System

Abstract

A system is provided.SOLUTION: A system comprising: means for generating virtual reviewers; means for assigning review tasks to the generated virtual reviewers; means for collecting review results from the virtual reviewers; and means for analyzing and providing the collected review results to a user.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] Traditional quality assurance (QA) processes in software development are time-consuming and costly due to their manual labor and high dependency on human resources. Furthermore, it is difficult to quickly obtain feedback from users specific to the target demographic, making rapid improvements difficult. In response to these challenges, we aim to realize a new development style, DevOpsQA, and to effectively resolve these issues. [Means for solving the problem]

[0005] The present invention provides the following means.

[0006] 1. A means of generating virtual reviewers.

[0007] 2. A means of assigning review tasks to the generated virtual reviewers.

[0008] 3. A means of collecting review results from virtual reviewers.

[0009] 4. A means of analyzing the collected review results and providing them to users.

[0010] Furthermore, by providing a means for receiving attribute data of the target demographic based on user input and generating a virtual reviewer that matches the attribute data using generative artificial intelligence, evaluations tailored to the target demographic can be made. Also, by providing a means for uploading the application to be reviewed to a server, reviews can be carried out quickly.

[0011] The virtual reviewer uses the application under review and evaluates it according to specified criteria, generating detailed review results. The server then analyzes and organizes the review results, extracting and providing meaningful insights to users. This process shortens the application development cycle and enables efficient quality control.

[0012] Finally, by providing an interface that allows users to plan improvements to their applications based on the review results they obtain, rapid and effective improvements are possible, reducing the time and cost required for traditional QA processes and enabling the rapid provision of highly accurate reviews tailored to the target audience.

[0013] A "virtual reviewer" is someone who is generated by generative artificial intelligence based on the attribute data of the target demographic and plays the role of a virtually set user.

[0014] "Generative AI" is a system that uses artificial intelligence technology to generate new data or entities based on input data.

[0015] A "review task" is a set of instructions assigned to a hypothetical reviewer to evaluate and review a particular application or system.

[0016] "Review results" are evaluation and feedback information generated by a virtual reviewer after performing a review task.

[0017] "Target demographic attribute data" is data based on specific aspects specified by the user (e.g., age, gender, interests, frequency of use).

[0018] The "server" is a computer system that receives input data from users, generates virtual reviewers using generative artificial intelligence, and collects and analyzes review results.

[0019] A "terminal" is a device that provides an interface for a user to connect to a server and input demographic data and upload applications to be reviewed.

[0020] A "user" is a person who provides an application to be reviewed and requests that it be reviewed by a virtual reviewer.

[0021] "Data for Analysis" means information collected by the Server and organized to extract meaningful insights through analysis of review results.

[0022] "Interface" refers to the operation screen and input method that users use to interact with the system.

[0023] "Improvement measures" are specific measures or proposals based on the review results to improve the functionality and usability of the application. [Brief explanation of the drawings]

[0024] [Figure 1]1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0025] 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.

[0026] First, the terms used in the following description will be explained.

[0027] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0028] 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.

[0029] 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.

[0030] 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.

[0031] 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."

[0032] [First embodiment]

[0033] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0034] 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.

[0035] 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).

[0036] 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.

[0037] 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.

[0038] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0039] 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.

[0040] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0041] 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.

[0042] 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.

[0043] 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.

[0044] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0045] To implement the present invention, a system is used that includes procedures for generating virtual reviewers using generative artificial intelligence, assigning review tasks, and collecting and analyzing review results. The operation of this system is specifically described below.

[0046] System configuration

[0047] 1. User input and application upload

[0048] The user uses the device interface to input demographic data (e.g., age, gender, interests, and frequency of use) about the target demographic, and uploads the application to be reviewed to the server via the device, allowing the user to send a review request to the server.

[0049] 2. Creating a Virtual Reviewer

[0050] The server receives target demographic attribute data sent by the user and uses generative artificial intelligence to generate a virtual reviewer based on that data. The generated virtual reviewer has the pre-set target demographic attributes and evaluates the application from a perspective based on those attributes.

[0051] 3. Assigning review tasks and conducting reviews

[0052] The server assigns the task of reviewing the uploaded application to the generated virtual reviewer, who then uses the application to perform this task and provides a detailed evaluation according to the specified criteria.

[0053] 4. Collection and analysis of review results

[0054] The server collects the review results generated by the virtual reviewers, analyzes them further, and extracts meaningful insights for users (e.g., UI improvements, performance bottlenecks). The analysis results are displayed in the form of a report on the platform and provided to users.

[0055] 5. Formulation of improvement measures

[0056] Users can use their devices to check the review results received from the server, and based on the feedback they receive, they can formulate improvements to the application and incorporate them into the next development cycle, thereby continuously improving the quality of the application.

[0057] Specific examples

[0058] For example, consider the development of an e-commerce application with new features. The development team specifies the target demographic as "women in their 20s who purchase frequently" and generates virtual reviewers. The virtual reviewers use the application from a perspective based on the specified target demographic and provide detailed feedback on features such as ease of use and UI design.

[0059] After the review is complete, the server collects the review results and generates an analysis report. This report includes major issues and improvement suggestions. The user (development team) refers to this report and formulates improvement measures for new functions. This enables high-quality application releases in a short period of time.

[0060] This invention significantly reduces the time and cost required for conventional QA processes and enables the rapid acquisition of highly accurate reviews tailored to the target demographic.

[0061] The processing flow will be explained below.

[0062] Step 1:

[0063] The user opens the interface on their device, enters the demographic data of the target demographic (e.g., age, gender, interests, frequency of use), and clicks the send button once the input is complete.

[0064] Step 2:

[0065] The device temporarily stores the target demographic attribute data entered by the user and prepares to send it to the server. At the same time, the user selects the application to be reviewed (e.g., APK file) and clicks the upload button to begin uploading the application.

[0066] Step 3:

[0067] The server receives the target demographic attribute data and uploaded application files sent from the device. The received data is temporarily stored and organized as input data for the generative AI.

[0068] Step 4:

[0069] The server uses generative artificial intelligence based on the received target demographic attribute data to generate a virtual reviewer that matches the specified attributes. This virtual reviewer is a virtual user with the specific attributes entered by the user.

[0070] Step 5:

[0071] The server assigns a review task for the application to be reviewed to the generated virtual reviewer, which includes viewpoints and evaluation criteria specified by the user.

[0072] Step 6:

[0073] The virtual reviewer uses the application under review in a simulated environment and evaluates it according to specified criteria, such as the ease of use of the UI, the intuitiveness of the features, and performance.

[0074] Step 7:

[0075] After completing the review task, the virtual reviewers send the generated evaluation results and feedback to the server, which includes detailed evaluation comments and suggestions for improvement.

[0076] Step 8:

[0077] The server collects the review results received from the virtual reviewers and organizes them as data for analysis. The server analyzes the collected data and extracts insights that are meaningful to the user.

[0078] Step 9:

[0079] The server generates the analysis results in the form of a report and displays it on the platform, and sends a notification to the user informing them that the review results are available for access.

[0080] Step 10:

[0081] Users access the server via their terminals and check the review results displayed on the platform. Users can view detailed reports and evaluate the feedback of each virtual reviewer.

[0082] Step 11:

[0083] The user will then formulate improvements to the application based on the review results obtained, and will incorporate this data into the next development cycle to create plans for new features and bug fixes.

[0084] In this way, the entire processing flow works together efficiently, allowing users to quickly obtain highly accurate reviews specific to their target demographic and improve the quality of their applications.

[0085] Example 1

[0086] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0087] The conventional evaluation process had problems in that it was difficult to quickly obtain highly accurate reviews specific to a target demographic, and the reviews were time-consuming and costly. In particular, in order to collect feedback from different target demographics, it was necessary to arrange for multiple reviewers to conduct the evaluations, which resulted in a significant burden of cost and time.

[0088] 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.

[0089] In this invention, the server includes means for receiving target demographic attribute data and the program to be reviewed based on user input, means for using a generative artificial intelligence model to generate virtual reviewers that match the received target demographic attribute data, means for assigning review tasks to the generated virtual reviewers and having them perform detailed evaluations of the program, means for collecting review results from the virtual reviewers and analyzing the results, and means for providing the analysis results to the user. This makes it possible to quickly obtain highly accurate reviews tailored to the target demographic, significantly reducing the time and cost required for conventional evaluation processes.

[0090] "User input" refers to a person or entity using the system providing demographic data and programs to be reviewed to the system.

[0091] "Target demographic attribute data" refers to information that indicates the characteristics of the target user demographic when writing a review (e.g., age, gender, interests, frequency of use).

[0092] "Reviewed Program" means software or an application that is subject to evaluation and review.

[0093] A "virtual reviewer" is a reviewer simulated based on specific attribute data using a generative artificial intelligence model.

[0094] A "generative artificial intelligence model" is an artificial intelligence algorithm that simulates human opinions and behaviors based on specific prompts.

[0095] A "review task" is a task that includes specific evaluation activities and checkpoints assigned to a virtual reviewer.

[0096] "Review results" are the feedback and evaluation scores obtained when a virtual reviewer performs a review task.

[0097] "Analysis results" are the results of further analyzing the collected review results and extracting insights that are useful to users (e.g., areas for improvement in the UI, performance bottlenecks).

[0098] "Display in report format" means that the analysis results are provided in a document or table format that is easy for the user to understand.

[0099] A "prompt sentence" is an input sentence given to a generative artificial intelligence model, and is a sentence that includes instructions for evaluation based on specific attributes and evaluation criteria.

[0100] MODE FOR CARRYING OUT THE INVENTION

[0101] To implement the present invention, a system is used in which a user inputs attribute data of the target demographic and the program to be reviewed, and a virtual reviewer is generated based on the input data using generative artificial intelligence. A specific implementation method of this system is described below.

[0102] User input and application upload

[0103] Users use their own devices to input demographic data (e.g., age, gender, interests, and frequency of use) about their target demographic. This data is important information for reflecting the characteristics of the target demographic. Users also upload the program (e.g., application) to be reviewed to the server via their devices. This sends a review request to the server in response to the user's request.

[0104] Virtual reviewer generation

[0105] The server receives the target demographic attribute data sent by the user and generates a virtual reviewer based on that data. The virtual reviewer is generated using a generative artificial intelligence model (e.g., GPT-3). By inputting the target demographic profile into this model, a virtual reviewer with specific attributes is generated.

[0106] Assigning review tasks and conducting reviews

[0107] The server assigns the uploaded program review task to the generated virtual reviewer, who then performs a detailed evaluation according to specified criteria (e.g., ease of use of functions, UI design, performance). The virtual reviewer operates the program in an AI simulation environment and performs the evaluation through simulated user operations.

[0108] Collection and analysis of review results

[0109] The server collects and analyzes the reviews generated by the virtual reviewers. The review results include specific feedback and rating scores that users may find interesting. The server further analyzes this data to extract useful insights for users. The analysis results are displayed on the platform in the form of a report.

[0110] Formulating improvement measures

[0111] Users can use their devices to check the review results and analysis reports provided by the server, and can formulate improvements to the application based on this feedback and incorporate them into the next development cycle, thereby continuously improving the quality of the application.

[0112] Specific examples

[0113] For example, consider the development of an e-commerce application with new features. The development team (users) specify the target demographic as "women in their 20s with high purchasing frequency" and generate a virtual reviewer. The virtual reviewer will use the application from a perspective based on the specified attributes and provide detailed feedback on features such as ease of use and UI design.

[0114] Example prompts used to generate virtual reviewers:

[0115] "Please rate in detail the ease of use of the new features for women in their 20s."

[0116] "Please review in detail how users who frequently use SNS experience the UI."

[0117] After the review is completed, the server collects the review results and generates an analysis report, which includes major issues and improvement suggestions. The development team (users) can refer to this report and quickly develop improvement plans for new features.

[0118] This invention significantly reduces the time and cost required for the conventional evaluation process, making it possible to quickly obtain highly accurate reviews specific to the target demographic.

[0119] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0120] Step 1: User Input

[0121] The user uses the device interface to input attribute data of the target demographic (e.g., age, gender, interests, frequency of use). Specifically, the user uses forms and check boxes to input data such as "20s," "female," and "frequent SNS user." The input data is organized within the device and sent to the server.

[0122] Input: Target demographic data (e.g., age, gender, interests, frequency of use)

[0123] Output: Attribute data sent to the server

[0124] Step 2: Upload your application

[0125] The user uploads the application to be reviewed to the server via their device. Specifically, the user selects the specified application file (e.g., apk file or ipa file) from their device and clicks the "Upload" button. The file is then transferred from the device to the server.

[0126] Input: Application file to be reviewed

[0127] Output: Application files uploaded to the server

[0128] Step 3: Receive target data

[0129] The server receives the target demographic attribute data sent by the user. The received data is stored in a database and used as a target demographic profile. Data integrity checks are also performed at this stage.

[0130] Input: Target demographic data submitted by the user

[0131] Output: Attribute data stored in a database

[0132] Step 4: Generate virtual reviewers

[0133] The server generates virtual reviewers using a generative artificial intelligence model (e.g., GPT-3). By inputting the target demographic profile into a prompt statement (e.g., "Please generate a virtual reviewer for a woman in her 20s"), the AI ​​generates a virtual reviewer with specific attributes.

[0134] Input: Target demographic profile

[0135] Output: Generated virtual reviewer

[0136] Step 5: Assign review tasks

[0137] The server assigns review tasks for uploaded applications to the generated virtual reviewers, including evaluation criteria (e.g., usability, design, performance).

[0138] Input: Generated virtual reviewer, application to be reviewed

[0139] Output: Tasks assigned to virtual reviewers

[0140] Step 6: Conducting the review

[0141] The virtual reviewer interacts with the application in a simulated environment to perform the assigned task: interacting with the UI, testing each feature, and recording their evaluation.

[0142] Input: Review task, Simulation environment

[0143] Output: Review results from the virtual reviewer

[0144] Step 7: Collect review results

[0145] The server collects the review results generated by the virtual reviewers, which are then stored in a database and treated as a dataset for later analysis.

[0146] Input: Review results from a hypothetical reviewer

[0147] Output: Review results stored in a database

[0148] Step 8: Analyze the results

[0149] The server analyzes the collected reviews and extracts useful insights for users. Specifically, it uses data analysis algorithms to identify key issues and improvement suggestions.

[0150] Input: Collected review results

[0151] Output: Analysis results (insights and improvement suggestions)

[0152] Step 9: View the results

[0153] The server displays the analysis results in the form of a report on the platform, which the user can view via their terminal.

[0154] Input: Analysis results

[0155] Output: Analysis results displayed in report format

[0156] Step 10: Developing improvement measures

[0157] The user formulates improvement measures for the application based on the review results and analysis reports provided by the server, and reflects these measures in the next development cycle to improve the quality of the application.

[0158] Input: Review results and analysis report

[0159] Output: Improvements

[0160] (Application example 1)

[0161] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0162] Conventional review systems often struggle to quickly obtain detailed evaluations tailored to a target demographic, resulting in problems with the quality and accuracy of reviews. Furthermore, when using human reviewers, the review tasks are time-consuming and costly, making it difficult to obtain efficient feedback. This often prevents users from quickly obtaining suggestions for improvements or product modifications, delaying the product development cycle. The present invention aims to solve these problems.

[0163] 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.

[0164] In this invention, the server includes means for generating virtual reviewers, means for assigning review tasks to the generated virtual reviewers, means for collecting review results from the virtual reviewers, means for analyzing the collected review results and providing them to the user, means for receiving attribute data of the target demographic based on user input and using the attribute data to send prompt sentences to the generative artificial intelligence to generate virtual reviewers, and means for performing detailed product evaluations based on the perspective of the target demographic and analyzing the results to provide insights to the user. This makes it possible to quickly obtain highly accurate reviews tailored to the target demographic.

[0165] A "virtual reviewer" is a virtual entity generated using generative artificial intelligence based on attribute data of a specific target demographic, and is a system that reviews applications and products from the perspective of an actual user.

[0166] A "review task" is an application or product evaluation task assigned to a virtual reviewer, providing detailed feedback based on specific criteria or attributes.

[0167] "Review results" refer to the evaluation information and feedback obtained by the virtual reviewers when they perform the review task, including specific comments about the product's usability, appearance, functionality, etc.

[0168] "Target demographic attribute data" is information used to define a specific target demographic, such as age, gender, interests, and frequency of use. Based on this data, generative AI generates virtual reviewers.

[0169] "Generative AI" is an AI that has the ability to generate virtual reviewers based on the attribute data of the target demographic entered by the user, and is a technology that receives prompt statements and generates an appropriate model.

[0170] A "prompt sentence" is an input sentence that specifies a specific target demographic and evaluation criteria to the generative artificial intelligence, which then generates a virtual reviewer.

[0171] A "detailed evaluation" is an evaluation of a product by a virtual reviewer according to specific criteria, including specific feedback on specific usability, design, performance, etc.

[0172] "Insights" are useful information and suggestions obtained by analyzing collected review results, and serve as guidelines for improving products and adding new features.

[0173] To implement the present invention, it is necessary to operate the system in the following procedure. This system is made up of a server, a terminal, and a user, and the specific procedure is as follows.

[0174] 1. User input and product data upload

[0175] Users use the device interface to input demographic data (e.g., age, gender, interests, frequency of use) of their target demographic and upload product data to be reviewed. This input data is sent to the server and used to generate virtual reviewers.

[0176] 2. Creating a Virtual Reviewer

[0177] The server uses generative artificial intelligence (specifically, OpenAI's GPT-4 model) to generate a virtual reviewer based on the target demographic attribute data it receives. The generative artificial intelligence sends prompts using the input attribute data, and generates a virtual reviewer based on the results.

[0178] Example prompt sentence:

[0179] "Generate a virtual reviewer with the following attributes: age: 20-29, gender: female, purchase frequency: high"

[0180] 3. Assigning and completing review tasks

[0181] The server assigns review tasks to the virtual reviewers, who then perform the tasks and provide detailed evaluations according to specific criteria (e.g., usability, appearance, cost-effectiveness).

[0182] 4. Collection and analysis of review results

[0183] The server collects and analyzes the reviews generated by the virtual reviewers. This analysis uses natural language processing technology based on generative artificial intelligence to extract key issues and improvement suggestions. These results are provided to users in the form of useful insights (e.g., areas for product improvement, the need for new features).

[0184] 5. Formulation of improvement measures

[0185] Users can use their devices to check the review results obtained from the server and formulate measures to improve the product. This feedback is reflected in the next development cycle, contributing to improving the quality of the product.

[0186] Examples:

[0187] For example, consider the case of reviewing product data for a fashion dress aimed at women in their 20s. The user inputs target demographic attribute data, such as "women in their 20s, frequent buyers," and uploads the product data for the fashion dress to the server. The server then generates a virtual reviewer based on this data and has that reviewer evaluate the dress. The review results include the quality of the design, functionality, cost-effectiveness, etc., and the user can use this information to formulate improvements to the product.

[0188] This system allows us to quickly obtain highly accurate reviews tailored to the target demographic, improving the quality and accuracy of reviews while shortening the development cycle. By using these methods, users can effectively evaluate and improve products.

[0189] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0190] Step 1:

[0191] The user uses the device interface to input demographic data for the target demographic and upload product data for review. The input demographic data includes age, gender, interests, and frequency of use, while the product data includes name, price, and description. This data is then sent to the server.

[0192] Input: Target demographic data (e.g., age, gender, interests, frequency of use), product data (e.g., name, price, description)

[0193] Output: Target demographic attribute data and product data sent to the server

[0194] Step 2:

[0195] The server then sends a prompt to the generative AI based on the target demographic attribute data it receives, generating a virtual reviewer. The prompt includes the target demographic attribute data, and a virtual reviewer is generated based on that.

[0196] Input: target demographic attribute data, prompt text

[0197] Output: Generated virtual reviewer

[0198] Step 3:

[0199] The server assigns review tasks to the virtual reviewers, who then perform detailed evaluations based on the uploaded product data and specified criteria (e.g., usability, appearance, and price-performance).

[0200] Input: Virtual reviewer, product data, rating criteria

[0201] Output: Detailed evaluation results

[0202] Step 4:

[0203] The server collects the evaluation results generated by the virtual reviewers and analyzes the data, using natural language processing techniques (e.g., generative artificial intelligence) to extract key issues and improvement suggestions.

[0204] Input: Review results

[0205] Output: Analyzed insights (e.g., key issues, improvement suggestions)

[0206] Step 5:

[0207] The user uses a terminal to check the review results obtained from the server and formulates measures to improve the product. The user then reflects the feedback in the next development cycle.

[0208] Input: Review results (analysis report)

[0209] Output: Improved product development plan

[0210] 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.

[0211] To implement the present invention, a system is used that includes procedures for generating virtual reviewers using generative artificial intelligence, review evaluation using an emotion engine, and collection and analysis of review results. The operation of this system is described in detail below.

[0212] System configuration

[0213] 1. User input and application upload

[0214] The user uses the device interface to input demographic data (e.g., age, gender, interests, and frequency of use) about the target demographic, and uploads the application to be reviewed to the server via the device, allowing the user to send a review request to the server.

[0215] 2. Creating a Virtual Reviewer

[0216] The server receives target demographic attribute data sent by the user and uses generative artificial intelligence to generate a virtual reviewer based on that data. The generated virtual reviewer has the pre-set target demographic attributes and evaluates the application from a perspective based on those attributes.

[0217] 3. Emotion Recognition by Emotion Engine

[0218] The server integrates an emotion engine when assigning review tasks for applications to the virtual reviewers. As the virtual reviewers use the applications, the emotion engine recognizes and records their emotional states in real time. This allows us to quantitatively understand the emotional state of the virtual reviewers.

[0219] 4. Assigning review tasks and conducting reviews

[0220] The server assigns a review task for the application to the generated virtual reviewer. The virtual reviewer uses the application to perform this task and provides a detailed evaluation according to the specified criteria. During the evaluation process, the emotion engine monitors the emotional state of the virtual reviewer and generates emotion analysis data.

[0221] 5. Collection and analysis of review results

[0222] The server collects the review results generated by the virtual reviewers and sentiment analysis data from the sentiment engine, and further analyzes this data to extract meaningful insights for users (e.g., emotional responses to UI usability, sentiment fluctuations for specific features).

[0223] 6. Providing insights and suggesting improvements

[0224] The server generates the analysis results in the form of a report and displays it on the platform. The report includes information that integrates review results and sentiment analysis, providing more meaningful insights for users. Users can review the report and formulate improvements for their applications.

[0225] 7. Formulation of improvement measures

[0226] Users can use their devices to check the reports received from the server and evaluate the detailed review results and sentiment analysis data. This allows users to make specific improvements based on their emotional state (e.g., changing the interface because a specific feature causes discomfort). Based on the feedback obtained, it is possible to plan new features and bug fixes for the next development cycle.

[0227] Specific examples

[0228] For example, when developing a new e-commerce application, a user may specify a target demographic of "women in their 20s who purchase frequently." Virtual reviewers are generated based on these attributes and work with an emotion engine to conduct reviews. Review results include not only evaluations of product display speed and UI, but also associated emotional fluctuations (e.g., stress levels when a product loads slowly).

[0229] After the review is complete, the server collects and analyzes all data and provides a detailed report to the user, who can then refer to the report and develop specific improvement measures that take into account the user's emotional state and incorporate them into the next release.

[0230] In this way, the system of the present invention makes it possible to quickly obtain highly accurate reviews and sentiment analysis specific to the target demographic, thereby improving the quality of applications.

[0231] The processing flow will be explained below.

[0232] Step 1:

[0233] The user opens the interface on their device, enters the demographic data of the target demographic (e.g., age, gender, interests, frequency of use), and clicks the send button once the input is complete.

[0234] Step 2:

[0235] The device temporarily stores the target demographic attribute data entered by the user and prepares to send it to the server.The user also selects the application to be reviewed via the device and uploads the application by clicking the upload button.

[0236] Step 3:

[0237] The server receives the target demographic attribute data and uploaded application files sent from the device, temporarily stores the received data, and organizes it as input data for the generative AI.

[0238] Step 4:

[0239] The server uses generative artificial intelligence to generate a virtual reviewer based on the received target demographic attribute data. The generated virtual reviewer has the specified attributes and evaluates the application from a perspective based on those attributes.

[0240] Step 5:

[0241] The server integrates an emotion engine with the generated virtual reviewer, which recognizes and records the emotional state of the virtual reviewer in real time.

[0242] Step 6:

[0243] The server assigns review tasks for uploaded applications to virtual reviewers, with the review tasks including user-specified viewpoints and evaluation criteria.

[0244] Step 7:

[0245] The virtual reviewer uses the application under review and provides a detailed evaluation according to the specified criteria. The emotion engine monitors the virtual reviewer's emotional state (e.g., stress level, satisfaction) while using the application and generates emotion analysis data.

[0246] Step 8:

[0247] The server collects the review results generated by the virtual reviewers and the sentiment analysis data generated by the sentiment engine. This collected data is temporarily stored and then used for the next analysis step.

[0248] Step 9:

[0249] The server analyzes the collected review results and sentiment analysis data to extract meaningful insights for users (e.g., emotional reactions to UI usability, sentiment fluctuations for specific features). The analysis results are generated in the form of a report and displayed on the platform.

[0250] Step 10:

[0251] The user uses a terminal to check the report retrieved from the server, which contains information that integrates review results and sentiment analysis.

[0252] Step 11:

[0253] Users can formulate improvements to their applications based on the review results and sentiment analysis data they obtain. Specifically, if a particular feature causes discomfort, they can consider specific measures such as changing the interface. The obtained feedback can be reflected in the next development cycle, allowing them to plan new features and bug fixes.

[0254] In this way, the system of the present invention can quickly obtain highly accurate reviews and sentiment analysis specific to the target demographic, effectively improving the quality of applications.

[0255] Example 2

[0256] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0257] Conventional review systems have difficulty evaluating each target demographic and providing detailed feedback, including emotional states. In particular, users lack the means to create virtual reviewers based on target attributes and collect and analyze their review results along with emotional data, making it difficult to gain insights needed to improve software under development.

[0258] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for receiving attribute data of the target demographic based on user input and generating a virtual reviewer that matches the attribute data, means for assigning a review task to the generated virtual reviewer, means for integrating an emotion engine and recognizing and recording the emotional state of the virtual reviewer in real time, means for collecting review results and emotion analysis data from the virtual reviewer, and means for analyzing the collected review results and emotion analysis data and providing them to the user. This enables highly accurate reviews and emotion analysis tailored to the target demographic.

[0259] "Target demographic attribute data" is information for defining a specific target user group, such as the user's age, gender, interests, and frequency of use.

[0260] A "virtual reviewer" is a fictitious reviewer with specific target demographic attributes, generated using a generative artificial intelligence model.

[0261] A "review task" is a specific instruction and task for evaluating and providing feedback on the software or application being reviewed.

[0262] An "emotion engine" is software or algorithms for recognizing and recording the emotional state of a virtual reviewer in real time.

[0263] "Sentiment Analysis Data" is information about the emotional state of a hypothetical reviewer recorded by the emotion engine, and is a record of emotions that fluctuate during the course of the review.

[0264] A "generative AI model" is an artificial intelligence model that uses natural language processing and machine learning techniques to generate new data and information (in this case, virtual reviewers).

[0265] A "prompt" is an instruction or input to a generative AI model to obtain a specific output.

[0266] "Review evaluation data" refers to evaluation information collected by virtual reviewers while performing a review task, and is detailed feedback on each function and interface of an application.

[0267] The present invention is a system that includes procedures for generating virtual reviewers using generative artificial intelligence, review evaluation using an emotion engine, and collection and analysis of review results. This system collects attribute data of the target demographic based on user input, generates virtual reviewers based on that data, and conducts reviews. In the process, it collects emotion data using an emotion engine, and finally analyzes this data and provides it to the user as a report. The details of this system are described below.

[0268] First, the user uses the device interface to input the target demographic's attribute data (e.g., age, gender, interests, frequency of use, etc.), then uploads the software to be reviewed to the server via the device and submits a review request.

[0269] The server receives the target demographic attribute data sent by the user and generates a virtual reviewer using a generative AI model (e.g., GPT-3 or BERT) based on that data. In this process, a prompt sentence is generated for the generative AI model and input into the model to generate a virtual reviewer.

[0270] The server then assigns review tasks to the virtual reviewers, and integrates an emotion engine (e.g., Microsoft Azure's emotion recognition API) to recognize and record the virtual reviewers' emotional states in real time.

[0271] The virtual reviewer uses the software under review to perform a detailed evaluation according to the specified criteria, during which the emotion engine monitors the reviewer's emotional state and generates sentiment analysis data.

[0272] The server collects the review results generated by the virtual reviewers and the sentiment analysis data generated by the sentiment engine, and organizes and analyzes this data using analytical tools (e.g., Python's Pandas and NumPy libraries) to extract meaningful insights for users.

[0273] Finally, the server generates a report of the analysis results and displays it on the platform. Users can then use their devices to review the report and formulate specific improvement measures based on the detailed review results and sentiment analysis data, which can then be incorporated into the next development cycle.

[0274] As a concrete example, consider the development of a new e-commerce application. The user inputs demographic data such as "women in their 20s who purchase frequently" as the target demographic, and generates a virtual reviewer. This virtual reviewer reviews the application together with an emotion engine. The review results include evaluations of the product display speed and UI, as well as emotional states such as stress felt when products are slow to load.

[0275] Example prompt sentence:

[0276] The target users are women in their 20s who purchase frequently. Please generate a virtual reviewer based on this profile and have them rate the following application.

[0277] In this way, the system of the present invention can quickly perform highly accurate reviews and sentiment analysis specific to the target demographic, allowing users to efficiently improve the quality of their applications.

[0278] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0279] Step 1: User Input and Application Upload

[0280] The user uses the device interface to input demographic data (age, gender, interests, frequency of use, etc.) of the target demographic. The input data is sent to the server in JSON format, which becomes the input.

[0281] The user selects the software to be reviewed (e.g., APK file) and clicks the upload button. The software file is uploaded to the server, which is the output.

[0282] Step 2: Receive demographic data for your target audience

[0283] The server receives the HTTP request, analyzes the target demographic attribute data sent by the user, and saves the received JSON data in an internal database. This becomes the input.

[0284] The attribute data is converted into a format suitable for generative AI models, which then becomes the output.

[0285] Step 3: Generate virtual reviewers

[0286] The server uses a generative AI model (e.g., GPT-3 or BERT) to generate a virtual reviewer. Specifically, it creates a prompt and inputs it into the generative AI model. The prompt is, "The target user is a woman in her 20s who frequently purchases. Please generate a virtual reviewer based on this profile and rate the following application." This is the input.

[0287] The generative AI model generates a virtual reviewer and returns that data to the server, which becomes the output.

[0288] Step 4: Assigning review tasks and integrating the sentiment engine

[0289] The server assigns review tasks for uploaded software to the virtual reviewers, which is the input.

[0290] The server integrates an emotion engine (e.g., Microsoft Azure's emotion recognition API) to recognize and record the emotional state of the virtual reviewer in real time. This is the specific operation. It configures the settings to call the emotion recognition API and sends an API request for each reviewer operation. It receives the response from the API and records the emotional data, which becomes the output.

[0291] Step 5: Performing review tasks

[0292] The virtual reviewers interact with the software according to the assigned review task and provide detailed evaluations based on the specified criteria, which serve as input.

[0293] During operation, the emotion engine monitors the emotional state and generates evaluation data and emotion analysis data, which are the output.

[0294] Step 6: Collect review results and sentiment analysis data

[0295] The server receives review rating data and emotional state data sent by virtual reviewers, which are the input.

[0296] This data is stored in a database and organized and analyzed using analytical tools (e.g., Python's Pandas or NumPy library). This is the specific operation, and the analyzed data is the output.

[0297] Step 7: Generate and serve reports

[0298] The server generates a report based on the analysis results. Specifically, it creates a detailed report by applying the analyzed data to a template. This is the input.

[0299] The generated report is displayed on the user interface and made accessible to the user - this is the output.

[0300] Step 8: Developing improvement measures

[0301] The user uses a terminal to check the report retrieved from the server, which becomes the input.

[0302] Users evaluate the detailed review results and sentiment analysis data and develop specific improvement measures. This is the output. Improvement measures may include, for example, changing the UI design or adding or removing features.

[0303] In this way, the user, server, and device work together at each processing step to achieve highly accurate reviews and sentiment analysis tailored to the target demographic.

[0304] (Application example 2)

[0305] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0306] Conventional review systems rely on user-provided feedback, making it difficult to accurately grasp emotional fluctuations and real-time emotional states. There was also a need for a method to quickly and accurately obtain reviews that are optimal for the target demographic.

[0307] 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 virtual reviewers, means for assigning review tasks to the generated virtual reviewers, means for collecting review results from the virtual reviewers, means for analyzing the collected review results and providing them to users, means for collecting and analyzing emotional data in real time, and means for monitoring and recording the emotional state of users from their gaze, facial expressions, etc. This makes it possible to quickly obtain highly accurate reviews and emotional analysis tailored to the target demographic and efficiently propose improvement measures for applications and products.

[0308] A "virtual reviewer" is a simulated subject for conducting a review, generated using generative artificial intelligence based on attribute data of a specific target demographic.

[0309] A "review task" is a series of tasks assigned to a virtual reviewer to evaluate a particular application or product and record the results.

[0310] "Emotional data" is data used to record and analyze changes in the user's emotional state in real time based on information such as their gaze, facial expressions, and voice.

[0311] "Means for collecting and analyzing emotional data in real time" refers to technology that continuously monitors the user's gaze, facial expressions, etc., and records and analyzes their emotional state on the spot.

[0312] "Generative AI" is an AI model that generates responses and evaluations for specific tasks based on input attribute data.

[0313] "Means for monitoring and recording emotional states from gaze, facial expressions, etc." refers to technology that detects a user's gaze and facial expressions through smart glasses or the like when the user is browsing an online shopping site, and collects and stores that emotional state in real time.

[0314] A "target demographic" is a group of users with specific attributes such as age, gender, interests, and frequency of use, and is the group that is analyzed to evaluate and improve products and services.

[0315] To implement the present invention, the following system configuration and processing procedures are used: Specifically, review evaluation using a virtual reviewer and an emotion engine, and review result collection and analysis procedures.

[0316] System configuration

[0317] 1. User input and application upload

[0318] Users use the device interface to input demographic data (age, gender, interests, frequency of use, etc.) about their target demographic, and also upload the application to be reviewed to the server via the device, allowing the user to send a review request to the server.

[0319] 2. Creating a Virtual Reviewer

[0320] The server receives target demographic attribute data sent by the user and uses generative artificial intelligence to generate a virtual reviewer based on that data. The generated virtual reviewer has the pre-set target demographic attributes and evaluates the application from a perspective based on those attributes.

[0321] 3. Emotion Recognition by Emotion Engine

[0322] The server integrates an emotion engine when assigning review tasks for applications to the virtual reviewers. As the virtual reviewers use the applications, the emotion engine recognizes and records their emotional states in real time. This allows us to quantitatively understand the emotional state of the virtual reviewers.

[0323] 4. Assigning review tasks and conducting reviews

[0324] The server assigns a review task for the application to the generated virtual reviewer. The virtual reviewer uses the application to perform this task and provides a detailed evaluation according to the specified criteria. During the evaluation process, the emotion engine monitors the emotional state of the virtual reviewer and generates emotion analysis data.

[0325] 5. Collection and analysis of review results

[0326] The server collects the review results generated by the virtual reviewers and sentiment analysis data from the sentiment engine, and further analyzes this data to extract meaningful insights for users (e.g., emotional responses to the usability of the user interface, sentiment fluctuations for specific features).

[0327] Specific examples

[0328] For example, consider a female user in her twenties searching for new clothes on an online shopping site. The user puts on smart glasses and launches a dedicated app. The emotion engine collects real-time emotional data from the user's gaze and facial expressions, recording their emotional state. At the same time, a virtual reviewer using generative artificial intelligence evaluates the same product page. As a result, the emotional data of the user and the virtual reviewer is sent to a server for analysis.

[0329] The analysis results are provided to users in the form of a report, which includes specific improvement measures based on the sentiment data. For example, the report may include insights such as "users feel frustrated when products take a long time to load" or "certain designs are particularly well-received."

[0330] Prompt Sentence Examples

[0331] The following prompt sentence is input into the generative artificial intelligence:

[0332] Generate virtual reviewers based on the following user attributes:

[0333] User Attributes:

[0334] Age: 20s

[0335] Gender: Female

[0336] Interests: Fashion, shopping

[0337] Frequency of use: High

[0338] We assign a hypothetical reviewer the task of rating a clothing product page on an online shopping website at the following URL:

[0339] URL: [example.com / fashion / clothing / item1234]

[0340] Output your impressions and emotional data.

[0341] This allows for quick and accurate target-specific reviews and sentiment analysis, which can be used to improve the quality of applications and products.

[0342] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0343] Step 1:

[0344] The user uses the device interface to input demographic data (age, gender, interests, frequency of use, etc.) of the target demographic. The application to be reviewed is then uploaded to the server via the device. This input data is sent to the server and used as the basis for generating virtual reviewers.

[0345] Step 2:

[0346] The server receives the target demographic attribute data sent by the user and generates a virtual reviewer based on that data using generative AI. Specifically, the received attribute data is input into the generative AI model as a prompt sentence, generating an appropriate virtual reviewer.

[0347] Step 3:

[0348] The server assigns a review task for the application to the virtual reviewer, who then evaluates the application according to the specified criteria. During this process, the virtual reviewer's evaluation is recorded.

[0349] Step 4:

[0350] The server integrates an emotion engine to collect real-time emotional data from users' gazes and facial expressions. When users browse online shopping sites through smart glasses, emotion data is recorded in real time using eye tracking and facial recognition technology. This data is sent to the server and integrated with the evaluation data from the virtual reviewers.

[0351] Step 5:

[0352] The server collects the review results generated by the virtual reviewers and sentiment analysis data from the emotion engine. This data includes the user's emotional fluctuations and eye movements, forming detailed user experience data. This allows for comprehensive review results.

[0353] Step 6:

[0354] The server analyzes the collected data and extracts meaningful insights for users. Using data analysis tools such as Apache Hadoop and Apache Spark, the data is processed and calculated to obtain insights such as sentiment fluctuations and the usability of the user interface. Based on this, improvement measures are proposed.

[0355] Step 7:

[0356] The user uses a terminal to check the report obtained from the server. This report integrates sentiment analysis data and review results and includes specific improvement measures. The user then formulates an improvement plan for the application or product based on the report.

[0357] In this way, by carrying out specific processing at each step, it is possible to quickly obtain highly accurate reviews and sentiment analysis specific to the target demographic, thereby improving the quality of applications and products.

[0358] 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.

[0359] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0360] 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.

[0361] [Second embodiment]

[0362] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0363] 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.

[0364] 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).

[0365] 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.

[0366] 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.

[0367] 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).

[0368] 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.

[0369] 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.

[0370] 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.

[0371] 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.

[0372] 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.

[0373] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0374] To implement the present invention, a system is used that includes procedures for generating virtual reviewers using generative artificial intelligence, assigning review tasks, and collecting and analyzing review results. The operation of this system is specifically described below.

[0375] System configuration

[0376] 1. User input and application upload

[0377] The user uses the device interface to input demographic data (e.g., age, gender, interests, and frequency of use) about the target demographic, and uploads the application to be reviewed to the server via the device, allowing the user to send a review request to the server.

[0378] 2. Creating a Virtual Reviewer

[0379] The server receives target demographic attribute data sent by the user and uses generative artificial intelligence to generate a virtual reviewer based on that data. The generated virtual reviewer has the pre-set target demographic attributes and evaluates the application from a perspective based on those attributes.

[0380] 3. Assigning review tasks and conducting reviews

[0381] The server assigns the task of reviewing the uploaded application to the generated virtual reviewer, who then uses the application to perform this task and provides a detailed evaluation according to the specified criteria.

[0382] 4. Collection and analysis of review results

[0383] The server collects the review results generated by the virtual reviewers, analyzes them further, and extracts meaningful insights for users (e.g., UI improvements, performance bottlenecks). The analysis results are displayed in the form of a report on the platform and provided to users.

[0384] 5. Formulation of improvement measures

[0385] Users can use their devices to check the review results received from the server, and based on the feedback they receive, they can formulate improvements to the application and incorporate them into the next development cycle, thereby continuously improving the quality of the application.

[0386] Specific examples

[0387] For example, consider the development of an e-commerce application with new features. The development team specifies the target demographic as "women in their 20s who purchase frequently" and generates virtual reviewers. The virtual reviewers use the application from a perspective based on the specified target demographic and provide detailed feedback on features such as ease of use and UI design.

[0388] After the review is complete, the server collects the review results and generates an analysis report. This report includes major issues and improvement suggestions. The user (development team) refers to this report and formulates improvement measures for new functions. This enables high-quality application releases in a short period of time.

[0389] This invention significantly reduces the time and cost required for conventional QA processes and enables the rapid acquisition of highly accurate reviews tailored to the target demographic.

[0390] The processing flow will be explained below.

[0391] Step 1:

[0392] The user opens the interface on their device, enters the demographic data of the target demographic (e.g., age, gender, interests, frequency of use), and clicks the send button once the input is complete.

[0393] Step 2:

[0394] The device temporarily stores the target demographic attribute data entered by the user and prepares to send it to the server. At the same time, the user selects the application to be reviewed (e.g., APK file) and clicks the upload button to begin uploading the application.

[0395] Step 3:

[0396] The server receives the target demographic attribute data and uploaded application files sent from the device. The received data is temporarily stored and organized as input data for the generative AI.

[0397] Step 4:

[0398] The server uses generative artificial intelligence based on the received target demographic attribute data to generate a virtual reviewer that matches the specified attributes. This virtual reviewer is a virtual user with the specific attributes entered by the user.

[0399] Step 5:

[0400] The server assigns a review task for the application to be reviewed to the generated virtual reviewer, which includes viewpoints and evaluation criteria specified by the user.

[0401] Step 6:

[0402] The virtual reviewer uses the application under review in a simulated environment and evaluates it according to specified criteria, such as the ease of use of the UI, the intuitiveness of the features, and performance.

[0403] Step 7:

[0404] After completing the review task, the virtual reviewers send the generated evaluation results and feedback to the server, which includes detailed evaluation comments and suggestions for improvement.

[0405] Step 8:

[0406] The server collects the review results received from the virtual reviewers and organizes them as data for analysis. The server analyzes the collected data and extracts insights that are meaningful to the user.

[0407] Step 9:

[0408] The server generates the analysis results in the form of a report and displays it on the platform, and sends a notification to the user informing them that the review results are available for access.

[0409] Step 10:

[0410] Users access the server via their terminals and check the review results displayed on the platform. Users can view detailed reports and evaluate the feedback of each virtual reviewer.

[0411] Step 11:

[0412] The user will then formulate improvements to the application based on the review results obtained, and will incorporate this data into the next development cycle to create plans for new features and bug fixes.

[0413] In this way, the entire processing flow works together efficiently, allowing users to quickly obtain highly accurate reviews specific to their target demographic and improve the quality of their applications.

[0414] Example 1

[0415] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0416] The conventional evaluation process had problems in that it was difficult to quickly obtain highly accurate reviews specific to a target demographic, and the reviews were time-consuming and costly. In particular, in order to collect feedback from different target demographics, it was necessary to arrange for multiple reviewers to conduct the evaluations, which resulted in a significant burden of cost and time.

[0417] 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.

[0418] In this invention, the server includes means for receiving target demographic attribute data and the program to be reviewed based on user input, means for using a generative artificial intelligence model to generate virtual reviewers that match the received target demographic attribute data, means for assigning review tasks to the generated virtual reviewers and having them perform detailed evaluations of the program, means for collecting review results from the virtual reviewers and analyzing the results, and means for providing the analysis results to the user. This makes it possible to quickly obtain highly accurate reviews tailored to the target demographic, significantly reducing the time and cost required for conventional evaluation processes.

[0419] "User input" refers to a person or entity using the system providing demographic data and programs to be reviewed to the system.

[0420] "Target demographic attribute data" refers to information that indicates the characteristics of the target user demographic when writing a review (e.g., age, gender, interests, frequency of use).

[0421] "Reviewed Program" means software or an application that is subject to evaluation and review.

[0422] A "virtual reviewer" is a reviewer simulated based on specific attribute data using a generative artificial intelligence model.

[0423] A "generative artificial intelligence model" is an artificial intelligence algorithm that simulates human opinions and behaviors based on specific prompts.

[0424] A "review task" is a task that includes specific evaluation activities and checkpoints assigned to a virtual reviewer.

[0425] "Review results" are the feedback and evaluation scores obtained when a virtual reviewer performs a review task.

[0426] "Analysis results" are the results of further analyzing the collected review results and extracting insights that are useful to users (e.g., areas for improvement in the UI, performance bottlenecks).

[0427] "Display in report format" means that the analysis results are provided in a document or table format that is easy for the user to understand.

[0428] A "prompt sentence" is an input sentence given to a generative artificial intelligence model, and is a sentence that includes instructions for evaluation based on specific attributes and evaluation criteria.

[0429] MODE FOR CARRYING OUT THE INVENTION

[0430] To implement the present invention, a system is used in which a user inputs attribute data of the target demographic and the program to be reviewed, and a virtual reviewer is generated based on the input data using generative artificial intelligence. A specific implementation method of this system is described below.

[0431] User input and application upload

[0432] Users use their own devices to input demographic data (e.g., age, gender, interests, and frequency of use) about their target demographic. This data is important information for reflecting the characteristics of the target demographic. Users also upload the program (e.g., application) to be reviewed to the server via their devices. This sends a review request to the server in response to the user's request.

[0433] Virtual reviewer generation

[0434] The server receives the target demographic attribute data sent by the user and generates a virtual reviewer based on that data. The virtual reviewer is generated using a generative artificial intelligence model (e.g., GPT-3). By inputting the target demographic profile into this model, a virtual reviewer with specific attributes is generated.

[0435] Assigning review tasks and conducting reviews

[0436] The server assigns the uploaded program review task to the generated virtual reviewer, who then performs a detailed evaluation according to specified criteria (e.g., ease of use of functions, UI design, performance). The virtual reviewer operates the program in an AI simulation environment and performs the evaluation through simulated user operations.

[0437] Collection and analysis of review results

[0438] The server collects and analyzes the reviews generated by the virtual reviewers. The review results include specific feedback and rating scores that users may find interesting. The server further analyzes this data to extract useful insights for users. The analysis results are displayed on the platform in the form of a report.

[0439] Formulating improvement measures

[0440] Users can use their devices to check the review results and analysis reports provided by the server, and can formulate improvements to the application based on this feedback and incorporate them into the next development cycle, thereby continuously improving the quality of the application.

[0441] Specific examples

[0442] For example, consider the development of an e-commerce application with new features. The development team (users) specify the target demographic as "women in their 20s with high purchasing frequency" and generate a virtual reviewer. The virtual reviewer will use the application from a perspective based on the specified attributes and provide detailed feedback on features such as ease of use and UI design.

[0443] Example prompts used to generate virtual reviewers:

[0444] "Please rate in detail the ease of use of the new features for women in their 20s."

[0445] "Please review in detail how users who frequently use SNS experience the UI."

[0446] After the review is completed, the server collects the review results and generates an analysis report, which includes major issues and improvement suggestions. The development team (users) can refer to this report and quickly develop improvement plans for new features.

[0447] This invention significantly reduces the time and cost required for the conventional evaluation process, making it possible to quickly obtain highly accurate reviews specific to the target demographic.

[0448] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0449] Step 1: User Input

[0450] The user uses the device interface to input attribute data of the target demographic (e.g., age, gender, interests, frequency of use). Specifically, the user uses forms and check boxes to input data such as "20s," "female," and "frequent SNS user." The input data is organized within the device and sent to the server.

[0451] Input: Target demographic data (e.g., age, gender, interests, frequency of use)

[0452] Output: Attribute data sent to the server

[0453] Step 2: Upload your application

[0454] The user uploads the application to be reviewed to the server via their device. Specifically, the user selects the specified application file (e.g., apk file or ipa file) from their device and clicks the "Upload" button. The file is then transferred from the device to the server.

[0455] Input: Application file to be reviewed

[0456] Output: Application files uploaded to the server

[0457] Step 3: Receive target data

[0458] The server receives the target demographic attribute data sent by the user. The received data is stored in a database and used as a target demographic profile. Data integrity checks are also performed at this stage.

[0459] Input: Target demographic data submitted by the user

[0460] Output: Attribute data stored in a database

[0461] Step 4: Generate virtual reviewers

[0462] The server generates virtual reviewers using a generative artificial intelligence model (e.g., GPT-3). By inputting the target demographic profile into a prompt statement (e.g., "Please generate a virtual reviewer for a woman in her 20s"), the AI ​​generates a virtual reviewer with specific attributes.

[0463] Input: Target demographic profile

[0464] Output: Generated virtual reviewer

[0465] Step 5: Assign review tasks

[0466] The server assigns review tasks for uploaded applications to the generated virtual reviewers, including evaluation criteria (e.g., usability, design, performance).

[0467] Input: Generated virtual reviewer, application to be reviewed

[0468] Output: Tasks assigned to virtual reviewers

[0469] Step 6: Conducting the review

[0470] The virtual reviewer interacts with the application in a simulated environment to perform the assigned task: interacting with the UI, testing each feature, and recording their evaluation.

[0471] Input: Review task, Simulation environment

[0472] Output: Review results from the virtual reviewer

[0473] Step 7: Collect review results

[0474] The server collects the review results generated by the virtual reviewers, which are then stored in a database and treated as a dataset for later analysis.

[0475] Input: Review results from a hypothetical reviewer

[0476] Output: Review results stored in a database

[0477] Step 8: Analyze the results

[0478] The server analyzes the collected reviews and extracts useful insights for users. Specifically, it uses data analysis algorithms to identify key issues and improvement suggestions.

[0479] Input: Collected review results

[0480] Output: Analysis results (insights and improvement suggestions)

[0481] Step 9: View the results

[0482] The server displays the analysis results in the form of a report on the platform, which the user can view via their terminal.

[0483] Input: Analysis results

[0484] Output: Analysis results displayed in report format

[0485] Step 10: Developing improvement measures

[0486] The user formulates improvement measures for the application based on the review results and analysis reports provided by the server, and reflects these measures in the next development cycle to improve the quality of the application.

[0487] Input: Review results and analysis report

[0488] Output: Improvements

[0489] (Application example 1)

[0490] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0491] Conventional review systems often struggle to quickly obtain detailed evaluations tailored to a target demographic, resulting in problems with the quality and accuracy of reviews. Furthermore, when using human reviewers, the review tasks are time-consuming and costly, making it difficult to obtain efficient feedback. This often prevents users from quickly obtaining suggestions for improvements or product modifications, delaying the product development cycle. The present invention aims to solve these problems.

[0492] 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.

[0493] In this invention, the server includes means for generating virtual reviewers, means for assigning review tasks to the generated virtual reviewers, means for collecting review results from the virtual reviewers, means for analyzing the collected review results and providing them to the user, means for receiving attribute data of the target demographic based on user input and using the attribute data to send prompt sentences to the generative artificial intelligence to generate virtual reviewers, and means for performing detailed product evaluations based on the perspective of the target demographic and analyzing the results to provide insights to the user. This makes it possible to quickly obtain highly accurate reviews tailored to the target demographic.

[0494] A "virtual reviewer" is a virtual entity generated using generative artificial intelligence based on attribute data of a specific target demographic, and is a system that reviews applications and products from the perspective of an actual user.

[0495] A "review task" is an application or product evaluation task assigned to a virtual reviewer, providing detailed feedback based on specific criteria or attributes.

[0496] "Review results" refer to the evaluation information and feedback obtained by the virtual reviewers when they perform the review task, including specific comments about the product's usability, appearance, functionality, etc.

[0497] "Target demographic attribute data" is information used to define a specific target demographic, such as age, gender, interests, and frequency of use. Based on this data, generative AI generates virtual reviewers.

[0498] "Generative AI" is an AI that has the ability to generate virtual reviewers based on the attribute data of the target demographic entered by the user, and is a technology that receives prompt statements and generates an appropriate model.

[0499] A "prompt sentence" is an input sentence that specifies a specific target demographic and evaluation criteria to the generative artificial intelligence, which then generates a virtual reviewer.

[0500] A "detailed evaluation" is an evaluation of a product by a virtual reviewer according to specific criteria, including specific feedback on specific usability, design, performance, etc.

[0501] "Insights" are useful information and suggestions obtained by analyzing collected review results, and serve as guidelines for improving products and adding new features.

[0502] To implement the present invention, it is necessary to operate the system in the following procedure. This system is made up of a server, a terminal, and a user, and the specific procedure is as follows.

[0503] 1. User input and product data upload

[0504] Users use the device interface to input demographic data (e.g., age, gender, interests, frequency of use) of their target demographic and upload product data to be reviewed. This input data is sent to the server and used to generate virtual reviewers.

[0505] 2. Creating a Virtual Reviewer

[0506] The server uses generative artificial intelligence (specifically, OpenAI's GPT-4 model) to generate a virtual reviewer based on the target demographic attribute data it receives. The generative artificial intelligence sends prompts using the input attribute data, and generates a virtual reviewer based on the results.

[0507] Example prompt sentence:

[0508] "Generate a virtual reviewer with the following attributes: age: 20-29, gender: female, purchase frequency: high"

[0509] 3. Assigning and completing review tasks

[0510] The server assigns review tasks to the virtual reviewers, who then perform the tasks and provide detailed evaluations according to specific criteria (e.g., usability, appearance, cost-effectiveness).

[0511] 4. Collection and analysis of review results

[0512] The server collects and analyzes the reviews generated by the virtual reviewers. This analysis uses natural language processing technology based on generative artificial intelligence to extract key issues and improvement suggestions. These results are provided to users in the form of useful insights (e.g., areas for product improvement, the need for new features).

[0513] 5. Formulation of improvement measures

[0514] Users can use their devices to check the review results obtained from the server and formulate measures to improve the product. This feedback is reflected in the next development cycle, contributing to improving the quality of the product.

[0515] Examples:

[0516] For example, consider the case of reviewing product data for a fashion dress aimed at women in their 20s. The user inputs target demographic attribute data, such as "women in their 20s, frequent buyers," and uploads the product data for the fashion dress to the server. The server then generates a virtual reviewer based on this data and has that reviewer evaluate the dress. The review results include the quality of the design, functionality, cost-effectiveness, etc., and the user can use this information to formulate improvements to the product.

[0517] This system allows us to quickly obtain highly accurate reviews tailored to the target demographic, improving the quality and accuracy of reviews while shortening the development cycle. By using these methods, users can effectively evaluate and improve products.

[0518] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0519] Step 1:

[0520] The user uses the device interface to input demographic data for the target demographic and upload product data for review. The input demographic data includes age, gender, interests, and frequency of use, while the product data includes name, price, and description. This data is then sent to the server.

[0521] Input: Target demographic data (e.g., age, gender, interests, frequency of use), product data (e.g., name, price, description)

[0522] Output: Target demographic attribute data and product data sent to the server

[0523] Step 2:

[0524] The server then sends a prompt to the generative AI based on the target demographic attribute data it receives, generating a virtual reviewer. The prompt includes the target demographic attribute data, and a virtual reviewer is generated based on that.

[0525] Input: target demographic attribute data, prompt text

[0526] Output: Generated virtual reviewer

[0527] Step 3:

[0528] The server assigns review tasks to the virtual reviewers, who then perform detailed evaluations based on the uploaded product data and specified criteria (e.g., usability, appearance, and price-performance).

[0529] Input: Virtual reviewer, product data, rating criteria

[0530] Output: Detailed evaluation results

[0531] Step 4:

[0532] The server collects the evaluation results generated by the virtual reviewers and analyzes the data, using natural language processing techniques (e.g., generative artificial intelligence) to extract key issues and improvement suggestions.

[0533] Input: Review results

[0534] Output: Analyzed insights (e.g., key issues, improvement suggestions)

[0535] Step 5:

[0536] The user uses a terminal to check the review results obtained from the server and formulates measures to improve the product. The user then reflects the feedback in the next development cycle.

[0537] Input: Review results (analysis report)

[0538] Output: Improved product development plan

[0539] 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.

[0540] To implement the present invention, a system is used that includes procedures for generating virtual reviewers using generative artificial intelligence, review evaluation using an emotion engine, and collection and analysis of review results. The operation of this system is described in detail below.

[0541] System configuration

[0542] 1. User input and application upload

[0543] The user uses the device interface to input demographic data (e.g., age, gender, interests, and frequency of use) about the target demographic, and uploads the application to be reviewed to the server via the device, allowing the user to send a review request to the server.

[0544] 2. Creating a Virtual Reviewer

[0545] The server receives target demographic attribute data sent by the user and uses generative artificial intelligence to generate a virtual reviewer based on that data. The generated virtual reviewer has the pre-set target demographic attributes and evaluates the application from a perspective based on those attributes.

[0546] 3. Emotion Recognition by Emotion Engine

[0547] The server integrates an emotion engine when assigning review tasks for applications to the virtual reviewers. As the virtual reviewers use the applications, the emotion engine recognizes and records their emotional states in real time. This allows us to quantitatively understand the emotional state of the virtual reviewers.

[0548] 4. Assigning review tasks and conducting reviews

[0549] The server assigns a review task for the application to the generated virtual reviewer. The virtual reviewer uses the application to perform this task and provides a detailed evaluation according to the specified criteria. During the evaluation process, the emotion engine monitors the emotional state of the virtual reviewer and generates emotion analysis data.

[0550] 5. Collection and analysis of review results

[0551] The server collects the review results generated by the virtual reviewers and sentiment analysis data from the sentiment engine, and further analyzes this data to extract meaningful insights for users (e.g., emotional responses to UI usability, sentiment fluctuations for specific features).

[0552] 6. Providing insights and suggesting improvements

[0553] The server generates the analysis results in the form of a report and displays it on the platform. The report includes information that integrates review results and sentiment analysis, providing more meaningful insights for users. Users can review the report and formulate improvements for their applications.

[0554] 7. Formulation of improvement measures

[0555] Users can use their devices to check the reports received from the server and evaluate the detailed review results and sentiment analysis data. This allows users to make specific improvements based on their emotional state (e.g., changing the interface because a specific feature causes discomfort). Based on the feedback obtained, it is possible to plan new features and bug fixes for the next development cycle.

[0556] Specific examples

[0557] For example, when developing a new e-commerce application, a user may specify a target demographic of "women in their 20s who purchase frequently." Virtual reviewers are generated based on these attributes and work with an emotion engine to conduct reviews. Review results include not only evaluations of product display speed and UI, but also associated emotional fluctuations (e.g., stress levels when a product loads slowly).

[0558] After the review is complete, the server collects and analyzes all data and provides a detailed report to the user, who can then refer to the report and develop specific improvement measures that take into account the user's emotional state and incorporate them into the next release.

[0559] In this way, the system of the present invention makes it possible to quickly obtain highly accurate reviews and sentiment analysis specific to the target demographic, thereby improving the quality of applications.

[0560] The processing flow will be explained below.

[0561] Step 1:

[0562] The user opens the interface on their device, enters the demographic data of the target demographic (e.g., age, gender, interests, frequency of use), and clicks the send button once the input is complete.

[0563] Step 2:

[0564] The device temporarily stores the target demographic attribute data entered by the user and prepares to send it to the server.The user also selects the application to be reviewed via the device and uploads the application by clicking the upload button.

[0565] Step 3:

[0566] The server receives the target demographic attribute data and uploaded application files sent from the device, temporarily stores the received data, and organizes it as input data for the generative AI.

[0567] Step 4:

[0568] The server uses generative artificial intelligence to generate a virtual reviewer based on the received target demographic attribute data. The generated virtual reviewer has the specified attributes and evaluates the application from a perspective based on those attributes.

[0569] Step 5:

[0570] The server integrates an emotion engine with the generated virtual reviewer, which recognizes and records the emotional state of the virtual reviewer in real time.

[0571] Step 6:

[0572] The server assigns review tasks for uploaded applications to virtual reviewers, with the review tasks including user-specified viewpoints and evaluation criteria.

[0573] Step 7:

[0574] The virtual reviewer uses the application under review and provides a detailed evaluation according to the specified criteria. The emotion engine monitors the virtual reviewer's emotional state (e.g., stress level, satisfaction) while using the application and generates emotion analysis data.

[0575] Step 8:

[0576] The server collects the review results generated by the virtual reviewers and the sentiment analysis data generated by the sentiment engine. This collected data is temporarily stored and then used for the next analysis step.

[0577] Step 9:

[0578] The server analyzes the collected review results and sentiment analysis data to extract meaningful insights for users (e.g., emotional reactions to UI usability, sentiment fluctuations for specific features). The analysis results are generated in the form of a report and displayed on the platform.

[0579] Step 10:

[0580] The user uses a terminal to check the report retrieved from the server, which contains information that integrates review results and sentiment analysis.

[0581] Step 11:

[0582] Users can formulate improvements to their applications based on the review results and sentiment analysis data they obtain. Specifically, if a particular feature causes discomfort, they can consider specific measures such as changing the interface. The obtained feedback can be reflected in the next development cycle, allowing them to plan new features and bug fixes.

[0583] In this way, the system of the present invention can quickly obtain highly accurate reviews and sentiment analysis specific to the target demographic, effectively improving the quality of applications.

[0584] Example 2

[0585] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0586] Conventional review systems have difficulty evaluating each target demographic and providing detailed feedback, including emotional states. In particular, users lack the means to create virtual reviewers based on target attributes and collect and analyze their review results along with emotional data, making it difficult to gain insights needed to improve software under development.

[0587] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for receiving attribute data of the target demographic based on user input and generating a virtual reviewer that matches the attribute data, means for assigning a review task to the generated virtual reviewer, means for integrating an emotion engine and recognizing and recording the emotional state of the virtual reviewer in real time, means for collecting review results and emotion analysis data from the virtual reviewer, and means for analyzing the collected review results and emotion analysis data and providing them to the user. This enables highly accurate reviews and emotion analysis tailored to the target demographic.

[0588] "Target demographic attribute data" is information for defining a specific target user group, such as the user's age, gender, interests, and frequency of use.

[0589] A "virtual reviewer" is a fictitious reviewer with specific target demographic attributes, generated using a generative artificial intelligence model.

[0590] A "review task" is a specific instruction and task for evaluating and providing feedback on the software or application being reviewed.

[0591] An "emotion engine" is software or algorithms for recognizing and recording the emotional state of a virtual reviewer in real time.

[0592] "Sentiment Analysis Data" is information about the emotional state of a hypothetical reviewer recorded by the emotion engine, and is a record of emotions that fluctuate during the course of the review.

[0593] A "generative AI model" is an artificial intelligence model that uses natural language processing and machine learning techniques to generate new data and information (in this case, virtual reviewers).

[0594] A "prompt" is an instruction or input to a generative AI model to obtain a specific output.

[0595] "Review evaluation data" refers to evaluation information collected by virtual reviewers while performing a review task, and is detailed feedback on each function and interface of an application.

[0596] The present invention is a system that includes procedures for generating virtual reviewers using generative artificial intelligence, review evaluation using an emotion engine, and collection and analysis of review results. This system collects attribute data of the target demographic based on user input, generates virtual reviewers based on that data, and conducts reviews. In the process, it collects emotion data using an emotion engine, and finally analyzes this data and provides it to the user as a report. The details of this system are described below.

[0597] First, the user uses the device interface to input the target demographic's attribute data (e.g., age, gender, interests, frequency of use, etc.), then uploads the software to be reviewed to the server via the device and submits a review request.

[0598] The server receives the target demographic attribute data sent by the user and generates a virtual reviewer using a generative AI model (e.g., GPT-3 or BERT) based on that data. In this process, a prompt sentence is generated for the generative AI model and input into the model to generate a virtual reviewer.

[0599] The server then assigns review tasks to the virtual reviewers, and integrates an emotion engine (e.g., Microsoft Azure's emotion recognition API) to recognize and record the virtual reviewers' emotional states in real time.

[0600] The virtual reviewer uses the software under review to perform a detailed evaluation according to the specified criteria, during which the emotion engine monitors the reviewer's emotional state and generates sentiment analysis data.

[0601] The server collects the review results generated by the virtual reviewers and the sentiment analysis data generated by the sentiment engine, and organizes and analyzes this data using analytical tools (e.g., Python's Pandas and NumPy libraries) to extract meaningful insights for users.

[0602] Finally, the server generates a report of the analysis results and displays it on the platform. Users can then use their devices to review the report and formulate specific improvement measures based on the detailed review results and sentiment analysis data, which can then be incorporated into the next development cycle.

[0603] As a concrete example, consider the development of a new e-commerce application. The user inputs demographic data such as "women in their 20s who purchase frequently" as the target demographic, and generates a virtual reviewer. This virtual reviewer reviews the application together with an emotion engine. The review results include evaluations of the product display speed and UI, as well as emotional states such as stress felt when products are slow to load.

[0604] Example prompt sentence:

[0605] The target users are women in their 20s who purchase frequently. Please generate a virtual reviewer based on this profile and have them rate the following application.

[0606] In this way, the system of the present invention can quickly perform highly accurate reviews and sentiment analysis specific to the target demographic, allowing users to efficiently improve the quality of their applications.

[0607] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0608] Step 1: User Input and Application Upload

[0609] The user uses the device interface to input demographic data (age, gender, interests, frequency of use, etc.) of the target demographic. The input data is sent to the server in JSON format, which becomes the input.

[0610] The user selects the software to be reviewed (e.g., APK file) and clicks the upload button. The software file is uploaded to the server, which is the output.

[0611] Step 2: Receive demographic data for your target audience

[0612] The server receives the HTTP request, analyzes the target demographic attribute data sent by the user, and saves the received JSON data in an internal database. This becomes the input.

[0613] The attribute data is converted into a format suitable for generative AI models, which then becomes the output.

[0614] Step 3: Generate virtual reviewers

[0615] The server uses a generative AI model (e.g., GPT-3 or BERT) to generate a virtual reviewer. Specifically, it creates a prompt and inputs it into the generative AI model. The prompt is, "The target user is a woman in her 20s who frequently purchases. Please generate a virtual reviewer based on this profile and rate the following application." This is the input.

[0616] The generative AI model generates a virtual reviewer and returns that data to the server, which becomes the output.

[0617] Step 4: Assigning review tasks and integrating the sentiment engine

[0618] The server assigns review tasks for uploaded software to the virtual reviewers, which is the input.

[0619] The server integrates an emotion engine (e.g., Microsoft Azure's emotion recognition API) to recognize and record the emotional state of the virtual reviewer in real time. This is the specific operation. It configures the settings to call the emotion recognition API and sends an API request for each reviewer operation. It receives the response from the API and records the emotional data, which becomes the output.

[0620] Step 5: Performing review tasks

[0621] The virtual reviewers interact with the software according to the assigned review task and provide detailed evaluations based on the specified criteria, which serve as input.

[0622] During operation, the emotion engine monitors the emotional state and generates evaluation data and emotion analysis data, which are the output.

[0623] Step 6: Collect review results and sentiment analysis data

[0624] The server receives review rating data and emotional state data sent by virtual reviewers, which are the input.

[0625] This data is stored in a database and organized and analyzed using analytical tools (e.g., Python's Pandas or NumPy library). This is the specific operation, and the analyzed data is the output.

[0626] Step 7: Generate and serve reports

[0627] The server generates a report based on the analysis results. Specifically, it creates a detailed report by applying the analyzed data to a template. This is the input.

[0628] The generated report is displayed on the user interface and made accessible to the user - this is the output.

[0629] Step 8: Developing improvement measures

[0630] The user uses a terminal to check the report retrieved from the server, which becomes the input.

[0631] Users evaluate the detailed review results and sentiment analysis data and develop specific improvement measures. This is the output. Improvement measures may include, for example, changing the UI design or adding or removing features.

[0632] In this way, the user, server, and device work together at each processing step to achieve highly accurate reviews and sentiment analysis tailored to the target demographic.

[0633] (Application example 2)

[0634] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0635] Conventional review systems rely on user-provided feedback, making it difficult to accurately grasp emotional fluctuations and real-time emotional states. There was also a need for a method to quickly and accurately obtain reviews that are optimal for the target demographic.

[0636] 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 virtual reviewers, means for assigning review tasks to the generated virtual reviewers, means for collecting review results from the virtual reviewers, means for analyzing the collected review results and providing them to users, means for collecting and analyzing emotional data in real time, and means for monitoring and recording the emotional state of users from their gaze, facial expressions, etc. This makes it possible to quickly obtain highly accurate reviews and emotional analysis tailored to the target demographic and efficiently propose improvement measures for applications and products.

[0637] A "virtual reviewer" is a simulated subject for conducting a review, generated using generative artificial intelligence based on attribute data of a specific target demographic.

[0638] A "review task" is a series of tasks assigned to a virtual reviewer to evaluate a particular application or product and record the results.

[0639] "Emotional data" is data used to record and analyze changes in the user's emotional state in real time based on information such as their gaze, facial expressions, and voice.

[0640] "Means for collecting and analyzing emotional data in real time" refers to technology that continuously monitors the user's gaze, facial expressions, etc., and records and analyzes their emotional state on the spot.

[0641] "Generative AI" is an AI model that generates responses and evaluations for specific tasks based on input attribute data.

[0642] "Means for monitoring and recording emotional states from gaze, facial expressions, etc." refers to technology that detects a user's gaze and facial expressions through smart glasses or the like when the user is browsing an online shopping site, and collects and stores that emotional state in real time.

[0643] A "target demographic" is a group of users with specific attributes such as age, gender, interests, and frequency of use, and is the group that is analyzed to evaluate and improve products and services.

[0644] To implement the present invention, the following system configuration and processing procedures are used: Specifically, review evaluation using a virtual reviewer and an emotion engine, and review result collection and analysis procedures.

[0645] System configuration

[0646] 1. User input and application upload

[0647] Users use the device interface to input demographic data (age, gender, interests, frequency of use, etc.) about their target demographic, and also upload the application to be reviewed to the server via the device, allowing the user to send a review request to the server.

[0648] 2. Creating a Virtual Reviewer

[0649] The server receives target demographic attribute data sent by the user and uses generative artificial intelligence to generate a virtual reviewer based on that data. The generated virtual reviewer has the pre-set target demographic attributes and evaluates the application from a perspective based on those attributes.

[0650] 3. Emotion Recognition by Emotion Engine

[0651] The server integrates an emotion engine when assigning review tasks for applications to the virtual reviewers. As the virtual reviewers use the applications, the emotion engine recognizes and records their emotional states in real time. This allows us to quantitatively understand the emotional state of the virtual reviewers.

[0652] 4. Assigning review tasks and conducting reviews

[0653] The server assigns a review task for the application to the generated virtual reviewer. The virtual reviewer uses the application to perform this task and provides a detailed evaluation according to the specified criteria. During the evaluation process, the emotion engine monitors the emotional state of the virtual reviewer and generates emotion analysis data.

[0654] 5. Collection and analysis of review results

[0655] The server collects the review results generated by the virtual reviewers and sentiment analysis data from the sentiment engine, and further analyzes this data to extract meaningful insights for users (e.g., emotional responses to the usability of the user interface, sentiment fluctuations for specific features).

[0656] Specific examples

[0657] For example, consider a female user in her twenties searching for new clothes on an online shopping site. The user puts on smart glasses and launches a dedicated app. The emotion engine collects real-time emotional data from the user's gaze and facial expressions, recording their emotional state. At the same time, a virtual reviewer using generative artificial intelligence evaluates the same product page. As a result, the emotional data of the user and the virtual reviewer is sent to a server for analysis.

[0658] The analysis results are provided to users in the form of a report, which includes specific improvement measures based on the sentiment data. For example, the report may include insights such as "users feel frustrated when products take a long time to load" or "certain designs are particularly well-received."

[0659] Prompt Sentence Examples

[0660] The following prompt sentence is input into the generative artificial intelligence:

[0661] Generate virtual reviewers based on the following user attributes:

[0662] User Attributes:

[0663] Age: 20s

[0664] Gender: Female

[0665] Interests: Fashion, shopping

[0666] Frequency of use: High

[0667] We assign a hypothetical reviewer the task of rating a clothing product page on an online shopping website at the following URL:

[0668] URL: [example.com / fashion / clothing / item1234]

[0669] Output your impressions and emotional data.

[0670] This allows for quick and accurate target-specific reviews and sentiment analysis, which can be used to improve the quality of applications and products.

[0671] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0672] Step 1:

[0673] The user uses the device interface to input demographic data (age, gender, interests, frequency of use, etc.) of the target demographic. The application to be reviewed is then uploaded to the server via the device. This input data is sent to the server and used as the basis for generating virtual reviewers.

[0674] Step 2:

[0675] The server receives the target demographic attribute data sent by the user and generates a virtual reviewer based on that data using generative AI. Specifically, the received attribute data is input into the generative AI model as a prompt sentence, generating an appropriate virtual reviewer.

[0676] Step 3:

[0677] The server assigns a review task for the application to the virtual reviewer, who then evaluates the application according to the specified criteria. During this process, the virtual reviewer's evaluation is recorded.

[0678] Step 4:

[0679] The server integrates an emotion engine to collect real-time emotional data from users' gazes and facial expressions. When users browse online shopping sites through smart glasses, emotion data is recorded in real time using eye tracking and facial recognition technology. This data is sent to the server and integrated with the evaluation data from the virtual reviewers.

[0680] Step 5:

[0681] The server collects the review results generated by the virtual reviewers and sentiment analysis data from the emotion engine. This data includes the user's emotional fluctuations and eye movements, forming detailed user experience data. This allows for comprehensive review results.

[0682] Step 6:

[0683] The server analyzes the collected data and extracts meaningful insights for users. Using data analysis tools such as Apache Hadoop and Apache Spark, the data is processed and calculated to obtain insights such as sentiment fluctuations and the usability of the user interface. Based on this, improvement measures are proposed.

[0684] Step 7:

[0685] The user uses a terminal to check the report obtained from the server. This report integrates sentiment analysis data and review results and includes specific improvement measures. The user then formulates an improvement plan for the application or product based on the report.

[0686] In this way, by carrying out specific processing at each step, it is possible to quickly obtain highly accurate reviews and sentiment analysis specific to the target demographic, thereby improving the quality of applications and products.

[0687] 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.

[0688] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0689] 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.

[0690] [Third embodiment]

[0691] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0692] 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.

[0693] 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).

[0694] 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.

[0695] 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.

[0696] 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).

[0697] 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.

[0698] 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.

[0699] 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.

[0700] 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.

[0701] 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.

[0702] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0703] To implement the present invention, a system is used that includes procedures for generating virtual reviewers using generative artificial intelligence, assigning review tasks, and collecting and analyzing review results. The operation of this system is specifically described below.

[0704] System configuration

[0705] 1. User input and application upload

[0706] The user uses the device interface to input demographic data (e.g., age, gender, interests, and frequency of use) about the target demographic, and uploads the application to be reviewed to the server via the device, allowing the user to send a review request to the server.

[0707] 2. Creating a Virtual Reviewer

[0708] The server receives target demographic attribute data sent by the user and uses generative artificial intelligence to generate a virtual reviewer based on that data. The generated virtual reviewer has the pre-set target demographic attributes and evaluates the application from a perspective based on those attributes.

[0709] 3. Assigning review tasks and conducting reviews

[0710] The server assigns the task of reviewing the uploaded application to the generated virtual reviewer, who then uses the application to perform this task and provides a detailed evaluation according to the specified criteria.

[0711] 4. Collection and analysis of review results

[0712] The server collects the review results generated by the virtual reviewers, analyzes them further, and extracts meaningful insights for users (e.g., UI improvements, performance bottlenecks). The analysis results are displayed in the form of a report on the platform and provided to users.

[0713] 5. Formulation of improvement measures

[0714] Users can use their devices to check the review results received from the server, and based on the feedback they receive, they can formulate improvements to the application and incorporate them into the next development cycle, thereby continuously improving the quality of the application.

[0715] Specific examples

[0716] For example, consider the development of an e-commerce application with new features. The development team specifies the target demographic as "women in their 20s who purchase frequently" and generates virtual reviewers. The virtual reviewers use the application from a perspective based on the specified target demographic and provide detailed feedback on features such as ease of use and UI design.

[0717] After the review is complete, the server collects the review results and generates an analysis report. This report includes major issues and improvement suggestions. The user (development team) refers to this report and formulates improvement measures for new functions. This enables high-quality application releases in a short period of time.

[0718] This invention significantly reduces the time and cost required for conventional QA processes and enables the rapid acquisition of highly accurate reviews tailored to the target demographic.

[0719] The processing flow will be explained below.

[0720] Step 1:

[0721] The user opens the interface on their device, enters the demographic data of the target demographic (e.g., age, gender, interests, frequency of use), and clicks the send button once the input is complete.

[0722] Step 2:

[0723] The device temporarily stores the target demographic attribute data entered by the user and prepares to send it to the server. At the same time, the user selects the application to be reviewed (e.g., APK file) and clicks the upload button to begin uploading the application.

[0724] Step 3:

[0725] The server receives the target demographic attribute data and uploaded application files sent from the device. The received data is temporarily stored and organized as input data for the generative AI.

[0726] Step 4:

[0727] The server uses generative artificial intelligence based on the received target demographic attribute data to generate a virtual reviewer that matches the specified attributes. This virtual reviewer is a virtual user with the specific attributes entered by the user.

[0728] Step 5:

[0729] The server assigns a review task for the application to be reviewed to the generated virtual reviewer, which includes viewpoints and evaluation criteria specified by the user.

[0730] Step 6:

[0731] The virtual reviewer uses the application under review in a simulated environment and evaluates it according to specified criteria, such as the ease of use of the UI, the intuitiveness of the features, and performance.

[0732] Step 7:

[0733] After completing the review task, the virtual reviewers send the generated evaluation results and feedback to the server, which includes detailed evaluation comments and suggestions for improvement.

[0734] Step 8:

[0735] The server collects the review results received from the virtual reviewers and organizes them as data for analysis. The server analyzes the collected data and extracts insights that are meaningful to the user.

[0736] Step 9:

[0737] The server generates the analysis results in the form of a report and displays it on the platform, and sends a notification to the user informing them that the review results are available for access.

[0738] Step 10:

[0739] Users access the server via their terminals and check the review results displayed on the platform. Users can view detailed reports and evaluate the feedback of each virtual reviewer.

[0740] Step 11:

[0741] The user will then formulate improvements to the application based on the review results obtained, and will incorporate this data into the next development cycle to create plans for new features and bug fixes.

[0742] In this way, the entire processing flow works together efficiently, allowing users to quickly obtain highly accurate reviews specific to their target demographic and improve the quality of their applications.

[0743] Example 1

[0744] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0745] The conventional evaluation process had problems in that it was difficult to quickly obtain highly accurate reviews specific to a target demographic, and the reviews were time-consuming and costly. In particular, in order to collect feedback from different target demographics, it was necessary to arrange for multiple reviewers to conduct the evaluations, which resulted in a significant burden of cost and time.

[0746] 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.

[0747] In this invention, the server includes means for receiving target demographic attribute data and the program to be reviewed based on user input, means for using a generative artificial intelligence model to generate virtual reviewers that match the received target demographic attribute data, means for assigning review tasks to the generated virtual reviewers and having them perform detailed evaluations of the program, means for collecting review results from the virtual reviewers and analyzing the results, and means for providing the analysis results to the user. This makes it possible to quickly obtain highly accurate reviews tailored to the target demographic, significantly reducing the time and cost required for conventional evaluation processes.

[0748] "User input" refers to a person or entity using the system providing demographic data and programs to be reviewed to the system.

[0749] "Target demographic attribute data" refers to information that indicates the characteristics of the target user demographic when writing a review (e.g., age, gender, interests, frequency of use).

[0750] "Reviewed Program" means software or an application that is subject to evaluation and review.

[0751] A "virtual reviewer" is a reviewer simulated based on specific attribute data using a generative artificial intelligence model.

[0752] A "generative artificial intelligence model" is an artificial intelligence algorithm that simulates human opinions and behaviors based on specific prompts.

[0753] A "review task" is a task that includes specific evaluation activities and checkpoints assigned to a virtual reviewer.

[0754] "Review results" are the feedback and evaluation scores obtained when a virtual reviewer performs a review task.

[0755] "Analysis results" are the results of further analyzing the collected review results and extracting insights that are useful to users (e.g., areas for improvement in the UI, performance bottlenecks).

[0756] "Display in report format" means that the analysis results are provided in a document or table format that is easy for the user to understand.

[0757] A "prompt sentence" is an input sentence given to a generative artificial intelligence model, and is a sentence that includes instructions for evaluation based on specific attributes and evaluation criteria.

[0758] MODE FOR CARRYING OUT THE INVENTION

[0759] To implement the present invention, a system is used in which a user inputs attribute data of the target demographic and the program to be reviewed, and a virtual reviewer is generated based on the input data using generative artificial intelligence. A specific implementation method of this system is described below.

[0760] User input and application upload

[0761] Users use their own devices to input demographic data (e.g., age, gender, interests, and frequency of use) about their target demographic. This data is important information for reflecting the characteristics of the target demographic. Users also upload the program (e.g., application) to be reviewed to the server via their devices. This sends a review request to the server in response to the user's request.

[0762] Virtual reviewer generation

[0763] The server receives the target demographic attribute data sent by the user and generates a virtual reviewer based on that data. The virtual reviewer is generated using a generative artificial intelligence model (e.g., GPT-3). By inputting the target demographic profile into this model, a virtual reviewer with specific attributes is generated.

[0764] Assigning review tasks and conducting reviews

[0765] The server assigns the uploaded program review task to the generated virtual reviewer, who then performs a detailed evaluation according to specified criteria (e.g., ease of use of functions, UI design, performance). The virtual reviewer operates the program in an AI simulation environment and performs the evaluation through simulated user operations.

[0766] Collection and analysis of review results

[0767] The server collects and analyzes the reviews generated by the virtual reviewers. The review results include specific feedback and rating scores that users may find interesting. The server further analyzes this data to extract useful insights for users. The analysis results are displayed on the platform in the form of a report.

[0768] Formulating improvement measures

[0769] Users can use their devices to check the review results and analysis reports provided by the server, and can formulate improvements to the application based on this feedback and incorporate them into the next development cycle, thereby continuously improving the quality of the application.

[0770] Specific examples

[0771] For example, consider the development of an e-commerce application with new features. The development team (users) specify the target demographic as "women in their 20s with high purchasing frequency" and generate a virtual reviewer. The virtual reviewer will use the application from a perspective based on the specified attributes and provide detailed feedback on features such as ease of use and UI design.

[0772] Example prompts used to generate virtual reviewers:

[0773] "Please rate in detail the ease of use of the new features for women in their 20s."

[0774] "Please review in detail how users who frequently use SNS experience the UI."

[0775] After the review is completed, the server collects the review results and generates an analysis report, which includes major issues and improvement suggestions. The development team (users) can refer to this report and quickly develop improvement plans for new features.

[0776] This invention significantly reduces the time and cost required for the conventional evaluation process, making it possible to quickly obtain highly accurate reviews specific to the target demographic.

[0777] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0778] Step 1: User Input

[0779] The user uses the device interface to input attribute data of the target demographic (e.g., age, gender, interests, frequency of use). Specifically, the user uses forms and check boxes to input data such as "20s," "female," and "frequent SNS user." The input data is organized within the device and sent to the server.

[0780] Input: Target demographic data (e.g., age, gender, interests, frequency of use)

[0781] Output: Attribute data sent to the server

[0782] Step 2: Upload your application

[0783] The user uploads the application to be reviewed to the server via their device. Specifically, the user selects the specified application file (e.g., apk file or ipa file) from their device and clicks the "Upload" button. The file is then transferred from the device to the server.

[0784] Input: Application file to be reviewed

[0785] Output: Application files uploaded to the server

[0786] Step 3: Receive target data

[0787] The server receives the target demographic attribute data sent by the user. The received data is stored in a database and used as a target demographic profile. Data integrity checks are also performed at this stage.

[0788] Input: Target demographic data submitted by the user

[0789] Output: Attribute data stored in a database

[0790] Step 4: Generate virtual reviewers

[0791] The server generates virtual reviewers using a generative artificial intelligence model (e.g., GPT-3). By inputting the target demographic profile into a prompt statement (e.g., "Please generate a virtual reviewer for a woman in her 20s"), the AI ​​generates a virtual reviewer with specific attributes.

[0792] Input: Target demographic profile

[0793] Output: Generated virtual reviewer

[0794] Step 5: Assign review tasks

[0795] The server assigns review tasks for uploaded applications to the generated virtual reviewers, including evaluation criteria (e.g., usability, design, performance).

[0796] Input: Generated virtual reviewer, application to be reviewed

[0797] Output: Tasks assigned to virtual reviewers

[0798] Step 6: Conducting the review

[0799] The virtual reviewer interacts with the application in a simulated environment to perform the assigned task: interacting with the UI, testing each feature, and recording their evaluation.

[0800] Input: Review task, Simulation environment

[0801] Output: Review results from the virtual reviewer

[0802] Step 7: Collect review results

[0803] The server collects the review results generated by the virtual reviewers, which are then stored in a database and treated as a dataset for later analysis.

[0804] Input: Review results from a hypothetical reviewer

[0805] Output: Review results stored in a database

[0806] Step 8: Analyze the results

[0807] The server analyzes the collected reviews and extracts useful insights for users. Specifically, it uses data analysis algorithms to identify key issues and improvement suggestions.

[0808] Input: Collected review results

[0809] Output: Analysis results (insights and improvement suggestions)

[0810] Step 9: View the results

[0811] The server displays the analysis results in the form of a report on the platform, which the user can view via their terminal.

[0812] Input: Analysis results

[0813] Output: Analysis results displayed in report format

[0814] Step 10: Developing improvement measures

[0815] The user formulates improvement measures for the application based on the review results and analysis reports provided by the server, and reflects these measures in the next development cycle to improve the quality of the application.

[0816] Input: Review results and analysis report

[0817] Output: Improvements

[0818] (Application example 1)

[0819] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0820] Conventional review systems often struggle to quickly obtain detailed evaluations tailored to a target demographic, resulting in problems with the quality and accuracy of reviews. Furthermore, when using human reviewers, the review tasks are time-consuming and costly, making it difficult to obtain efficient feedback. This often prevents users from quickly obtaining suggestions for improvements or product modifications, delaying the product development cycle. The present invention aims to solve these problems.

[0821] 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.

[0822] In this invention, the server includes means for generating virtual reviewers, means for assigning review tasks to the generated virtual reviewers, means for collecting review results from the virtual reviewers, means for analyzing the collected review results and providing them to the user, means for receiving attribute data of the target demographic based on user input and using the attribute data to send prompt sentences to the generative artificial intelligence to generate virtual reviewers, and means for performing detailed product evaluations based on the perspective of the target demographic and analyzing the results to provide insights to the user. This makes it possible to quickly obtain highly accurate reviews tailored to the target demographic.

[0823] A "virtual reviewer" is a virtual entity generated using generative artificial intelligence based on attribute data of a specific target demographic, and is a system that reviews applications and products from the perspective of an actual user.

[0824] A "review task" is an application or product evaluation task assigned to a virtual reviewer, providing detailed feedback based on specific criteria or attributes.

[0825] "Review results" refer to the evaluation information and feedback obtained by the virtual reviewers when they perform the review task, including specific comments about the product's usability, appearance, functionality, etc.

[0826] "Target demographic attribute data" is information used to define a specific target demographic, such as age, gender, interests, and frequency of use. Based on this data, generative AI generates virtual reviewers.

[0827] "Generative AI" is an AI that has the ability to generate virtual reviewers based on the attribute data of the target demographic entered by the user, and is a technology that receives prompt statements and generates an appropriate model.

[0828] A "prompt sentence" is an input sentence that specifies a specific target demographic and evaluation criteria to the generative artificial intelligence, which then generates a virtual reviewer.

[0829] A "detailed evaluation" is an evaluation of a product by a virtual reviewer according to specific criteria, including specific feedback on specific usability, design, performance, etc.

[0830] "Insights" are useful information and suggestions obtained by analyzing collected review results, and serve as guidelines for improving products and adding new features.

[0831] To implement the present invention, it is necessary to operate the system in the following procedure. This system is made up of a server, a terminal, and a user, and the specific procedure is as follows.

[0832] 1. User input and product data upload

[0833] Users use the device interface to input demographic data (e.g., age, gender, interests, frequency of use) of their target demographic and upload product data to be reviewed. This input data is sent to the server and used to generate virtual reviewers.

[0834] 2. Creating a Virtual Reviewer

[0835] The server uses generative artificial intelligence (specifically, OpenAI's GPT-4 model) to generate a virtual reviewer based on the target demographic attribute data it receives. The generative artificial intelligence sends prompts using the input attribute data, and generates a virtual reviewer based on the results.

[0836] Example prompt sentence:

[0837] "Generate a virtual reviewer with the following attributes: age: 20-29, gender: female, purchase frequency: high"

[0838] 3. Assigning and completing review tasks

[0839] The server assigns review tasks to the virtual reviewers, who then perform the tasks and provide detailed evaluations according to specific criteria (e.g., usability, appearance, cost-effectiveness).

[0840] 4. Collection and analysis of review results

[0841] The server collects and analyzes the reviews generated by the virtual reviewers. This analysis uses natural language processing technology based on generative artificial intelligence to extract key issues and improvement suggestions. These results are provided to users in the form of useful insights (e.g., areas for product improvement, the need for new features).

[0842] 5. Formulation of improvement measures

[0843] Users can use their devices to check the review results obtained from the server and formulate measures to improve the product. This feedback is reflected in the next development cycle, contributing to improving the quality of the product.

[0844] Examples:

[0845] For example, consider the case of reviewing product data for a fashion dress aimed at women in their 20s. The user inputs target demographic attribute data, such as "women in their 20s, frequent buyers," and uploads the product data for the fashion dress to the server. The server then generates a virtual reviewer based on this data and has that reviewer evaluate the dress. The review results include the quality of the design, functionality, cost-effectiveness, etc., and the user can use this information to formulate improvements to the product.

[0846] This system allows us to quickly obtain highly accurate reviews tailored to the target demographic, improving the quality and accuracy of reviews while shortening the development cycle. By using these methods, users can effectively evaluate and improve products.

[0847] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0848] Step 1:

[0849] The user uses the device interface to input demographic data for the target demographic and upload product data for review. The input demographic data includes age, gender, interests, and frequency of use, while the product data includes name, price, and description. This data is then sent to the server.

[0850] Input: Target demographic data (e.g., age, gender, interests, frequency of use), product data (e.g., name, price, description)

[0851] Output: Target demographic attribute data and product data sent to the server

[0852] Step 2:

[0853] The server then sends a prompt to the generative AI based on the target demographic attribute data it receives, generating a virtual reviewer. The prompt includes the target demographic attribute data, and a virtual reviewer is generated based on that.

[0854] Input: target demographic attribute data, prompt text

[0855] Output: Generated virtual reviewer

[0856] Step 3:

[0857] The server assigns review tasks to the virtual reviewers, who then perform detailed evaluations based on the uploaded product data and specified criteria (e.g., usability, appearance, and price-performance).

[0858] Input: Virtual reviewer, product data, rating criteria

[0859] Output: Detailed evaluation results

[0860] Step 4:

[0861] The server collects the evaluation results generated by the virtual reviewers and analyzes the data, using natural language processing techniques (e.g., generative artificial intelligence) to extract key issues and improvement suggestions.

[0862] Input: Review results

[0863] Output: Analyzed insights (e.g., key issues, improvement suggestions)

[0864] Step 5:

[0865] The user uses a terminal to check the review results obtained from the server and formulates measures to improve the product. The user then reflects the feedback in the next development cycle.

[0866] Input: Review results (analysis report)

[0867] Output: Improved product development plan

[0868] 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.

[0869] To implement the present invention, a system is used that includes procedures for generating virtual reviewers using generative artificial intelligence, review evaluation using an emotion engine, and collection and analysis of review results. The operation of this system is described in detail below.

[0870] System configuration

[0871] 1. User input and application upload

[0872] The user uses the device interface to input demographic data (e.g., age, gender, interests, and frequency of use) about the target demographic, and uploads the application to be reviewed to the server via the device, allowing the user to send a review request to the server.

[0873] 2. Creating a Virtual Reviewer

[0874] The server receives target demographic attribute data sent by the user and uses generative artificial intelligence to generate a virtual reviewer based on that data. The generated virtual reviewer has the pre-set target demographic attributes and evaluates the application from a perspective based on those attributes.

[0875] 3. Emotion Recognition by Emotion Engine

[0876] The server integrates an emotion engine when assigning review tasks for applications to the virtual reviewers. As the virtual reviewers use the applications, the emotion engine recognizes and records their emotional states in real time. This allows us to quantitatively understand the emotional state of the virtual reviewers.

[0877] 4. Assigning review tasks and conducting reviews

[0878] The server assigns a review task for the application to the generated virtual reviewer. The virtual reviewer uses the application to perform this task and provides a detailed evaluation according to the specified criteria. During the evaluation process, the emotion engine monitors the emotional state of the virtual reviewer and generates emotion analysis data.

[0879] 5. Collection and analysis of review results

[0880] The server collects the review results generated by the virtual reviewers and sentiment analysis data from the sentiment engine, and further analyzes this data to extract meaningful insights for users (e.g., emotional responses to UI usability, sentiment fluctuations for specific features).

[0881] 6. Providing insights and suggesting improvements

[0882] The server generates the analysis results in the form of a report and displays it on the platform. The report includes information that integrates review results and sentiment analysis, providing more meaningful insights for users. Users can review the report and formulate improvements for their applications.

[0883] 7. Formulation of improvement measures

[0884] Users can use their devices to check the reports received from the server and evaluate the detailed review results and sentiment analysis data. This allows users to make specific improvements based on their emotional state (e.g., changing the interface because a specific feature causes discomfort). Based on the feedback obtained, it is possible to plan new features and bug fixes for the next development cycle.

[0885] Specific examples

[0886] For example, when developing a new e-commerce application, a user may specify a target demographic of "women in their 20s who purchase frequently." Virtual reviewers are generated based on these attributes and work with an emotion engine to conduct reviews. Review results include not only evaluations of product display speed and UI, but also associated emotional fluctuations (e.g., stress levels when a product loads slowly).

[0887] After the review is complete, the server collects and analyzes all data and provides a detailed report to the user, who can then refer to the report and develop specific improvement measures that take into account the user's emotional state and incorporate them into the next release.

[0888] In this way, the system of the present invention makes it possible to quickly obtain highly accurate reviews and sentiment analysis specific to the target demographic, thereby improving the quality of applications.

[0889] The processing flow will be explained below.

[0890] Step 1:

[0891] The user opens the interface on their device, enters the demographic data of the target demographic (e.g., age, gender, interests, frequency of use), and clicks the send button once the input is complete.

[0892] Step 2:

[0893] The device temporarily stores the target demographic attribute data entered by the user and prepares to send it to the server.The user also selects the application to be reviewed via the device and uploads the application by clicking the upload button.

[0894] Step 3:

[0895] The server receives the target demographic attribute data and uploaded application files sent from the device, temporarily stores the received data, and organizes it as input data for the generative AI.

[0896] Step 4:

[0897] The server uses generative artificial intelligence to generate a virtual reviewer based on the received target demographic attribute data. The generated virtual reviewer has the specified attributes and evaluates the application from a perspective based on those attributes.

[0898] Step 5:

[0899] The server integrates an emotion engine with the generated virtual reviewer, which recognizes and records the emotional state of the virtual reviewer in real time.

[0900] Step 6:

[0901] The server assigns review tasks for uploaded applications to virtual reviewers, with the review tasks including user-specified viewpoints and evaluation criteria.

[0902] Step 7:

[0903] The virtual reviewer uses the application under review and provides a detailed evaluation according to the specified criteria. The emotion engine monitors the virtual reviewer's emotional state (e.g., stress level, satisfaction) while using the application and generates emotion analysis data.

[0904] Step 8:

[0905] The server collects the review results generated by the virtual reviewers and the sentiment analysis data generated by the sentiment engine. This collected data is temporarily stored and then used for the next analysis step.

[0906] Step 9:

[0907] The server analyzes the collected review results and sentiment analysis data to extract meaningful insights for users (e.g., emotional reactions to UI usability, sentiment fluctuations for specific features). The analysis results are generated in the form of a report and displayed on the platform.

[0908] Step 10:

[0909] The user uses a terminal to check the report retrieved from the server, which contains information that integrates review results and sentiment analysis.

[0910] Step 11:

[0911] Users can formulate improvements to their applications based on the review results and sentiment analysis data they obtain. Specifically, if a particular feature causes discomfort, they can consider specific measures such as changing the interface. The obtained feedback can be reflected in the next development cycle, allowing them to plan new features and bug fixes.

[0912] In this way, the system of the present invention can quickly obtain highly accurate reviews and sentiment analysis specific to the target demographic, effectively improving the quality of applications.

[0913] Example 2

[0914] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0915] Conventional review systems have difficulty evaluating each target demographic and providing detailed feedback, including emotional states. In particular, users lack the means to create virtual reviewers based on target attributes and collect and analyze their review results along with emotional data, making it difficult to gain insights needed to improve software under development.

[0916] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for receiving attribute data of the target demographic based on user input and generating a virtual reviewer that matches the attribute data, means for assigning a review task to the generated virtual reviewer, means for integrating an emotion engine and recognizing and recording the emotional state of the virtual reviewer in real time, means for collecting review results and emotion analysis data from the virtual reviewer, and means for analyzing the collected review results and emotion analysis data and providing them to the user. This enables highly accurate reviews and emotion analysis tailored to the target demographic.

[0917] "Target demographic attribute data" is information for defining a specific target user group, such as the user's age, gender, interests, and frequency of use.

[0918] A "virtual reviewer" is a fictitious reviewer with specific target demographic attributes, generated using a generative artificial intelligence model.

[0919] A "review task" is a specific instruction and task for evaluating and providing feedback on the software or application being reviewed.

[0920] An "emotion engine" is software or algorithms for recognizing and recording the emotional state of a virtual reviewer in real time.

[0921] "Sentiment Analysis Data" is information about the emotional state of a hypothetical reviewer recorded by the emotion engine, and is a record of emotions that fluctuate during the course of the review.

[0922] A "generative AI model" is an artificial intelligence model that uses natural language processing and machine learning techniques to generate new data and information (in this case, virtual reviewers).

[0923] A "prompt" is an instruction or input to a generative AI model to obtain a specific output.

[0924] "Review evaluation data" refers to evaluation information collected by virtual reviewers while performing a review task, and is detailed feedback on each function and interface of an application.

[0925] The present invention is a system that includes procedures for generating virtual reviewers using generative artificial intelligence, review evaluation using an emotion engine, and collection and analysis of review results. This system collects attribute data of the target demographic based on user input, generates virtual reviewers based on that data, and conducts reviews. In the process, it collects emotion data using an emotion engine, and finally analyzes this data and provides it to the user as a report. The details of this system are described below.

[0926] First, the user uses the device interface to input the target demographic's attribute data (e.g., age, gender, interests, frequency of use, etc.), then uploads the software to be reviewed to the server via the device and submits a review request.

[0927] The server receives the target demographic attribute data sent by the user and generates a virtual reviewer using a generative AI model (e.g., GPT-3 or BERT) based on that data. In this process, a prompt sentence is generated for the generative AI model and input into the model to generate a virtual reviewer.

[0928] The server then assigns review tasks to the virtual reviewers, and integrates an emotion engine (e.g., Microsoft Azure's emotion recognition API) to recognize and record the virtual reviewers' emotional states in real time.

[0929] The virtual reviewer uses the software under review to perform a detailed evaluation according to the specified criteria, during which the emotion engine monitors the reviewer's emotional state and generates sentiment analysis data.

[0930] The server collects the review results generated by the virtual reviewers and the sentiment analysis data generated by the sentiment engine, and organizes and analyzes this data using analytical tools (e.g., Python's Pandas and NumPy libraries) to extract meaningful insights for users.

[0931] Finally, the server generates a report of the analysis results and displays it on the platform. Users can then use their devices to review the report and formulate specific improvement measures based on the detailed review results and sentiment analysis data, which can then be incorporated into the next development cycle.

[0932] As a concrete example, consider the development of a new e-commerce application. The user inputs demographic data such as "women in their 20s who purchase frequently" as the target demographic, and generates a virtual reviewer. This virtual reviewer reviews the application together with an emotion engine. The review results include evaluations of the product display speed and UI, as well as emotional states such as stress felt when products are slow to load.

[0933] Example prompt sentence:

[0934] The target users are women in their 20s who purchase frequently. Please generate a virtual reviewer based on this profile and have them rate the following application.

[0935] In this way, the system of the present invention can quickly perform highly accurate reviews and sentiment analysis specific to the target demographic, allowing users to efficiently improve the quality of their applications.

[0936] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0937] Step 1: User Input and Application Upload

[0938] The user uses the device interface to input demographic data (age, gender, interests, frequency of use, etc.) of the target demographic. The input data is sent to the server in JSON format, which becomes the input.

[0939] The user selects the software to be reviewed (e.g., APK file) and clicks the upload button. The software file is uploaded to the server, which is the output.

[0940] Step 2: Receive demographic data for your target audience

[0941] The server receives the HTTP request, analyzes the target demographic attribute data sent by the user, and saves the received JSON data in an internal database. This becomes the input.

[0942] The attribute data is converted into a format suitable for generative AI models, which then becomes the output.

[0943] Step 3: Generate virtual reviewers

[0944] The server uses a generative AI model (e.g., GPT-3 or BERT) to generate a virtual reviewer. Specifically, it creates a prompt and inputs it into the generative AI model. The prompt is, "The target user is a woman in her 20s who frequently purchases. Please generate a virtual reviewer based on this profile and rate the following application." This is the input.

[0945] The generative AI model generates a virtual reviewer and returns that data to the server, which becomes the output.

[0946] Step 4: Assigning review tasks and integrating the sentiment engine

[0947] The server assigns review tasks for uploaded software to the virtual reviewers, which is the input.

[0948] The server integrates an emotion engine (e.g., Microsoft Azure's emotion recognition API) to recognize and record the emotional state of the virtual reviewer in real time. This is the specific operation. It configures the settings to call the emotion recognition API and sends an API request for each reviewer operation. It receives the response from the API and records the emotional data, which becomes the output.

[0949] Step 5: Performing review tasks

[0950] The virtual reviewers interact with the software according to the assigned review task and provide detailed evaluations based on the specified criteria, which serve as input.

[0951] During operation, the emotion engine monitors the emotional state and generates evaluation data and emotion analysis data, which are the output.

[0952] Step 6: Collect review results and sentiment analysis data

[0953] The server receives review rating data and emotional state data sent by virtual reviewers, which are the input.

[0954] This data is stored in a database and organized and analyzed using analytical tools (e.g., Python's Pandas or NumPy library). This is the specific operation, and the analyzed data is the output.

[0955] Step 7: Generate and serve reports

[0956] The server generates a report based on the analysis results. Specifically, it creates a detailed report by applying the analyzed data to a template. This is the input.

[0957] The generated report is displayed on the user interface and made accessible to the user - this is the output.

[0958] Step 8: Developing improvement measures

[0959] The user uses a terminal to check the report retrieved from the server, which becomes the input.

[0960] Users evaluate the detailed review results and sentiment analysis data and develop specific improvement measures. This is the output. Improvement measures may include, for example, changing the UI design or adding or removing features.

[0961] In this way, the user, server, and device work together at each processing step to achieve highly accurate reviews and sentiment analysis tailored to the target demographic.

[0962] (Application example 2)

[0963] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0964] Conventional review systems rely on user-provided feedback, making it difficult to accurately grasp emotional fluctuations and real-time emotional states. There was also a need for a method to quickly and accurately obtain reviews that are optimal for the target demographic.

[0965] 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 virtual reviewers, means for assigning review tasks to the generated virtual reviewers, means for collecting review results from the virtual reviewers, means for analyzing the collected review results and providing them to users, means for collecting and analyzing emotional data in real time, and means for monitoring and recording the emotional state of users from their gaze, facial expressions, etc. This makes it possible to quickly obtain highly accurate reviews and emotional analysis tailored to the target demographic and efficiently propose improvement measures for applications and products.

[0966] A "virtual reviewer" is a simulated subject for conducting a review, generated using generative artificial intelligence based on attribute data of a specific target demographic.

[0967] A "review task" is a series of tasks assigned to a virtual reviewer to evaluate a particular application or product and record the results.

[0968] "Emotional data" is data used to record and analyze changes in the user's emotional state in real time based on information such as their gaze, facial expressions, and voice.

[0969] "Means for collecting and analyzing emotional data in real time" refers to technology that continuously monitors the user's gaze, facial expressions, etc., and records and analyzes their emotional state on the spot.

[0970] "Generative AI" is an AI model that generates responses and evaluations for specific tasks based on input attribute data.

[0971] "Means for monitoring and recording emotional states from gaze, facial expressions, etc." refers to technology that detects a user's gaze and facial expressions through smart glasses or the like when the user is browsing an online shopping site, and collects and stores that emotional state in real time.

[0972] A "target demographic" is a group of users with specific attributes such as age, gender, interests, and frequency of use, and is the group that is analyzed to evaluate and improve products and services.

[0973] To implement the present invention, the following system configuration and processing procedures are used: Specifically, review evaluation using a virtual reviewer and an emotion engine, and review result collection and analysis procedures.

[0974] System configuration

[0975] 1. User input and application upload

[0976] Users use the device interface to input demographic data (age, gender, interests, frequency of use, etc.) about their target demographic, and also upload the application to be reviewed to the server via the device, allowing the user to send a review request to the server.

[0977] 2. Creating a Virtual Reviewer

[0978] The server receives target demographic attribute data sent by the user and uses generative artificial intelligence to generate a virtual reviewer based on that data. The generated virtual reviewer has the pre-set target demographic attributes and evaluates the application from a perspective based on those attributes.

[0979] 3. Emotion Recognition by Emotion Engine

[0980] The server integrates an emotion engine when assigning review tasks for applications to the virtual reviewers. As the virtual reviewers use the applications, the emotion engine recognizes and records their emotional states in real time. This allows us to quantitatively understand the emotional state of the virtual reviewers.

[0981] 4. Assigning review tasks and conducting reviews

[0982] The server assigns a review task for the application to the generated virtual reviewer. The virtual reviewer uses the application to perform this task and provides a detailed evaluation according to the specified criteria. During the evaluation process, the emotion engine monitors the emotional state of the virtual reviewer and generates emotion analysis data.

[0983] 5. Collection and analysis of review results

[0984] The server collects the review results generated by the virtual reviewers and sentiment analysis data from the sentiment engine, and further analyzes this data to extract meaningful insights for users (e.g., emotional responses to the usability of the user interface, sentiment fluctuations for specific features).

[0985] Specific examples

[0986] For example, consider a female user in her twenties searching for new clothes on an online shopping site. The user puts on smart glasses and launches a dedicated app. The emotion engine collects real-time emotional data from the user's gaze and facial expressions, recording their emotional state. At the same time, a virtual reviewer using generative artificial intelligence evaluates the same product page. As a result, the emotional data of the user and the virtual reviewer is sent to a server for analysis.

[0987] The analysis results are provided to users in the form of a report, which includes specific improvement measures based on the sentiment data. For example, the report may include insights such as "users feel frustrated when products take a long time to load" or "certain designs are particularly well-received."

[0988] Prompt Sentence Examples

[0989] The following prompt sentence is input into the generative artificial intelligence:

[0990] Generate virtual reviewers based on the following user attributes:

[0991] User Attributes:

[0992] Age: 20s

[0993] Gender: Female

[0994] Interests: Fashion, shopping

[0995] Frequency of use: High

[0996] We assign a hypothetical reviewer the task of rating a clothing product page on an online shopping website at the following URL:

[0997] URL: [example.com / fashion / clothing / item1234]

[0998] Output your impressions and emotional data.

[0999] This allows for quick and accurate target-specific reviews and sentiment analysis, which can be used to improve the quality of applications and products.

[1000] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1001] Step 1:

[1002] The user uses the device interface to input demographic data (age, gender, interests, frequency of use, etc.) of the target demographic. The application to be reviewed is then uploaded to the server via the device. This input data is sent to the server and used as the basis for generating virtual reviewers.

[1003] Step 2:

[1004] The server receives the target demographic attribute data sent by the user and generates a virtual reviewer based on that data using generative AI. Specifically, the received attribute data is input into the generative AI model as a prompt sentence, generating an appropriate virtual reviewer.

[1005] Step 3:

[1006] The server assigns a review task for the application to the virtual reviewer, who then evaluates the application according to the specified criteria. During this process, the virtual reviewer's evaluation is recorded.

[1007] Step 4:

[1008] The server integrates an emotion engine to collect real-time emotional data from users' gazes and facial expressions. When users browse online shopping sites through smart glasses, emotion data is recorded in real time using eye tracking and facial recognition technology. This data is sent to the server and integrated with the evaluation data from the virtual reviewers.

[1009] Step 5:

[1010] The server collects the review results generated by the virtual reviewers and sentiment analysis data from the emotion engine. This data includes the user's emotional fluctuations and eye movements, forming detailed user experience data. This allows for comprehensive review results.

[1011] Step 6:

[1012] The server analyzes the collected data and extracts meaningful insights for users. Using data analysis tools such as Apache Hadoop and Apache Spark, the data is processed and calculated to obtain insights such as sentiment fluctuations and the usability of the user interface. Based on this, improvement measures are proposed.

[1013] Step 7:

[1014] The user uses a terminal to check the report obtained from the server. This report integrates sentiment analysis data and review results and includes specific improvement measures. The user then formulates an improvement plan for the application or product based on the report.

[1015] In this way, by carrying out specific processing at each step, it is possible to quickly obtain highly accurate reviews and sentiment analysis specific to the target demographic, thereby improving the quality of applications and products.

[1016] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1017] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1018] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1019] [Fourth embodiment]

[1020] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1021] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1022] 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).

[1023] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1024] 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.

[1025] 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).

[1026] 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.

[1027] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1028] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1029] 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.

[1030] 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.

[1031] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1032] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1033] To implement the present invention, a system is used that includes procedures for generating virtual reviewers using generative artificial intelligence, assigning review tasks, and collecting and analyzing review results. The operation of this system is specifically described below.

[1034] System configuration

[1035] 1. User input and application upload

[1036] The user uses the device interface to input demographic data (e.g., age, gender, interests, and frequency of use) about the target demographic, and uploads the application to be reviewed to the server via the device, allowing the user to send a review request to the server.

[1037] 2. Creating a Virtual Reviewer

[1038] The server receives target demographic attribute data sent by the user and uses generative artificial intelligence to generate a virtual reviewer based on that data. The generated virtual reviewer has the pre-set target demographic attributes and evaluates the application from a perspective based on those attributes.

[1039] 3. Assigning review tasks and conducting reviews

[1040] The server assigns the task of reviewing the uploaded application to the generated virtual reviewer, who then uses the application to perform this task and provides a detailed evaluation according to the specified criteria.

[1041] 4. Collection and analysis of review results

[1042] The server collects the review results generated by the virtual reviewers, analyzes them further, and extracts meaningful insights for users (e.g., UI improvements, performance bottlenecks). The analysis results are displayed in the form of a report on the platform and provided to users.

[1043] 5. Formulation of improvement measures

[1044] Users can use their devices to check the review results received from the server, and based on the feedback they receive, they can formulate improvements to the application and incorporate them into the next development cycle, thereby continuously improving the quality of the application.

[1045] Specific examples

[1046] For example, consider the development of an e-commerce application with new features. The development team specifies the target demographic as "women in their 20s who purchase frequently" and generates virtual reviewers. The virtual reviewers use the application from a perspective based on the specified target demographic and provide detailed feedback on features such as ease of use and UI design.

[1047] After the review is complete, the server collects the review results and generates an analysis report. This report includes major issues and improvement suggestions. The user (development team) refers to this report and formulates improvement measures for new functions. This enables high-quality application releases in a short period of time.

[1048] This invention significantly reduces the time and cost required for conventional QA processes and enables the rapid acquisition of highly accurate reviews tailored to the target demographic.

[1049] The processing flow will be explained below.

[1050] Step 1:

[1051] The user opens the interface on their device, enters the demographic data of the target demographic (e.g., age, gender, interests, frequency of use), and clicks the send button once the input is complete.

[1052] Step 2:

[1053] The device temporarily stores the target demographic attribute data entered by the user and prepares to send it to the server. At the same time, the user selects the application to be reviewed (e.g., APK file) and clicks the upload button to begin uploading the application.

[1054] Step 3:

[1055] The server receives the target demographic attribute data and uploaded application files sent from the device. The received data is temporarily stored and organized as input data for the generative AI.

[1056] Step 4:

[1057] The server uses generative artificial intelligence based on the received target demographic attribute data to generate a virtual reviewer that matches the specified attributes. This virtual reviewer is a virtual user with the specific attributes entered by the user.

[1058] Step 5:

[1059] The server assigns a review task for the application to be reviewed to the generated virtual reviewer, which includes viewpoints and evaluation criteria specified by the user.

[1060] Step 6:

[1061] The virtual reviewer uses the application under review in a simulated environment and evaluates it according to specified criteria, such as the ease of use of the UI, the intuitiveness of the features, and performance.

[1062] Step 7:

[1063] After completing the review task, the virtual reviewers send the generated evaluation results and feedback to the server, which includes detailed evaluation comments and suggestions for improvement.

[1064] Step 8:

[1065] The server collects the review results received from the virtual reviewers and organizes them as data for analysis. The server analyzes the collected data and extracts insights that are meaningful to the user.

[1066] Step 9:

[1067] The server generates the analysis results in the form of a report and displays it on the platform, and sends a notification to the user informing them that the review results are available for access.

[1068] Step 10:

[1069] Users access the server via their terminals and check the review results displayed on the platform. Users can view detailed reports and evaluate the feedback of each virtual reviewer.

[1070] Step 11:

[1071] The user will then formulate improvements to the application based on the review results obtained, and will incorporate this data into the next development cycle to create plans for new features and bug fixes.

[1072] In this way, the entire processing flow works together efficiently, allowing users to quickly obtain highly accurate reviews specific to their target demographic and improve the quality of their applications.

[1073] Example 1

[1074] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1075] The conventional evaluation process had problems in that it was difficult to quickly obtain highly accurate reviews specific to a target demographic, and the reviews were time-consuming and costly. In particular, in order to collect feedback from different target demographics, it was necessary to arrange for multiple reviewers to conduct the evaluations, which resulted in a significant burden of cost and time.

[1076] 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.

[1077] In this invention, the server includes means for receiving target demographic attribute data and the program to be reviewed based on user input, means for using a generative artificial intelligence model to generate virtual reviewers that match the received target demographic attribute data, means for assigning review tasks to the generated virtual reviewers and having them perform detailed evaluations of the program, means for collecting review results from the virtual reviewers and analyzing the results, and means for providing the analysis results to the user. This makes it possible to quickly obtain highly accurate reviews tailored to the target demographic, significantly reducing the time and cost required for conventional evaluation processes.

[1078] "User input" refers to a person or entity using the system providing demographic data and programs to be reviewed to the system.

[1079] "Target demographic attribute data" refers to information that indicates the characteristics of the target user demographic when writing a review (e.g., age, gender, interests, frequency of use).

[1080] "Reviewed Program" means software or an application that is subject to evaluation and review.

[1081] A "virtual reviewer" is a reviewer simulated based on specific attribute data using a generative artificial intelligence model.

[1082] A "generative artificial intelligence model" is an artificial intelligence algorithm that simulates human opinions and behaviors based on specific prompts.

[1083] A "review task" is a task that includes specific evaluation activities and checkpoints assigned to a virtual reviewer.

[1084] "Review results" are the feedback and evaluation scores obtained when a virtual reviewer performs a review task.

[1085] "Analysis results" are the results of further analyzing the collected review results and extracting insights that are useful to users (e.g., areas for improvement in the UI, performance bottlenecks).

[1086] "Display in report format" means that the analysis results are provided in a document or table format that is easy for the user to understand.

[1087] A "prompt sentence" is an input sentence given to a generative artificial intelligence model, and is a sentence that includes instructions for evaluation based on specific attributes and evaluation criteria.

[1088] MODE FOR CARRYING OUT THE INVENTION

[1089] To implement the present invention, a system is used in which a user inputs attribute data of the target demographic and the program to be reviewed, and a virtual reviewer is generated based on the input data using generative artificial intelligence. A specific implementation method of this system is described below.

[1090] User input and application upload

[1091] Users use their own devices to input demographic data (e.g., age, gender, interests, and frequency of use) about their target demographic. This data is important information for reflecting the characteristics of the target demographic. Users also upload the program (e.g., application) to be reviewed to the server via their devices. This sends a review request to the server in response to the user's request.

[1092] Virtual reviewer generation

[1093] The server receives the target demographic attribute data sent by the user and generates a virtual reviewer based on that data. The virtual reviewer is generated using a generative artificial intelligence model (e.g., GPT-3). By inputting the target demographic profile into this model, a virtual reviewer with specific attributes is generated.

[1094] Assigning review tasks and conducting reviews

[1095] The server assigns the uploaded program review task to the generated virtual reviewer, who then performs a detailed evaluation according to specified criteria (e.g., ease of use of functions, UI design, performance). The virtual reviewer operates the program in an AI simulation environment and performs the evaluation through simulated user operations.

[1096] Collection and analysis of review results

[1097] The server collects and analyzes the reviews generated by the virtual reviewers. The review results include specific feedback and rating scores that users may find interesting. The server further analyzes this data to extract useful insights for users. The analysis results are displayed on the platform in the form of a report.

[1098] Formulating improvement measures

[1099] Users can use their devices to check the review results and analysis reports provided by the server, and can formulate improvements to the application based on this feedback and incorporate them into the next development cycle, thereby continuously improving the quality of the application.

[1100] Specific examples

[1101] For example, consider the development of an e-commerce application with new features. The development team (users) specify the target demographic as "women in their 20s with high purchasing frequency" and generate a virtual reviewer. The virtual reviewer will use the application from a perspective based on the specified attributes and provide detailed feedback on features such as ease of use and UI design.

[1102] Example prompts used to generate virtual reviewers:

[1103] "Please rate in detail the ease of use of the new features for women in their 20s."

[1104] "Please review in detail how users who frequently use SNS experience the UI."

[1105] After the review is completed, the server collects the review results and generates an analysis report, which includes major issues and improvement suggestions. The development team (users) can refer to this report and quickly develop improvement plans for new features.

[1106] This invention significantly reduces the time and cost required for the conventional evaluation process, making it possible to quickly obtain highly accurate reviews specific to the target demographic.

[1107] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1108] Step 1: User Input

[1109] The user uses the device interface to input attribute data of the target demographic (e.g., age, gender, interests, frequency of use). Specifically, the user uses forms and check boxes to input data such as "20s," "female," and "frequent SNS user." The input data is organized within the device and sent to the server.

[1110] Input: Target demographic data (e.g., age, gender, interests, frequency of use)

[1111] Output: Attribute data sent to the server

[1112] Step 2: Upload your application

[1113] The user uploads the application to be reviewed to the server via their device. Specifically, the user selects the specified application file (e.g., apk file or ipa file) from their device and clicks the "Upload" button. The file is then transferred from the device to the server.

[1114] Input: Application file to be reviewed

[1115] Output: Application files uploaded to the server

[1116] Step 3: Receive target data

[1117] The server receives the target demographic attribute data sent by the user. The received data is stored in a database and used as a target demographic profile. Data integrity checks are also performed at this stage.

[1118] Input: Target demographic data submitted by the user

[1119] Output: Attribute data stored in a database

[1120] Step 4: Generate virtual reviewers

[1121] The server generates virtual reviewers using a generative artificial intelligence model (e.g., GPT-3). By inputting the target demographic profile into a prompt statement (e.g., "Please generate a virtual reviewer for a woman in her 20s"), the AI ​​generates a virtual reviewer with specific attributes.

[1122] Input: Target demographic profile

[1123] Output: Generated virtual reviewer

[1124] Step 5: Assign review tasks

[1125] The server assigns review tasks for uploaded applications to the generated virtual reviewers, including evaluation criteria (e.g., usability, design, performance).

[1126] Input: Generated virtual reviewer, application to be reviewed

[1127] Output: Tasks assigned to virtual reviewers

[1128] Step 6: Conducting the review

[1129] The virtual reviewer interacts with the application in a simulated environment to perform the assigned task: interacting with the UI, testing each feature, and recording their evaluation.

[1130] Input: Review task, Simulation environment

[1131] Output: Review results from the virtual reviewer

[1132] Step 7: Collect review results

[1133] The server collects the review results generated by the virtual reviewers, which are then stored in a database and treated as a dataset for later analysis.

[1134] Input: Review results from a hypothetical reviewer

[1135] Output: Review results stored in a database

[1136] Step 8: Analyze the results

[1137] The server analyzes the collected reviews and extracts useful insights for users. Specifically, it uses data analysis algorithms to identify key issues and improvement suggestions.

[1138] Input: Collected review results

[1139] Output: Analysis results (insights and improvement suggestions)

[1140] Step 9: View the results

[1141] The server displays the analysis results in the form of a report on the platform, which the user can view via their terminal.

[1142] Input: Analysis results

[1143] Output: Analysis results displayed in report format

[1144] Step 10: Developing improvement measures

[1145] The user formulates improvement measures for the application based on the review results and analysis reports provided by the server, and reflects these measures in the next development cycle to improve the quality of the application.

[1146] Input: Review results and analysis report

[1147] Output: Improvements

[1148] (Application example 1)

[1149] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1150] Conventional review systems often struggle to quickly obtain detailed evaluations tailored to a target demographic, resulting in problems with the quality and accuracy of reviews. Furthermore, when using human reviewers, the review tasks are time-consuming and costly, making it difficult to obtain efficient feedback. This often prevents users from quickly obtaining suggestions for improvements or product modifications, delaying the product development cycle. The present invention aims to solve these problems.

[1151] 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.

[1152] In this invention, the server includes means for generating virtual reviewers, means for assigning review tasks to the generated virtual reviewers, means for collecting review results from the virtual reviewers, means for analyzing the collected review results and providing them to the user, means for receiving attribute data of the target demographic based on user input and using the attribute data to send prompt sentences to the generative artificial intelligence to generate virtual reviewers, and means for performing detailed product evaluations based on the perspective of the target demographic and analyzing the results to provide insights to the user. This makes it possible to quickly obtain highly accurate reviews tailored to the target demographic.

[1153] A "virtual reviewer" is a virtual entity generated using generative artificial intelligence based on attribute data of a specific target demographic, and is a system that reviews applications and products from the perspective of an actual user.

[1154] A "review task" is an application or product evaluation task assigned to a virtual reviewer, providing detailed feedback based on specific criteria or attributes.

[1155] "Review results" refer to the evaluation information and feedback obtained by the virtual reviewers when they perform the review task, including specific comments about the product's usability, appearance, functionality, etc.

[1156] "Target demographic attribute data" is information used to define a specific target demographic, such as age, gender, interests, and frequency of use. Based on this data, generative AI generates virtual reviewers.

[1157] "Generative AI" is an AI that has the ability to generate virtual reviewers based on the attribute data of the target demographic entered by the user, and is a technology that receives prompt statements and generates an appropriate model.

[1158] A "prompt sentence" is an input sentence that specifies a specific target demographic and evaluation criteria to the generative artificial intelligence, which then generates a virtual reviewer.

[1159] A "detailed evaluation" is an evaluation of a product by a virtual reviewer according to specific criteria, including specific feedback on specific usability, design, performance, etc.

[1160] "Insights" are useful information and suggestions obtained by analyzing collected review results, and serve as guidelines for improving products and adding new features.

[1161] To implement the present invention, it is necessary to operate the system in the following procedure. This system is made up of a server, a terminal, and a user, and the specific procedure is as follows.

[1162] 1. User input and product data upload

[1163] Users use the device interface to input demographic data (e.g., age, gender, interests, frequency of use) of their target demographic and upload product data to be reviewed. This input data is sent to the server and used to generate virtual reviewers.

[1164] 2. Creating a Virtual Reviewer

[1165] The server uses generative artificial intelligence (specifically, OpenAI's GPT-4 model) to generate a virtual reviewer based on the target demographic attribute data it receives. The generative artificial intelligence sends prompts using the input attribute data, and generates a virtual reviewer based on the results.

[1166] Example prompt sentence:

[1167] "Generate a virtual reviewer with the following attributes: age: 20-29, gender: female, purchase frequency: high"

[1168] 3. Assigning and completing review tasks

[1169] The server assigns review tasks to the virtual reviewers, who then perform the tasks and provide detailed evaluations according to specific criteria (e.g., usability, appearance, cost-effectiveness).

[1170] 4. Collection and analysis of review results

[1171] The server collects and analyzes the reviews generated by the virtual reviewers. This analysis uses natural language processing technology based on generative artificial intelligence to extract key issues and improvement suggestions. These results are provided to users in the form of useful insights (e.g., areas for product improvement, the need for new features).

[1172] 5. Formulation of improvement measures

[1173] Users can use their devices to check the review results obtained from the server and formulate measures to improve the product. This feedback is reflected in the next development cycle, contributing to improving the quality of the product.

[1174] Examples:

[1175] For example, consider the case of reviewing product data for a fashion dress aimed at women in their 20s. The user inputs target demographic attribute data, such as "women in their 20s, frequent buyers," and uploads the product data for the fashion dress to the server. The server then generates a virtual reviewer based on this data and has that reviewer evaluate the dress. The review results include the quality of the design, functionality, cost-effectiveness, etc., and the user can use this information to formulate improvements to the product.

[1176] This system allows us to quickly obtain highly accurate reviews tailored to the target demographic, improving the quality and accuracy of reviews while shortening the development cycle. By using these methods, users can effectively evaluate and improve products.

[1177] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1178] Step 1:

[1179] The user uses the device interface to input demographic data for the target demographic and upload product data for review. The input demographic data includes age, gender, interests, and frequency of use, while the product data includes name, price, and description. This data is then sent to the server.

[1180] Input: Target demographic data (e.g., age, gender, interests, frequency of use), product data (e.g., name, price, description)

[1181] Output: Target demographic attribute data and product data sent to the server

[1182] Step 2:

[1183] The server then sends a prompt to the generative AI based on the target demographic attribute data it receives, generating a virtual reviewer. The prompt includes the target demographic attribute data, and a virtual reviewer is generated based on that.

[1184] Input: target demographic attribute data, prompt text

[1185] Output: Generated virtual reviewer

[1186] Step 3:

[1187] The server assigns review tasks to the virtual reviewers, who then perform detailed evaluations based on the uploaded product data and specified criteria (e.g., usability, appearance, and price-performance).

[1188] Input: Virtual reviewer, product data, rating criteria

[1189] Output: Detailed evaluation results

[1190] Step 4:

[1191] The server collects the evaluation results generated by the virtual reviewers and analyzes the data, using natural language processing techniques (e.g., generative artificial intelligence) to extract key issues and improvement suggestions.

[1192] Input: Review results

[1193] Output: Analyzed insights (e.g., key issues, improvement suggestions)

[1194] Step 5:

[1195] The user uses a terminal to check the review results obtained from the server and formulates measures to improve the product. The user then reflects the feedback in the next development cycle.

[1196] Input: Review results (analysis report)

[1197] Output: Improved product development plan

[1198] 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.

[1199] To implement the present invention, a system is used that includes procedures for generating virtual reviewers using generative artificial intelligence, review evaluation using an emotion engine, and collection and analysis of review results. The operation of this system is described in detail below.

[1200] System configuration

[1201] 1. User input and application upload

[1202] The user uses the device interface to input demographic data (e.g., age, gender, interests, and frequency of use) about the target demographic, and uploads the application to be reviewed to the server via the device, allowing the user to send a review request to the server.

[1203] 2. Creating a Virtual Reviewer

[1204] The server receives target demographic attribute data sent by the user and uses generative artificial intelligence to generate a virtual reviewer based on that data. The generated virtual reviewer has the pre-set target demographic attributes and evaluates the application from a perspective based on those attributes.

[1205] 3. Emotion Recognition by Emotion Engine

[1206] The server integrates an emotion engine when assigning review tasks for applications to the virtual reviewers. As the virtual reviewers use the applications, the emotion engine recognizes and records their emotional states in real time. This allows us to quantitatively understand the emotional state of the virtual reviewers.

[1207] 4. Assigning review tasks and conducting reviews

[1208] The server assigns a review task for the application to the generated virtual reviewer. The virtual reviewer uses the application to perform this task and provides a detailed evaluation according to the specified criteria. During the evaluation process, the emotion engine monitors the emotional state of the virtual reviewer and generates emotion analysis data.

[1209] 5. Collection and analysis of review results

[1210] The server collects the review results generated by the virtual reviewers and sentiment analysis data from the sentiment engine, and further analyzes this data to extract meaningful insights for users (e.g., emotional responses to UI usability, sentiment fluctuations for specific features).

[1211] 6. Providing insights and suggesting improvements

[1212] The server generates the analysis results in the form of a report and displays it on the platform. The report includes information that integrates review results and sentiment analysis, providing more meaningful insights for users. Users can review the report and formulate improvements for their applications.

[1213] 7. Formulation of improvement measures

[1214] Users can use their devices to check the reports received from the server and evaluate the detailed review results and sentiment analysis data. This allows users to make specific improvements based on their emotional state (e.g., changing the interface because a specific feature causes discomfort). Based on the feedback obtained, it is possible to plan new features and bug fixes for the next development cycle.

[1215] Specific examples

[1216] For example, when developing a new e-commerce application, a user may specify a target demographic of "women in their 20s who purchase frequently." Virtual reviewers are generated based on these attributes and work with an emotion engine to conduct reviews. Review results include not only evaluations of product display speed and UI, but also associated emotional fluctuations (e.g., stress levels when a product loads slowly).

[1217] After the review is complete, the server collects and analyzes all data and provides a detailed report to the user, who can then refer to the report and develop specific improvement measures that take into account the user's emotional state and incorporate them into the next release.

[1218] In this way, the system of the present invention makes it possible to quickly obtain highly accurate reviews and sentiment analysis specific to the target demographic, thereby improving the quality of applications.

[1219] The processing flow will be explained below.

[1220] Step 1:

[1221] The user opens the interface on their device, enters the demographic data of the target demographic (e.g., age, gender, interests, frequency of use), and clicks the send button once the input is complete.

[1222] Step 2:

[1223] The device temporarily stores the target demographic attribute data entered by the user and prepares to send it to the server.The user also selects the application to be reviewed via the device and uploads the application by clicking the upload button.

[1224] Step 3:

[1225] The server receives the target demographic attribute data and uploaded application files sent from the device, temporarily stores the received data, and organizes it as input data for the generative AI.

[1226] Step 4:

[1227] The server uses generative artificial intelligence to generate a virtual reviewer based on the received target demographic attribute data. The generated virtual reviewer has the specified attributes and evaluates the application from a perspective based on those attributes.

[1228] Step 5:

[1229] The server integrates an emotion engine with the generated virtual reviewer, which recognizes and records the emotional state of the virtual reviewer in real time.

[1230] Step 6:

[1231] The server assigns review tasks for uploaded applications to virtual reviewers, with the review tasks including user-specified viewpoints and evaluation criteria.

[1232] Step 7:

[1233] The virtual reviewer uses the application under review and provides a detailed evaluation according to the specified criteria. The emotion engine monitors the virtual reviewer's emotional state (e.g., stress level, satisfaction) while using the application and generates emotion analysis data.

[1234] Step 8:

[1235] The server collects the review results generated by the virtual reviewers and the sentiment analysis data generated by the sentiment engine. This collected data is temporarily stored and then used for the next analysis step.

[1236] Step 9:

[1237] The server analyzes the collected review results and sentiment analysis data to extract meaningful insights for users (e.g., emotional reactions to UI usability, sentiment fluctuations for specific features). The analysis results are generated in the form of a report and displayed on the platform.

[1238] Step 10:

[1239] The user uses a terminal to check the report retrieved from the server, which contains information that integrates review results and sentiment analysis.

[1240] Step 11:

[1241] Users can formulate improvements to their applications based on the review results and sentiment analysis data they obtain. Specifically, if a particular feature causes discomfort, they can consider specific measures such as changing the interface. The obtained feedback can be reflected in the next development cycle, allowing them to plan new features and bug fixes.

[1242] In this way, the system of the present invention can quickly obtain highly accurate reviews and sentiment analysis specific to the target demographic, effectively improving the quality of applications.

[1243] Example 2

[1244] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1245] Conventional review systems have difficulty evaluating each target demographic and providing detailed feedback, including emotional states. In particular, users lack the means to create virtual reviewers based on target attributes and collect and analyze their review results along with emotional data, making it difficult to gain insights needed to improve software under development.

[1246] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for receiving attribute data of the target demographic based on user input and generating a virtual reviewer that matches the attribute data, means for assigning a review task to the generated virtual reviewer, means for integrating an emotion engine and recognizing and recording the emotional state of the virtual reviewer in real time, means for collecting review results and emotion analysis data from the virtual reviewer, and means for analyzing the collected review results and emotion analysis data and providing them to the user. This enables highly accurate reviews and emotion analysis tailored to the target demographic.

[1247] "Target demographic attribute data" is information for defining a specific target user group, such as the user's age, gender, interests, and frequency of use.

[1248] A "virtual reviewer" is a fictitious reviewer with specific target demographic attributes, generated using a generative artificial intelligence model.

[1249] A "review task" is a specific instruction and task for evaluating and providing feedback on the software or application being reviewed.

[1250] An "emotion engine" is software or algorithms for recognizing and recording the emotional state of a virtual reviewer in real time.

[1251] "Sentiment Analysis Data" is information about the emotional state of a hypothetical reviewer recorded by the emotion engine, and is a record of emotions that fluctuate during the course of the review.

[1252] A "generative AI model" is an artificial intelligence model that uses natural language processing and machine learning techniques to generate new data and information (in this case, virtual reviewers).

[1253] A "prompt" is an instruction or input to a generative AI model to obtain a specific output.

[1254] "Review evaluation data" refers to evaluation information collected by virtual reviewers while performing a review task, and is detailed feedback on each function and interface of an application.

[1255] The present invention is a system that includes procedures for generating virtual reviewers using generative artificial intelligence, review evaluation using an emotion engine, and collection and analysis of review results. This system collects attribute data of the target demographic based on user input, generates virtual reviewers based on that data, and conducts reviews. In the process, it collects emotion data using an emotion engine, and finally analyzes this data and provides it to the user as a report. The details of this system are described below.

[1256] First, the user uses the device interface to input the target demographic's attribute data (e.g., age, gender, interests, frequency of use, etc.), then uploads the software to be reviewed to the server via the device and submits a review request.

[1257] The server receives the target demographic attribute data sent by the user and generates a virtual reviewer using a generative AI model (e.g., GPT-3 or BERT) based on that data. In this process, a prompt sentence is generated for the generative AI model and input into the model to generate a virtual reviewer.

[1258] The server then assigns review tasks to the virtual reviewers, and integrates an emotion engine (e.g., Microsoft Azure's emotion recognition API) to recognize and record the virtual reviewers' emotional states in real time.

[1259] The virtual reviewer uses the software under review to perform a detailed evaluation according to the specified criteria, during which the emotion engine monitors the reviewer's emotional state and generates sentiment analysis data.

[1260] The server collects the review results generated by the virtual reviewers and the sentiment analysis data generated by the sentiment engine, and organizes and analyzes this data using analytical tools (e.g., Python's Pandas and NumPy libraries) to extract meaningful insights for users.

[1261] Finally, the server generates a report of the analysis results and displays it on the platform. Users can then use their devices to review the report and formulate specific improvement measures based on the detailed review results and sentiment analysis data, which can then be incorporated into the next development cycle.

[1262] As a concrete example, consider the development of a new e-commerce application. The user inputs demographic data such as "women in their 20s who purchase frequently" as the target demographic, and generates a virtual reviewer. This virtual reviewer reviews the application together with an emotion engine. The review results include evaluations of the product display speed and UI, as well as emotional states such as stress felt when products are slow to load.

[1263] Example prompt sentence:

[1264] The target users are women in their 20s who purchase frequently. Please generate a virtual reviewer based on this profile and have them rate the following application.

[1265] In this way, the system of the present invention can quickly perform highly accurate reviews and sentiment analysis specific to the target demographic, allowing users to efficiently improve the quality of their applications.

[1266] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1267] Step 1: User Input and Application Upload

[1268] The user uses the device interface to input demographic data (age, gender, interests, frequency of use, etc.) of the target demographic. The input data is sent to the server in JSON format, which becomes the input.

[1269] The user selects the software to be reviewed (e.g., APK file) and clicks the upload button. The software file is uploaded to the server, which is the output.

[1270] Step 2: Receive demographic data for your target audience

[1271] The server receives the HTTP request, analyzes the target demographic attribute data sent by the user, and saves the received JSON data in an internal database. This becomes the input.

[1272] The attribute data is converted into a format suitable for generative AI models, which then becomes the output.

[1273] Step 3: Generate virtual reviewers

[1274] The server uses a generative AI model (e.g., GPT-3 or BERT) to generate a virtual reviewer. Specifically, it creates a prompt and inputs it into the generative AI model. The prompt is, "The target user is a woman in her 20s who frequently purchases. Please generate a virtual reviewer based on this profile and rate the following application." This is the input.

[1275] The generative AI model generates a virtual reviewer and returns that data to the server, which becomes the output.

[1276] Step 4: Assigning review tasks and integrating the sentiment engine

[1277] The server assigns review tasks for uploaded software to the virtual reviewers, which is the input.

[1278] The server integrates an emotion engine (e.g., Microsoft Azure's emotion recognition API) to recognize and record the emotional state of the virtual reviewer in real time. This is the specific operation. It configures the settings to call the emotion recognition API and sends an API request for each reviewer operation. It receives the response from the API and records the emotional data, which becomes the output.

[1279] Step 5: Performing review tasks

[1280] The virtual reviewers interact with the software according to the assigned review task and provide detailed evaluations based on the specified criteria, which serve as input.

[1281] During operation, the emotion engine monitors the emotional state and generates evaluation data and emotion analysis data, which are the output.

[1282] Step 6: Collect review results and sentiment analysis data

[1283] The server receives review rating data and emotional state data sent by virtual reviewers, which are the input.

[1284] This data is stored in a database and organized and analyzed using analytical tools (e.g., Python's Pandas or NumPy library). This is the specific operation, and the analyzed data is the output.

[1285] Step 7: Generate and serve reports

[1286] The server generates a report based on the analysis results. Specifically, it creates a detailed report by applying the analyzed data to a template. This is the input.

[1287] The generated report is displayed on the user interface and made accessible to the user - this is the output.

[1288] Step 8: Developing improvement measures

[1289] The user uses a terminal to check the report retrieved from the server, which becomes the input.

[1290] Users evaluate the detailed review results and sentiment analysis data and develop specific improvement measures. This is the output. Improvement measures may include, for example, changing the UI design or adding or removing features.

[1291] In this way, the user, server, and device work together at each processing step to achieve highly accurate reviews and sentiment analysis tailored to the target demographic.

[1292] (Application example 2)

[1293] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1294] Conventional review systems rely on user-provided feedback, making it difficult to accurately grasp emotional fluctuations and real-time emotional states. There was also a need for a method to quickly and accurately obtain reviews that are optimal for the target demographic.

[1295] 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 virtual reviewers, means for assigning review tasks to the generated virtual reviewers, means for collecting review results from the virtual reviewers, means for analyzing the collected review results and providing them to users, means for collecting and analyzing emotional data in real time, and means for monitoring and recording the emotional state of users from their gaze, facial expressions, etc. This makes it possible to quickly obtain highly accurate reviews and emotional analysis tailored to the target demographic and efficiently propose improvement measures for applications and products.

[1296] A "virtual reviewer" is a simulated subject for conducting a review, generated using generative artificial intelligence based on attribute data of a specific target demographic.

[1297] A "review task" is a series of tasks assigned to a virtual reviewer to evaluate a particular application or product and record the results.

[1298] "Emotional data" is data used to record and analyze changes in the user's emotional state in real time based on information such as their gaze, facial expressions, and voice.

[1299] "Means for collecting and analyzing emotional data in real time" refers to technology that continuously monitors the user's gaze, facial expressions, etc., and records and analyzes their emotional state on the spot.

[1300] "Generative AI" is an AI model that generates responses and evaluations for specific tasks based on input attribute data.

[1301] "Means for monitoring and recording emotional states from gaze, facial expressions, etc." refers to technology that detects a user's gaze and facial expressions through smart glasses or the like when the user is browsing an online shopping site, and collects and stores that emotional state in real time.

[1302] A "target demographic" is a group of users with specific attributes such as age, gender, interests, and frequency of use, and is the group that is analyzed to evaluate and improve products and services.

[1303] To implement the present invention, the following system configuration and processing procedures are used: Specifically, review evaluation using a virtual reviewer and an emotion engine, and review result collection and analysis procedures.

[1304] System configuration

[1305] 1. User input and application upload

[1306] Users use the device interface to input demographic data (age, gender, interests, frequency of use, etc.) about their target demographic, and also upload the application to be reviewed to the server via the device, allowing the user to send a review request to the server.

[1307] 2. Creating a Virtual Reviewer

[1308] The server receives target demographic attribute data sent by the user and uses generative artificial intelligence to generate a virtual reviewer based on that data. The generated virtual reviewer has the pre-set target demographic attributes and evaluates the application from a perspective based on those attributes.

[1309] 3. Emotion Recognition by Emotion Engine

[1310] The server integrates an emotion engine when assigning review tasks for applications to the virtual reviewers. As the virtual reviewers use the applications, the emotion engine recognizes and records their emotional states in real time. This allows us to quantitatively understand the emotional state of the virtual reviewers.

[1311] 4. Assigning review tasks and conducting reviews

[1312] The server assigns a review task for the application to the generated virtual reviewer. The virtual reviewer uses the application to perform this task and provides a detailed evaluation according to the specified criteria. During the evaluation process, the emotion engine monitors the emotional state of the virtual reviewer and generates emotion analysis data.

[1313] 5. Collection and analysis of review results

[1314] The server collects the review results generated by the virtual reviewers and sentiment analysis data from the sentiment engine, and further analyzes this data to extract meaningful insights for users (e.g., emotional responses to the usability of the user interface, sentiment fluctuations for specific features).

[1315] Specific examples

[1316] For example, consider a female user in her twenties searching for new clothes on an online shopping site. The user puts on smart glasses and launches a dedicated app. The emotion engine collects real-time emotional data from the user's gaze and facial expressions, recording their emotional state. At the same time, a virtual reviewer using generative artificial intelligence evaluates the same product page. As a result, the emotional data of the user and the virtual reviewer is sent to a server for analysis.

[1317] The analysis results are provided to users in the form of a report, which includes specific improvement measures based on the sentiment data. For example, the report may include insights such as "users feel frustrated when products take a long time to load" or "certain designs are particularly well-received."

[1318] Prompt Sentence Examples

[1319] The following prompt sentence is input into the generative artificial intelligence:

[1320] Generate virtual reviewers based on the following user attributes:

[1321] User Attributes:

[1322] Age: 20s

[1323] Gender: Female

[1324] Interests: Fashion, shopping

[1325] Frequency of use: High

[1326] We assign a hypothetical reviewer the task of rating a clothing product page on an online shopping website at the following URL:

[1327] URL: [example.com / fashion / clothing / item1234]

[1328] Output your impressions and emotional data.

[1329] This allows for quick and accurate target-specific reviews and sentiment analysis, which can be used to improve the quality of applications and products.

[1330] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1331] Step 1:

[1332] The user uses the device interface to input demographic data (age, gender, interests, frequency of use, etc.) of the target demographic. The application to be reviewed is then uploaded to the server via the device. This input data is sent to the server and used as the basis for generating virtual reviewers.

[1333] Step 2:

[1334] The server receives the target demographic attribute data sent by the user and generates a virtual reviewer based on that data using generative AI. Specifically, the received attribute data is input into the generative AI model as a prompt sentence, generating an appropriate virtual reviewer.

[1335] Step 3:

[1336] The server assigns a review task for the application to the virtual reviewer, who then evaluates the application according to the specified criteria. During this process, the virtual reviewer's evaluation is recorded.

[1337] Step 4:

[1338] The server integrates an emotion engine to collect real-time emotional data from users' gazes and facial expressions. When users browse online shopping sites through smart glasses, emotion data is recorded in real time using eye tracking and facial recognition technology. This data is sent to the server and integrated with the evaluation data from the virtual reviewers.

[1339] Step 5:

[1340] The server collects the review results generated by the virtual reviewers and sentiment analysis data from the emotion engine. This data includes the user's emotional fluctuations and eye movements, forming detailed user experience data. This allows for comprehensive review results.

[1341] Step 6:

[1342] The server analyzes the collected data and extracts meaningful insights for users. Using data analysis tools such as Apache Hadoop and Apache Spark, the data is processed and calculated to obtain insights such as sentiment fluctuations and the usability of the user interface. Based on this, improvement measures are proposed.

[1343] Step 7:

[1344] The user uses a terminal to check the report obtained from the server. This report integrates sentiment analysis data and review results and includes specific improvement measures. The user then formulates an improvement plan for the application or product based on the report.

[1345] In this way, by carrying out specific processing at each step, it is possible to quickly obtain highly accurate reviews and sentiment analysis specific to the target demographic, thereby improving the quality of applications and products.

[1346] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1347] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1348] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1349] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1350] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1351] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1352] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1353] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1354] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1355] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1356] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1357] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1358] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1359] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1360] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1361] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1362] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1363] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1364] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1365] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1366] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1367] The following is further disclosed regarding the above embodiment.

[1368] (Claim 1)

[1369] a means for generating a virtual reviewer;

[1370] a means for assigning review tasks to the generated virtual reviewers;

[1371] a means for collecting review results from the virtual reviewers;

[1372] A means for analyzing the collected review results and providing them to users;

[1373] A system including:

[1374] (Claim 2)

[1375] 2. The system according to claim 1, further comprising means for receiving attribute data of a target demographic based on user input and generating a virtual reviewer that matches the attribute data using generative artificial intelligence.

[1376] (Claim 3)

[1377] 10. The system of claim 1, further comprising means for uploading the application to be reviewed to the server in response to a user request.

[1378] (Claim 4)

[1379] 10. The system of claim 1, further comprising means for a virtual reviewer to use the application to be reviewed, evaluate it according to specified criteria, and generate an evaluation result.

[1380] (Claim 5)

[1381] 10. The system of claim 1, further comprising means for analyzing the review results, extracting meaningful insights for the user, and providing the same in the form of a report.

[1382] (Claim 6)

[1383] 2. The system according to claim 1, further comprising means for providing a user with an interface for formulating an improvement plan for the application based on the review results.

[1384] "Example 1"

[1385] (Claim 1)

[1386] means for receiving demographic data and a program to be reviewed based on user input;

[1387] A means for using a generative artificial intelligence model to generate a virtual reviewer that matches the attribute data of the received target demographic;

[1388] A means for assigning review tasks to the generated virtual reviewers to have them perform detailed evaluations of the program;

[1389] A means for collecting review results from virtual reviewers and further analyzing the results;

[1390] a means for providing the analysis results to a user;

[1391] A system including:

[1392] (Claim 2)

[1393] 2. The system according to claim 1, further comprising means for generating a prompt statement relating to program evaluation and inputting the prompt statement to the generative artificial intelligence model.

[1394] (Claim 3)

[1395] 2. The system according to claim 1, further comprising means for displaying the collected and analyzed review results in a report format, and allowing a user to check the report using a terminal.

[1396] "Application Example 1"

[1397] (Claim 1)

[1398] a means for generating a virtual reviewer;

[1399] a means for assigning review tasks to the generated virtual reviewers;

[1400] a means for collecting review results from the virtual reviewers;

[1401] A means for analyzing the collected review results and providing them to users;

[1402] A means for receiving attribute data of a target demographic based on a user's input, and using the attribute data to send a prompt sentence to a generative artificial intelligence to generate a virtual reviewer;

[1403] A means to conduct detailed evaluations of products based on the perspective of the target demographic, analyze the results, and provide insights to users;

[1404] A system including:

[1405] (Claim 2)

[1406] 2. The system according to claim 1, further comprising means for sending a prompt sentence to the generative artificial intelligence based on attribute data of the target demographic and analyzing the result.

[1407] (Claim 3)

[1408] 10. The system according to claim 1, further comprising means for uploading data to be reviewed to a server in response to a user request.

[1409] "Example 2: Combining Emotion Engines"

[1410] (Claim 1)

[1411] A means for receiving attribute data of a target demographic based on a user's input and generating a virtual reviewer that matches the attribute data;

[1412] a means for assigning review tasks to the generated virtual reviewers;

[1413] a means for integrating an emotion engine to recognize and record the emotional state of the virtual reviewer in real time;

[1414] a means for collecting review results and sentiment analysis data from the virtual reviewers;

[1415] A means for analyzing the collected review results and sentiment analysis data and providing them to users;

[1416] A system including:

[1417] (Claim 2)

[1418] 10. The system according to claim 1, further comprising means for uploading the software to be reviewed to the server in response to a user request.

[1419] (Claim 3)

[1420] 10. The system of claim 1, further comprising means for generating a virtual reviewer using a generative artificial intelligence model and generating prompts necessary for the generation.

[1421] "Application example 2 when combining emotion engines"

[1422] (Claim 1)

[1423] a means for generating a virtual reviewer;

[1424] a means for assigning review tasks to the generated virtual reviewers;

[1425] a means for collecting review results from the virtual reviewers;

[1426] A means for analyzing the collected review results and providing them to users;

[1427] A means of collecting and analyzing emotion data in real time,

[1428] A means for monitoring and recording the user's emotional state based on their gaze, facial expression, etc.;

[1429] A system including:

[1430] (Claim 2)

[1431] 2. The system according to claim 1, further comprising means for receiving attribute data of a target demographic based on user input and generating a virtual reviewer that matches the attribute data using generative artificial intelligence.

[1432] (Claim 3)

[1433] 10. The system of claim 1, further comprising means for uploading the application to be reviewed to the server in response to a user request. [Explanation of symbols]

[1434] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. a means for generating a virtual reviewer; a means for assigning review tasks to the generated virtual reviewers; a means for collecting review results from the virtual reviewers; A means for analyzing the collected review results and providing them to users; A system including:

2. The system according to claim 1, further comprising means for receiving attribute data of a target demographic based on a user's input, and generating a virtual reviewer that matches the attribute data using generative artificial intelligence.

3. 10. The system of claim 1, further comprising means for uploading the application to be reviewed to the server in response to a user request.

4. The system of claim 1 , further comprising means for a virtual reviewer to use the application to be reviewed, evaluate it according to specified criteria, and generate an evaluation result.

5. 10. The system of claim 1, further comprising means for analyzing the review results, extracting meaningful insights for the user, and providing the insights in the form of a report.

6. 2. The system according to claim 1, further comprising means for providing a user with an interface for formulating an improvement plan for the application based on the review results.

Images