, NumPy, generative AI models

[1546] As a concrete example, the following prompt sentence is input into the generative AI model:

[1547] Example prompt sentence:

[1548] To simulate adoption of a new online shopping service, create a simulated user and simulate their choice of the following products:

[1549] Product A: Price 100 yen, quality 80

[1550] Product B: Price 150 yen, quality 90

[1551] Product C: Price 200 yen, quality 85

[1552] The pseudo-user preferences are as follows:

[1553] User 1: Price-sensitive (-0.5), Quality-sensitive (1.0)

[1554] User 2: Price-sensitive (-0.3), Quality-sensitive (0.8)

[1555] Please let me know the simulation results.

[1556] By inputting this prompt into a generative AI model, it is possible to simulate the purchasing behavior of pseudo-users and verify their acceptance of new services.

[1557] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1558] Step 1:

[1559] The server uses generative AI to generate pseudo-users. Data on user behavior patterns and preferences is used as input. Based on this data, the generative AI generates a large number of pseudo-users that mimic the behavior of actual users. The output is a list of the generated pseudo-users.

[1560] Step 2:

[1561] The server runs a simulation to provide a new business service to the generated pseudo-users. As input, it uses a list of pseudo-users and detailed information about the new business service. The simulation reproduces which products the pseudo-users select and what purchasing behavior they exhibit. The output is data on the products selected and purchasing behavior of each pseudo-user.

[1562] Step 3:

[1563] The server collects and analyzes the results of the simulation. It uses the simulation result data as input. It uses data analysis tools to identify which products were most popular and what purchasing behaviors were most common. The output is a report of the analysis results.

[1564] Step 4:

[1565] The server checks the effectiveness of the new business service based on the analysis results. It uses the analysis result report as input. It evaluates whether the new business service will be accepted by the market and identifies areas for improvement or enhancement, if necessary. The output is an evaluation report of the new business service.

[1566] Step 5:

[1567] The server identifies areas for improvement and enhancement of the new business service based on the evaluation report and reflects them in the next simulation. The evaluation report is used as input. A new simulation reflecting the improvements and enhancements is prepared and verified again using simulated users. The output is the simulation results of the improved new business service.

[1568] Example 3

[1569] Next, a third embodiment of the third embodiment will be described. 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."

[1570] There is a lack of effective methods to verify whether a new business service will be accepted by target users before it is launched on the market. It is also difficult to identify specific areas for improvement or strengthening based on simulation results. This makes it difficult to obtain appropriate feedback to increase the success rate of the business.

[1571] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[1572] In this invention, the server includes means for generating a large number of pseudo users using generative AI, means for running a simulation to verify whether a new business service will be accepted by the pseudo users, means for collecting simulation result data, means for inputting the collected data into a generative AI model and analyzing the data, means for feeding back the analysis results output from the generative AI model to users, and means for verifying the effectiveness of the new business service based on the simulation results. This makes it possible to effectively verify the acceptability of a new business service before it is released to the market and identify specific areas for improvement and enhancement.

[1573] "Generative AI" is an artificial intelligence technology that generates new data and models based on large amounts of data.

[1574] A "pseudo-user" is a virtual user created to mimic the behavior and preferences of a real user.

[1575] "Simulation" is a technology that imitates real-world systems and processes and virtually reproduces their behavior.

[1576] "Simulation result data" refers to various data and information obtained when a simulation is executed.

[1577] A "generative AI model" is a model that uses generative AI to perform specific tasks or generate data.

[1578] "Data analysis" is the process of extracting useful information and insights from collected data.

[1579] "Feedback" refers to returning the results or evaluation of a system or process to the user.

[1580] "New business services" are services or products that are planned to be newly introduced to the market.

[1581] "Effectiveness" is an indicator of how effective something is in achieving a specific purpose or goal.

[1582] "Areas for improvement" are parts or elements of the current system or process that need improvement.

[1583] "Improvement points" are parts or elements of the current system or process that need to be further strengthened.

[1584] The present invention provides a system for effectively verifying the acceptability of a new business service before it is launched into the market and identifying specific areas for improvement or strengthening. Specific embodiments of this system will be described below.

[1585] Hardware and software used

[1586] Hardware: Servers, terminals

[1587] Software: Generative AI models (e.g., GPT-4), simulation software (e.g., AnyLogic)

[1588] System configuration

[1589] 1. The server uses generative AI to generate a large number of pseudo-users. Pseudo-users are virtual users created to mimic the behavior and preferences of real users.

[1590] 2. The server runs a simulation to verify whether the new business service will be accepted by the pseudo-users. Using simulation software (e.g., AnyLogic), the behavioral patterns and reactions of the pseudo-users are modeled.

[1591] 3. The server collects the simulation results data, which are stored in a database and used for subsequent analysis.

[1592] 4. The server inputs the collected data into a generative AI model (e.g., GPT-4) and analyzes the data. The generative AI model uses the simulation results to identify areas for improvement, such as pricing for the new business service, delivery method, and selection of target users.

[1593] 5. The server sends the analysis results output from the generative AI model to the terminal.

[1594] 6. The terminal displays the analysis results to the user, who can then check areas for improvement in the new business service based on the displayed analysis results and make any necessary corrections.

[1595] Specific examples

[1596] For example, consider the case of launching an "online learning platform" as a new business service.

[1597] 1. The server launches AnyLogic and selects the simulation model of the "Online Learning Platform."

[1598] 2. The server inputs the pricing as "1,000 yen per month," the delivery method as "mobile app," and the target users as "working adults in their 20s and 30s."

[1599] 3. The server sets scenarios such as "study on weekday evenings" and "study all at once on weekends" as the behavioral patterns of the pseudo-users.

[1600] 4. The server executes the configured simulation model and simulates how a pseudo-user uses the platform.

[1601] 5. The server records the user's response to each scenario (e.g., study time, frequency of use, satisfaction, etc.).

[1602] 6. The server collects the simulation result data (e.g., study time, frequency of use, satisfaction level, etc.) and stores it in a database.

[1603] 7. The server converts the collected data into CSV format and formats it so that it can be input into the generative AI model.

[1604] 8. The server inputs the formatted data into GPT-4.

[1605] 9. From the simulation results, GPT-4 identifies areas for improvement, such as "the optimal pricing would be 1,000 yen per month," "the preferred delivery method is a mobile app," and "the optimal target users are working adults in their 20s and 30s."

[1606] 10. GPT-4 outputs the analysis results in text format.

[1607] 11. The server sends the analysis results output by GPT-4 to the terminal.

[1608] 12. The device displays analysis results to the user, such as "The optimal price setting is 1,000 yen per month," "The preferred delivery method is a mobile app," and "The optimal target users are working adults in their 20s and 30s."

[1609] 13. The user adjusts the pricing and delivery methods of the online learning platform based on the displayed analysis results.

[1610] Prompt Sentence Examples

[1611] "If you're launching an online learning platform as a new business service, use the simulation results to identify areas for improvement in pricing, delivery, and target users."

[1612] In this way, the system can confirm the effectiveness of the new business service based on the simulation results and identify specific areas for improvement and strengthening. The flow of the identification process in the third embodiment will be described with reference to FIG.

[1613] Step 1:

[1614] The server uses generative AI to generate a large number of pseudo-users.

[1615] Input: Data to mimic user behavior patterns and preferences

[1616] Data processing: Generative AI generates pseudo-user profiles based on input data.

[1617] Output: Pseudo user profile data

[1618] Specific operation: The server starts the generative AI model and inputs the user's behavioral patterns and preference data. The generative AI model generates a pseudo-user profile based on this data and stores it in a database.

[1619] Step 2:

[1620] The server executes a simulation to verify whether the new business service will be accepted by the pseudo-users.

[1621] Input: Pseudo user profile data, parameters of new business services (e.g., pricing, delivery method, target user attributes)

[1622] Data calculation: Simulation software evaluates the likelihood of new business services being accepted based on the behavioral patterns of simulated users.

[1623] Output: Simulation result data (e.g., study time, frequency of use, satisfaction, etc.)

[1624] Specific operation: The server starts the simulation software and inputs the profile data of the pseudo-user and the parameters of the new business service. The simulation software runs the simulation based on this data and generates the result data.

[1625] Step 3:

[1626] The server collects the result data of the simulation.

[1627] Input: Simulation result data

[1628] Data processing: The resulting data is stored in a database and formatted so that it can be input into the generative AI model.

[1629] Output: Formatted data

[1630] Specific operation: The server retrieves the result data from the simulation software, stores it in a database, and then converts it into CSV format, which is a format that can be input into the generative AI model.

[1631] Step 4:

[1632] The server inputs the collected data into a generative AI model and analyzes the data.

[1633] Input: Formatted data

[1634] Data calculation: Generative AI models use simulation results to identify areas for improvement in new business services, such as pricing, delivery methods, and target user selection.

[1635] Output: Analysis results (e.g., areas for improvement and enhancement)

[1636] How it works: The server inputs the formatted data into the generative AI model, which analyzes the data and identifies areas for improvement and strengthening. The analysis results are output in text format.

[1637] Step 5:

[1638] The server transmits the analysis results output from the generative AI model to the terminal.

[1639] Input: Analysis results

[1640] Data processing: Converting the analysis results into a format that is easy for users to understand.

[1641] Output: Feedback data

[1642] Specific operation: The server receives the analysis results output by the generative AI model, converts them into a format that is easy for the user to understand, and then sends the feedback data to the device.

[1643] Step 6:

[1644] The terminal displays the analysis results to the user.

[1645] Input: Feedback data

[1646] Data processing: Converting feedback data into a display format.

[1647] Output: Displayed analysis results

[1648] Specific operation: The terminal receives the feedback data from the server and displays it to the user. The user can then check the areas for improvement in the new business service based on the displayed analysis results and make any necessary corrections.

[1649] (Application example 3)

[1650] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[1651] When developing a new business service, it is extremely important to evaluate in advance whether the service will be accepted by the market. However, it is difficult to predict the actual market reaction, and there is a lack of concrete methods for optimizing elements such as pricing, delivery methods, and target user selection. For this reason, there is a need for a system that can perform simulations in advance, confirm the effectiveness of new business services based on the reactions of simulated users, and identify areas for improvement.

[1652] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means. In this invention, the server includes means for generating a large number of pseudo users using generative AI, means for running a simulation to verify whether a new business service will be accepted by the pseudo users, means for confirming the effectiveness of the new business service based on the results of the simulation, means for identifying areas for improvement such as pricing, provision method, and selection of target users based on the results of the simulation, and means for proposing optimal pricing and provision method. This makes it possible to predict and optimize market acceptance of the new business service.

[1653] "Generative AI" is a system that uses artificial intelligence technology to generate new data from large amounts of data.

[1654] A "pseudo user" is a virtual user created by imitating the behavior and preferences of an actual user.

[1655] "Simulation" is a method of virtually recreating a real-world situation and analyzing the results.

[1656] "New business services" refer to products and services that are newly introduced to the market.

[1657] "Effectiveness" is an indicator of how effective something is for a specific purpose.

[1658] "Pricing" is the process of determining the selling price of a product or service.

[1659] "Delivery method" refers to the means or format by which a product or service is delivered to a user.

[1660] "Target users" are the user group that is expected to use a particular product or service.

[1661] "Improvements" are specific changes that will resolve current problems and improve the situation.

[1662] A "proposal" is the act of suggesting a solution or improvement to a specific problem.

[1663] The system for implementing this invention consists of three main elements: a server, a terminal, and a user. The following explains how each element functions.

[1664] server

[1665] The server uses generative AI to generate a large number of pseudo-users. These pseudo-users are generated based on data that mimics the behavioral patterns and preferences of actual users. These pseudo-users are used in simulations to verify whether new business services will be accepted.

[1666] The results of the simulation are analyzed on the server and used to confirm the effectiveness of the new business service. Specifically, the simulation results are used to identify areas for improvement such as pricing, delivery methods, and target user selection, and optimal pricing and delivery methods are proposed.

[1667] Terminal

[1668] The terminal provides an interface for users to access the system. Mobile devices such as smartphones and tablets are mainly used. The application on the terminal displays the simulation results and proposals sent from the server and helps users make decisions based on them.

[1669] User

[1670] Users can access the system via their devices and check the simulation results and proposals for new business services. Based on this information, users can identify areas for improvement in services and formulate optimal strategies before actually launching them on the market.

[1671] Hardware and software used

[1672] Hardware: Servers (with high-performance computing power), smartphones, tablets

[1673] Software: Python, Pandas (data processing library), NumPy (numerical calculation library), Scikit-learn (machine learning library)

[1674] Specific examples

[1675] For example, consider the case of simulating pricing for a new product, a "smartwatch," and identifying the optimal price range. The server uses generative AI to generate pseudo-users and run the simulation. Based on the simulation results, it identifies areas for improvement in pricing and delivery methods and proposes the optimal price range.

[1676] Prompt Sentence Examples

[1677] "Please simulate pricing for a new product, a smartwatch, and identify the optimal price point. Also, please analyze the acceptability of a subscription model versus a one-time purchase model."

[1678] In this way, a system is realized in which servers, terminals, and users work together to predict and optimize market acceptance of new business services.

[1679] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[1680] Step 1:

[1681] The server uses generative AI to generate a large number of pseudo-users. As input, it uses data to imitate user behavior patterns and preferences. As data processing, it creates pseudo-user profiles based on this data. The output is a list of the generated pseudo-users.

[1682] Step 2:

[1683] The server uses the generated pseudo users to run a simulation of the new business service. As input, it uses a list of pseudo users and information about the new business service. As data calculation, it simulates the reactions of the pseudo users and determines whether each pseudo user will accept the service. The output is data of the simulation results.

[1684] Step 3:

[1685] The server analyzes the simulation results and confirms the effectiveness of the new business service. The simulation result data is used as input. The data is processed by statistically analyzing the simulation results and extracting the service acceptance rate and response patterns. The output is the evaluation result of the effectiveness.

[1686] Step 4:

[1687] The server identifies areas for improvement based on the simulation results, such as pricing, delivery methods, and target user selection. It uses the simulation results and effectiveness evaluation results as input. It analyzes the impact of each element as data calculation and extracts areas for improvement for optimization. The output is a list of areas for improvement.

[1688] Step 5:

[1689] The server proposes optimal pricing and delivery methods based on the identified improvements. As input, it uses a list of improvements. As data processing, it generates specific proposals based on the improvements. The output is a proposal for optimal pricing and delivery methods.

[1690] Step 6:

[1691] The terminal displays the simulation results and suggestions sent from the server. It receives data from the server as input. It visually displays the results and suggestions through a user interface. The output is a display that can be confirmed by the user.

[1692] Step 7:

[1693] The user accesses the system through a terminal and checks the simulation results and proposals. The information displayed on the terminal is used as input. The specific action of the user is to make a decision based on the displayed information. The output is the user's decision.

[1694] 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.

[1695] "Example 1"

[1696] In one embodiment of the present invention, the generative AI generates a pseudo-user using an emotion engine that recognizes the user's emotions. The emotion engine understands the user's emotional state and adjusts the pseudo-user's behavioral patterns based on that understanding. For example, if the user expresses joy, the emotion engine causes the pseudo-user to mimic the joyful behavioral patterns. This allows the pseudo-user to more accurately reflect the actual user's emotional state.

[1697] "Example 2"

[1698] In another embodiment of the present invention, the results of the simulation are used to analyze how the new business service affects the user's emotional state, for example, whether the new business service induces a feeling of pleasure or discomfort in the user, and the analysis results are used to identify areas for improvement or enhancement of the new business service.

[1699] "Example 3"

[1700] In one embodiment of the present invention, the generative AI generates a pseudo-user using an emotion engine that recognizes the user's emotions. The emotion engine understands the user's emotional state and adjusts the pseudo-user's behavioral patterns based on that understanding. For example, if the user expresses joy, the emotion engine causes the pseudo-user to mimic the joyful behavioral patterns. This allows the pseudo-user to more accurately reflect the actual user's emotional state.

[1701] The processing flow of each embodiment will be described below.

[1702] "Example 1"

[1703] Step 1: The generative AI activates an emotion engine that recognizes the user's emotions.

[1704] Step 2: The emotion engine understands the user's emotional state.

[1705] Step 3: The emotion engine adjusts the behavioral patterns of the simulated user based on the understood emotional state. For example, if the user shows an emotion of joy, the emotion engine makes the simulated user imitate the behavioral patterns of joy.

[1706] "Example 2"

[1707] Step 1: When the new business service is launched, the simulation begins.

[1708] Step 2: Based on the simulation results, analyze how the new business service affects the user's emotional state.

[1709] Step 3: The analysis results are used to identify areas for improvement or enhancement of the new business service. For example, we analyze whether the new business service evokes feelings of delight or discomfort in users.

[1710] Example 1

[1711] Next, a description will be given of Example 1 of Form 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."

[1712] Conventional methods for evaluating new business services have difficulty accurately imitating the behavior and emotions of actual users, making it difficult to accurately evaluate the effectiveness of the service. Furthermore, there was a lack of means to generate pseudo-users that reflected users' behavioral patterns and preferences, making the simulation results unreliable. Furthermore, because the behavioral simulation of pseudo-users did not take into account emotional states, it was not possible to accurately identify areas for improvement or strengthening of the service.

[1713] 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.

[1714] In this invention, the server includes means for generating a large number of pseudo-users using generative AI, means for running a simulation to verify whether a new business service will be accepted by the pseudo-users, means for verifying the effectiveness of the new business service based on the results of the simulation, means for collecting past user behavior data and user profile data, means for preprocessing the collected data, means for training a generative AI model using the preprocessed data, means for recognizing the emotional state of a user using an emotion engine and adjusting the behavior pattern of the pseudo-user, and means for performing behavioral simulations using the generated pseudo-users. This makes it possible to generate pseudo-users that accurately reflect the behavior patterns and emotional state of users, and to evaluate the effectiveness of the new business service with high accuracy.

[1715] "Generative AI" is AI that has the ability to learn data and generate new data and information based on that data.

[1716] A "pseudo user" is a virtual user created by imitating the behavioral patterns and preferences of an actual user.

[1717] A "start-up service" is a newly developed product or service that needs to be evaluated for effectiveness before being introduced to the market.

[1718] "Simulation" is a process of virtually testing how a new business service will be accepted by pseudo users.

[1719] "Effectiveness" is an indicator that shows the likelihood that a new business service will be successful in the market and the degree to which it will be accepted by users.

[1720] "User behavior data" is a record of a user's past behavior, including purchase history and browsing history.

[1721] "User profile data" refers to user attribute information, including age, gender, region, and interests.

[1722] "Preprocessing" is the process of converting collected data into a form suitable for analysis and model training, and includes cleaning and normalizing the data.

[1723] A "generative artificial intelligence model" is a part of generative artificial intelligence, and is a model for generating new data and information based on training data.

[1724] An "emotion engine" is a technology that recognizes a user's emotional state and adjusts behavioral patterns based on that emotion.

[1725] "Behavioral simulation" is a process in which the behavior of users is reproduced in a virtual environment using the generated pseudo-users, and new business services are evaluated.

[1726] The present invention relates to a system for generating a large number of pseudo-users using generative artificial intelligence and evaluating the effectiveness of new business services. Specific embodiments of this system will be described below.

[1727] The server first collects historical user behavior data and user profile data. This data collection involves running queries from a database or retrieving data from an external API. For example, the server runs an SQL query to retrieve purchase history and browsing history from a database, and calls an external API to collect social media user profile data.

[1728] The server then preprocesses the collected data, which includes data cleaning, normalization, and feature engineering. Specifically, the server uses Python's Pandas library to manipulate data frames, impute missing values, remove outliers, normalize numerical data, and one-hot encode categorical data.

[1729] Using the preprocessed data, the server trains a generative artificial intelligence model. This training uses a deep learning framework (e.g., TensorFlow, PyTorch). The server trains the neural network using TensorFlow's Model.fit method, inputting the training data to train the model. The training process aims to minimize the loss function and improve accuracy.

[1730] Furthermore, the server recognizes the user's emotional state using an emotion engine (e.g., IBM Watson Tone Analyzer) that extracts emotions from the user's text data and calculates an emotion score, e.g., from the user's reviews and comments.

[1731] The server generates a pseudo-user by combining the output of the generative AI model with the emotion score of the emotion engine. The generated pseudo-user mimics the behavioral patterns and preferences of real users and has behavioral patterns that reflect their emotional state.

[1732] Finally, the server performs a behavioral simulation using the generated pseudo-user. This verifies whether the new business service will be accepted by the pseudo-user and confirms its effectiveness based on the results of the simulation. Specifically, it simulates what products the pseudo-user will view and purchase on an online shopping site. For example, the following prompt sentence is input into the generative AI model:

[1733] Example prompt: "Predict what products a woman in her 30s, living in Tokyo, interested in fashion, would view based on her recent purchase history."

[1734] Based on this prompt, the generative AI model generates a pseudo-user's behavioral pattern and imitates user behavior on an online shopping site.

[1735] The above is a specific embodiment for carrying out the present invention.

[1736] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1737] Step 1:

[1738] The server collects historical user behavior data and user profile data.

[1739] Input: Data from a database or external API

[1740] What it does: The server executes SQL queries to retrieve purchase and browsing history from a database, and calls external APIs to collect social media user profile data.

[1741] Output: Collected user behavior and profile data

[1742] Step 2:

[1743] The server pre-processes the collected data.

[1744] Input: Collected user behavior and profile data

[1745] What it does: The server uses Python's Pandas library to manipulate data frames, impute missing values, remove outliers, normalize numerical data, and one-hot ...

Claims

1. A means for generating pseudo-users using data to mimic user behavior patterns and preferences; A means for executing a simulation to verify whether a new business service is acceptable to the pseudo user by inputting a specific prompt sentence into a generating AI model; A verification means for verifying the validity of the new business service by inputting the results of the simulation into a generating AI model; a means for recognizing an emotional state of the pseudo-user using an emotion engine, and adjusting a behavior pattern of the pseudo-user based on the recognized emotional state; A means for collecting behavioral data of the pseudo user by simulating the adjusted behavioral pattern of the pseudo user; A means for analyzing the collected behavioral data using a clustering algorithm; means for outputting the results of said analysis as a report; A system including:

2. A means for generating a rating and a review of a product based on a behavioral pattern of the pseudo user; means for storing the generated reviews in a database; The system of claim 1 , further comprising:

3. 2. The system of claim 1, further comprising means for identifying improvements and enhancements to the new business service by inputting specific prompt sentences into the generative AI model.

Images