System

Abstract

A system is provided.SOLUTION: A system, comprising: an artificial intelligence unit configured to perform idea generation; a unit configured to create different versions of a service based on a plurality of improvement ideas generated by the artificial intelligence unit; a unit configured to randomly distribute the different versions of the service to user terminals; a unit configured to collect operation logs from the user terminals; a unit configured to analyze the collected operation logs and evaluate performance of each version; and a unit configured to select an optimal version based on an evaluation result and update the service.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 service improvement processes have the problem of requiring a great deal of time and cost to manually generate ideas, implement them, and evaluate them. Furthermore, because only a small number of ideas are tried, there is a high chance that potentially effective improvement proposals will be overlooked. This delays service improvement and makes it difficult to optimize the user experience. [Means for solving the problem]

[0005] In order to solve the above problems, the present invention provides the following means. First, multiple improvement ideas are automatically generated using artificial intelligence means. Next, a server creates different versions of a service based on the generated improvement ideas. These versions are then randomly distributed to user terminals, and operation logs on the terminals are collected. The collected operation logs are sent to a server in real time, which analyzes them and evaluates the performance of each version. Finally, an optimal version is automatically selected based on the evaluation results, and the selected version is applied to all users, thereby providing a system for improving services quickly and at low cost.

[0006] An "artificial intelligence tool" is a device or system that analyzes data and automatically generates new ideas and improvements using specific algorithms or models.

[0007] "Improvement Ideas" are specific suggestions for improving the functionality, performance, or user experience of a service or product.

[0008] "Different versions of a service" refer to multiple implementations of the same service that reflect different improvement ideas.

[0009] A "user terminal" is an electronic device that records operation logs of a service and is used to use a service delivered from a server.

[0010] An "operation log" is information that records behavioral data such as clicks, viewing time, and page transitions that occur when a user uses a service.

[0011] The "collection means" refers to a process or device for capturing operation logs from user terminals.

[0012] "Means for analysis" refers to the process or device that evaluates the collected operation logs using statistical or machine learning techniques and compares and judges the effectiveness of different versions of the service.

[0013] The "means for selecting the optimal version" refers to a process or device that automatically selects the service version that is determined to be the most effective based on the analysis results.

[0014] A "means for updating a service" is a process or device for applying the selected optimal version to all users. [Brief explanation of the drawings]

[0015] [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

[0016] 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.

[0017] First, the terms used in the following description will be explained.

[0018] 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).

[0019] 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.

[0020] 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.

[0021] 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.

[0022] 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."

[0023] [First embodiment]

[0024] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0025] 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.

[0026] 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).

[0027] 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.

[0028] 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.

[0029] 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.

[0030] 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.

[0031] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0032] 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.

[0033] 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.

[0034] 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.

[0035] 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."

[0036] The present invention is a system that uses AI to automatically generate ideas for improving a service, conducts A / B testing based on those ideas, selects the optimal improvement plan, and applies it to the service. Specific embodiments of the system are described below.

[0037] The server first collects a large amount of user data and past feedback, and then uses artificial intelligence to generate improvement ideas based on that data. The generated ideas include, for example, "changing the color of the purchase button from red to blue" or "updating the homepage layout." Based on these generated ideas, the server designs different versions of the service.

[0038] The server then randomly distributes different versions of the designed service to user devices, which then record operation logs for each version distributed and send data such as click rates and duration of visits to the server in real time.

[0039] The server collects the sent operation logs and stores them in a database for analysis. This data is then analyzed using statistical methods and machine learning algorithms to determine which version is most effective. For example, if the click rate for the version with a blue purchase button is higher than that of the red button, the blue button is deemed to be the best.

[0040] The server then selects the optimal version based on the analysis results and updates the service to apply the selected version to all users, resulting in quick and low-cost service improvements and an optimized user experience.

[0041] A specific example would be the following case: The server generates two ideas, "changing the color of the buy button" and "changing the layout," and runs an A / B test on each. The device records the user's actions and sends them to the server. After analyzing the data, the server finds that the version that changes the buy button to blue has a higher click rate, so the blue buy button is officially adopted.

[0042] By repeating this process, it is possible to keep the service in an optimal state at all times. This allows the present invention to provide a system that realizes continuous service improvement at low cost.

[0043] The processing flow will be explained below.

[0044] Step 1:

[0045] The server collects user data and past service improvement examples and stores them in a database.

[0046] Step 2:

[0047] The server uses artificial intelligence tools to generate multiple improvement ideas based on the collected data, such as "change the color of the purchase button from red to blue" or "update the homepage layout."

[0048] Step 3:

[0049] The server designs and builds different versions of the service based on the generated improvement ideas, for example, version A uses a red buy button and version B uses a blue buy button.

[0050] Step 4:

[0051] The server is configured to randomly distribute different versions of the service to user devices. The distribution ratio is determined, and version A and version B are distributed to 50% of users each.

[0052] Step 5:

[0053] Based on the version delivered to the device, user operation logs (clicks, time spent, page transitions, etc.) are recorded in real time.

[0054] Step 6:

[0055] The terminal periodically sends the recorded operation log to the server, which receives it and stores it in a database.

[0056] Step 7:

[0057] Analyze the operation logs collected by the server, using statistical algorithms and machine learning techniques to evaluate the performance of the service (e.g., click-through rate, duration of visit, user engagement).

[0058] Step 8:

[0059] The server compares the effectiveness of each version based on the analysis results. For example, if version B with a blue buy button has a higher click-through rate than version A with a red button, it determines that version B is the best.

[0060] Step 9:

[0061] The server selects the optimal version, updates the service to officially apply that version to all users, and configures the selected version of the service to be distributed to all devices.

[0062] Step 10:

[0063] The server prepares to start the next cycle of generating and testing improvement ideas, enabling a continuous cycle of service improvement.

[0064] Through these steps, the system continues to autonomously improve services and optimize the user experience.

[0065] Example 1

[0066] 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."

[0067] In recent years, formulating improvement proposals aimed at improving the user experience of services has become important, but the process requires a great deal of time and effort. Current methods also present challenges, such as the time it takes to optimize services and the inability to quickly implement effective improvements. In particular, the process of comparatively evaluating multiple improvement proposals is complex, requiring efficient data collection and analysis. Another major problem is the difficulty of collecting and applying data in real time when evaluating the effectiveness of improvement proposals.

[0068] 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.

[0069] In this invention, the server includes artificial intelligence means for generating ideas, means for creating different versions of a service based on a plurality of improvement ideas generated by the artificial intelligence means, means for randomly distributing the different versions of the service to user terminals, means for collecting operation logs from the user terminals, means for saving the collected operation logs in a database in real time, means for analyzing the saved operation logs using statistical analysis or a machine learning algorithm and evaluating the performance of each version, and means for selecting an optimal version based on the evaluation results and automatically updating the service, thereby enabling service improvement and optimization quickly and at low cost.

[0070] "Artificial intelligence means" is a computer program that generates improvement ideas based on user data and feedback.

[0071] A "means for creating a service" is a software or hardware implementation for creating and designing different versions of a service based on generated improvement ideas.

[0072] The "random distribution means" is a processing system for randomly distributing different versions of a service to user terminals.

[0073] A "means for collecting operation logs" is a process or program for collecting various data (click rate, duration of stay, etc.) when a user operates a service.

[0074] "Means for saving to a database in real time" refers to a processing system for instantly storing collected operation logs in a database.

[0075] "Statistical analysis or machine learning algorithms" are computer programs and methods for analyzing operation log data and evaluating the performance of each version.

[0076] "Means for selecting the optimal version and automatically updating the service" refers to a processing system that selects the most effective service version based on the analysis results and applies that version to all users.

[0077] A "generative AI model" is a model that uses artificial intelligence technology to analyze user data and automatically generate new improvement ideas.

[0078] A "prompt sentence" is an instruction sentence that is input to an artificial intelligence model to cause it to output specific information.

[0079] The present invention is a system that uses AI to automatically generate ideas for improving services, conducts A / B testing based on those ideas, selects the optimal improvement plan, and applies it to the service. A specific embodiment of the system is shown below.

[0080] The server collects a large amount of user data and past feedback. This data is stored using the MySQL database management system (DBMS). For example, this data includes user purchase history, page visit time, and feedback comments.

[0081] The server uses a generative AI model (e.g., GPT-3) based on the collected data to generate new service improvement ideas. At this time, the server inputs a prompt into the AI ​​model. An example of a prompt is, "Please generate new service improvement ideas based on user feedback." The AI ​​model generates a specific suggestion, such as "Change the color of the purchase button from red to blue."

[0082] The server then designs different versions of the service based on the generated improvement ideas. Specifically, it creates a version of the webpage in which the color of the purchase button is changed from red to blue. This process involves generating or modifying the HTML and CSS files of the webpage. The development environment is Visual Studio Code, and Git is used for version control.

[0083] The server randomly distributes different versions of the designed service to user devices using load balancers and content delivery networks (CDNs), which distribute specific versions to users randomly.

[0084] The user device records operation logs for each version of the service it receives. The recorded data includes click rates, scroll distances, and page visit times. The user device uses a JavaScript-based analysis library (e.g., Google Analytics) to send this log data to the server in real time.

[0085] The server stores the received operation logs in a database. The stored data is analyzed using statistical analysis software (e.g., the pandas library in R and Python) or machine learning algorithms (e.g., scikit-learn). The server compares and evaluates the click-through rate and dwell time of each version to determine the optimal version. For example, if the click-through rate of the version with a blue purchase button is higher than that of the red one, the blue version is deemed optimal.

[0086] Finally, the server selects the optimal version based on the analysis results and updates the service to apply that version to all users. The updated HTML and CSS files are deployed to the web server and are immediately reflected on the website accessed by all users.

[0087] In this way, the present invention enables rapid and low-cost service improvements, ensuring an optimal user experience at all times, and by continuously collecting, analyzing, and applying data at each stage of the process, it enables effective optimization of the service.

[0088] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0089] Step 1: Collect user data

[0090] The server collects large amounts of user data and past feedback. This data includes user purchase history, page visit time, and feedback comments. As input, it obtains real-time data and past log data when users use websites and services. It stores the data using the MySQL database management system (DBMS) and processes it by organizing it by segment and date. As output, it obtains organized user data.

[0091] Specifically, the server connects to a MySQL database and executes the SQL query "SELECT FROM user_feedbacks" to collect user feedback.

[0092] Step 2: Generate improvement ideas

[0093] The server uses a generative AI model (e.g., GPT-3) based on the collected data to generate new service improvement ideas. Organized user data and a prompt, "Generate new service improvement ideas based on user feedback," are input to the AI ​​model. The generative AI model analyzes the data and performs data calculations to generate new ideas. Specific AI-generated suggestions, such as "Change the color of the purchase button from red to blue," are obtained as output.

[0094] Specifically, the process involves inputting a prompt into the generative AI model and receiving ideas generated by the AI.

[0095] Step 3: Design different versions of the service

[0096] The server designs different versions of the service based on the generated improvement ideas. The input involves using the AI-generated improvement ideas to generate the HTML and CSS files for the web page. The data processing involves designing each version of the web page and versioning the code. The output includes different versions of the service.

[0097] Specifically, the server uses Visual Studio Code to edit the HTML and CSS files to create a version where the buy button color is changed from red to blue.

[0098] Step 4: Random distribution of services

[0099] The server randomly distributes the different versions of the designed service to user devices. As input, it uses different versions of a web page, including configuration for distribution via a load balancer or content delivery network (CDN). As data calculation, it executes an algorithm for randomly distributing the different versions to user devices. As output, it obtains the different versions of the service randomly distributed to user devices.

[0100] Specifically, a load balancer is used to randomly distribute different versions of a service page to user terminals.

[0101] Step 5: Record and send the operation log

[0102] The user terminal records operation logs for the different versions of the service received. The input includes data such as clicks, scrolls, and page visit times when the user uses the service. The data is processed by recording each operation log and sending it to the server in real time. The output is the operation log data sent to the server in real time.

[0103] Specifically, the user's device uses JavaScript to monitor click events and executes the code "document.getElementById('buy_button').addEventListener('click', function() { ...});". This data is sent to the server via Google Analytics.

[0104] Step 6: Data analysis and evaluation of the optimal version

[0105] The server stores the received operation logs in a database. The input contains the operation log data sent in real time. For data processing and data calculation, the operation log data is analyzed using statistical analysis software (e.g., R or the Python pandas library) or machine learning algorithms (e.g., scikit-learn). The output is an evaluation of the click rate and dwell time for each version.

[0106] Specifically, the server uses Python's pandas to calculate the click rate for each version using the code "df.groupby('version')['click_rate'].mean()".

[0107] Step 7: Select and apply the optimal version

[0108] The server selects the optimal version based on the analysis results and updates the service to apply that version to all users. The input includes the evaluation results of each version. The data calculation determines the optimal version and deploys the updated HTML and CSS files to the web server. The output is the optimal version of the service provided to all users.

[0109] Specifically, the server updates the HTML and CSS files and deploys them to the web server to apply the selected optimal version to all users.

[0110] (Application example 1)

[0111] 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."

[0112] In conventional service improvement systems, the generation and testing of improvement ideas is done manually, which is time-consuming and costly, making it difficult to achieve optimal improvements quickly.In addition, there are limited means of collecting and analyzing user response data in real time, making it difficult to accurately evaluate the effectiveness of improvement proposals.

[0113] 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.

[0114] In this invention, the server includes an artificial intelligence means for generating ideas, a means for creating different versions of a service based on a plurality of improvement ideas generated by the artificial intelligence means, a means for randomly distributing the different versions of the service to user terminals, a means for collecting operation logs from the user terminals, a means for analyzing the collected operation logs and evaluating the performance of each version, a means for selecting an optimal version based on the evaluation results and updating the service, a means for installing an application on a smart device and measuring a user's click rate and stay time, and a means for transmitting the measured performance data to the server and analyzing the data in real time, thereby enabling the user experience to be quickly and efficiently optimized.

[0115] "Artificial intelligence means" refers to technology that automatically generates improvement ideas based on user data and feedback.

[0116] "Different versions of a service" are multiple different forms of a service provided to users, designed based on multiple improvement ideas generated by AI.

[0117] A "user terminal" is a device, such as a smartphone or tablet, that connects to and operates the service via the Internet.

[0118] "Operation logs" refer to the operation history and behavioral data of users when using a service, including click rates and duration of stay.

[0119] "Performance" refers to the results of evaluating user reactions and usage patterns for different versions of a service, and refers to specific indicators such as click-through rate and length of stay.

[0120] The "optimal version" is the version of the service that is determined to be the most effective as a result of evaluation based on the collected and analyzed operation logs.

[0121] "Smart device" refers to a device that allows users to connect to the Internet and use services, and includes smartphones and tablets.

[0122] "Means for analyzing data in real time" refers to technology that quickly processes operation log data collected in real time and performs immediate evaluation and analysis.

[0123] "Click-through rate" is an indicator that represents the ratio of the number of times users click on a particular link or button divided by the total number of times it is displayed.

[0124] "Dwell time" refers to the time from when a user accesses a service until when they leave, and indicates how much time the user uses the service.

[0125] The present invention is a system that uses AI to automatically generate ideas for improving services, conducts A / B testing based on those ideas, selects the optimal improvement plan, and applies it to the service. Specific embodiments of the system are described below.

[0126] The server first collects a large amount of user data and past feedback, and then uses AI to generate improvement ideas based on that data. The generated ideas include, for example, changing the color of the purchase button from red to blue, or updating the homepage layout. Based on these generated ideas, the server designs different versions of the service.

[0127] The server then randomly distributes different versions of the designed service to user devices. The user devices record user actions for each distributed version and measure performance data such as click-through rate and time spent. The measured data is sent to the server in real time and collected on the server side.

[0128] The server stores the collected operation logs in a database for analysis, using statistical methods and machine learning algorithms to evaluate the performance of each version and select the most effective version.

[0129] For example, if the version with a blue buy button has a higher click-through rate than the version with a red button, the server will determine that the blue button is the best.The server will then select the best version based on this analysis and update the service to apply the selected version to all users.

[0130] This process makes it possible to keep the service in an optimal state at all times, improving the user experience. As a specific example, the server generates two ideas - "changing the color of the purchase button" and "changing the layout" - and subjects each to A / B testing. The user's device records the user's actions for each version and sends this to the server. After analyzing this data, the server determines that the change in the color of the purchase button, which has the highest click-through rate, is optimal and is officially adopted.

[0131] The technologies used include machine learning frameworks (e.g., TensorFlow, PyTorch) for implementing AI models, databases (e.g., MySQL, PostgreSQL) for data collection and analysis, APIs (e.g., RESTful APIs) for real-time data transmission, etc. Smartphones, tablets, etc. are used as user devices.

[0132] As a concrete example, consider an implementation in a mail-order app. If user A uses the blue button version and user B uses the red button version, the click rate and duration for each version are recorded and analyzed on the server. As a result, the blue button version has a higher click rate, so the blue button version is adopted. Examples of prompts include "Calculate the user click rate for the version where the color of the purchase button is changed from red to blue" and "Analyze the difference in click rates between the blue button and the red button, and select which is optimal."

[0133] The above configuration and processing provide a system that efficiently realizes optimization of the user experience.

[0134] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0135] Step 1:

[0136] The server collects large amounts of user data and past feedback. This data is obtained from user operation history, feedback forms, reviews, etc. Based on the input data, it processes the data to identify user behavior patterns and problems. The output is user data organized in a format that can be input into an AI model.

[0137] Step 2:

[0138] The server inputs the organized user data into an AI model to generate improvement ideas. This program analyzes the data using a machine learning framework (e.g., TensorFlow, PyTorch) and generates improvement ideas. The input is the processed user data, and the output is multiple improvement ideas (e.g., "change the color of the purchase button" or "change the layout of the homepage").

[0139] Step 3:

[0140] The server creates different versions of the service based on the generated improvement ideas. At this stage, different UI / UX designs are created and multiple versions are prepared for testing. The input is the improvement ideas generated by the AI ​​model, and the output is different versions of the service.

[0141] Step 4:

[0142] The server executes a means for randomly delivering different versions of a service to a user terminal. The user terminal displays the received version of the service, and the user uses the service. The input is the information of the different service versions and the user terminal, and the output is the delivery of the service version to the user terminal.

[0143] Step 5:

[0144] The user terminal records user operations for each version of the service. Specifically, it collects operation logs such as click rates and stay times. The input is the operations performed by the user, and the output is the operation log.

[0145] Step 6:

[0146] The user terminal sends the recorded operation log to the server in real time. For real-time data transmission, a RESTful API is used. The input is the operation log, and the output is the data sent to the server.

[0147] Step 7:

[0148] The server stores the received operation logs in a database and analyzes them using statistical methods and machine learning algorithms. This analysis evaluates the performance of each version of the service. The input is the operation logs sent in real time, and the output is the evaluation results.

[0149] Step 8:

[0150] The server selects the optimal version based on the analysis results. The selected version is the one with the highest performance indicators, such as click-through rate and dwell time. The input is the evaluation results of each version, and the output is the identification information of the optimal version.

[0151] Step 9:

[0152] The server automatically applies the selected optimal version to all users. This update process allows all users to enjoy the optimized service. The input is the identification information of the optimal version, and the output is the updated service.

[0153] The above is the processing flow of a system that optimizes the user experience across all steps.

[0154] 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.

[0155] The present invention is a system that uses AI to automatically generate ideas for improving services, conducts A / B testing to select the optimal improvement plan, and applies it to the service, while also taking into account user emotional data for optimization. Specific embodiments of this system are described below.

[0156] First, the server collects user data and past feedback and stores it in a database. Based on this data, it generates multiple improvement ideas using artificial intelligence. Examples of generated ideas include "changing the color of the purchase button" and "changing the layout of the homepage." Based on these ideas, the server designs different versions of the service.

[0157] The server then randomly distributes different versions of the designed service to the user's device. Along with the user's operation log using the distributed version, the device also collects the user's emotional data through an emotion engine. The emotion engine generates emotional data by analyzing the user's facial expressions, voice, or biometric signals.

[0158] The operation logs and emotional data recorded by the device are sent in real time to a server, which stores them in a database and analyzes them. Statistical methods and machine learning algorithms are used for analysis to evaluate the performance of each version. The collected emotional data is also reflected in the performance evaluation. For example, if version B, which has a blue purchase button, has a high click rate and receives a large number of positive emotional responses from users, version B is deemed optimal.

[0159] The server then selects the best version based on the analysis and updates the service to distribute that version to all users. This update is automatic and applies the best version.

[0160] As a concrete example, the server generates two ideas, "changing the color of the buy button" and "changing the layout," and conducts an A / B test. The device collects user emotional data (such as satisfaction and excitement levels read from facial expressions) through an emotion engine along with user operations. The server collects and analyzes this data, and if it determines that a blue buy button is optimal in terms of both click rate and positive emotional response, it officially adopts this version.

[0161] In this way, the present invention provides a system that improves services quickly and at low cost, optimizing the user experience.

[0162] The processing flow will be explained below.

[0163] Step 1:

[0164] The server collects user data and past improvement examples, including user behavior logs, feedback data, session information, etc. This data is then stored in a database.

[0165] Step 2:

[0166] Based on the data collected by the server, multiple improvement ideas are generated using artificial intelligence. Examples of generated ideas include "changing the color of the purchase button" and "changing the layout of the homepage."

[0167] Step 3:

[0168] The server designs and builds different versions of the service based on the generated improvement ideas, for example, version A uses a red buy button and version B uses a blue buy button.

[0169] Step 4:

[0170] The server is configured to randomly distribute different versions of the service to user devices. An expected distribution ratio is set, for example, version A is distributed to 50% of users and version B is distributed to the remaining 50%.

[0171] Step 5:

[0172] The device records operation logs for each version distributed. The recorded content includes clicks, time spent, page transitions, etc. The emotion engine also obtains emotional data from the user's facial expressions, voice, and biometric signals.

[0173] Step 6:

[0174] The device transmits the recorded operation log and emotion data to the server in real time, and the server stores this data in a database.

[0175] Step 7:

[0176] The server analyzes the operation logs and sentiment data stored in the database, and uses statistical methods and machine learning algorithms to evaluate the performance of each version and compare the data.

[0177] Step 8:

[0178] The server compares and verifies the effectiveness of each version based on the analysis results. For example, if version B with a blue purchase button has a higher click rate than version A with a red button, and emotional data indicates higher user satisfaction, version B is deemed optimal.

[0179] Step 9:

[0180] The server selects the optimal version and updates the service to apply that version to all users. The system automatically configures the selected version of the service to be distributed to all devices.

[0181] Step 10:

[0182] The server is ready to start the next idea generation and testing cycle, allowing the next improvement cycle to begin quickly.

[0183] This flow allows the system to constantly try out the latest improvement ideas and provide the optimal version to all users quickly and at low cost.

[0184] Example 2

[0185] 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."

[0186] Current service improvement systems evaluate services based solely on user operation data, making it difficult to optimize services while reflecting the user's emotions and psychological state. This limits the improvement in user experience and makes it difficult to achieve truly effective improvements. Furthermore, traditional methods often require manual data collection and analysis when comparing different versions of a service, resulting in inefficiencies.

[0187] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0188] In this invention, the server includes artificial intelligence means for generating ideas, means for creating different versions of a service based on a plurality of improvement ideas generated by the artificial intelligence means, means for randomly distributing the different versions of the service to user terminals, means for collecting operation logs from user terminals, means for collecting user emotion data using an emotion engine, means for analyzing the collected operation logs and emotion data and evaluating the performance of each version, and means for selecting an optimal version based on the evaluation results and updating the service. This enables automatic improvement and optimization of services based on both user operation logs and emotion data.

[0189] An "artificial intelligence means" is a device or method that uses an AI model to generate a particular output based on data.

[0190] "Improvement ideas" are proposed changes or new ideas to improve user experience or service performance.

[0191] "Service versions" are multiple variations of the same service that differ in specific functionality, appearance, operability, etc.

[0192] A "user terminal" is a device used by a user to access and operate a service, and includes a personal computer, smartphone, etc.

[0193] An "operation log" is data related to the actions and operations of a user when using a service.

[0194] "Emotional data" is data that indicates the user's psychological state and emotional responses, and is collected from facial expressions, tone of voice, biometric signals, etc.

[0195] An "emotion engine" is a device or method that analyzes a user's facial expressions, voice, and biological signals to generate emotion data.

[0196] "Performance" is an evaluation metric based on user actions and emotions for a particular service version.

[0197] "Server" means a computer system for providing services and collecting and analyzing data.

[0198] The present invention is a system that uses AI to automatically generate ideas for improving services, conducts A / B testing to select the optimal improvement plan, and applies it to the service, and further takes user emotional data into consideration when optimizing. A specific embodiment of this system is described below.

[0199] First, the server collects user data and past feedback and stores it in a database. This database uses a database management system such as MySQL or PostgreSQL. The collected data includes user behavior logs and feedback comments. The server efficiently organizes this data and stores it for analysis.

[0200] Next, the server uses a generative AI model (e.g., OpenAI's GPT-4) based on the stored data to generate multiple service improvement ideas. The prompt used is, "Based on past user feedback, please provide ideas for improving the color of the purchase button and the layout of the homepage." Generated ideas include, "Change the color of the purchase button to blue" and "Simplify the homepage layout."

[0201] Based on these ideas, the server designs different versions of the service: a blue buy button version and a layout-changed version, each with different HTML / CSS code and content placement.

[0202] The server then randomly distributes the designed service versions to user terminals, distributing version A to a certain group of users and version B to another group of users, allowing for a fair comparison of the performance of each version.

[0203] The device displays the received service version to the user. For example, the browser renders the page to reflect the new layout and changes to the purchase button. At the same time, the device collects the user's operation log and emotion data. Emotion data is collected using an emotion engine (e.g., camera, microphone, biosensor). The timing when the user clicks the purchase button and their reaction to the new layout are captured here.

[0204] The operation log and emotion data collected by the device are sent to the server in real time. Specifically, each time data is collected, it is sent to the server via an HTTP request. After the server receives this data, it is stored in a database.

[0205] The server analyzes the received data and uses statistical methods (e.g., calculating click-through rates) and machine learning algorithms to evaluate the performance of each version, including sentiment data. For example, it may determine that the blue buy button version has a higher click-through rate and a more positive sentiment response than other versions.

[0206] Finally, the server chooses the best version and updates the service to distribute that version to all users. The update is automatic and applies the best version.

[0207] In this way, a system can be provided in which the server and the terminal cooperate to realize automatic improvement and optimization of services and improve the user experience.

[0208] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0209] Step 1:

[0210] The server collects user data and past feedback and stores it in a database. Specifically, it uses a database management system such as MySQL or PostgreSQL. User action logs and feedback comments are used as input. By storing this data in the database, the server can use it for later analysis. For example, it can collect the number of times a user visited a particular page and feedback comments and insert them into the corresponding tables in the database.

[0211] Step 2:

[0212] The server uses a generative AI model (e.g., GPT-4) to generate multiple service improvement ideas based on the stored user data and feedback. The input is a prompt: "Based on past user feedback, please provide ideas for improving the color of the purchase button and the layout of the homepage." The generative AI model generates ideas based on this prompt and the data, and outputs improvement ideas such as "Change the color of the purchase button to blue" and "Simplify the homepage layout." These ideas are stored on the server and used in the next step.

[0213] Step 3:

[0214] The server designs different versions of the service based on the generated improvement ideas. It uses the improvement ideas generated in the previous step as input. For example, it designs a "blue purchase button version" and a "layout change version" and writes the HTML / CSS code for each. As output, different versions of the service are prepared.

[0215] Step 4:

[0216] The server randomly distributes the designed service versions to user devices. As input, it uses the designed service version and user data. The server distributes version A to a specific group of users and version B to another group of users. As output, it obtains the different service versions distributed to the user devices. This distribution is done in real time via HTTP requests.

[0217] Step 5:

[0218] The terminal displays the received service version to the user. Specifically, the browser renders the page to reflect the new layout and purchase button changes. As input, it uses the service version delivered by the server. As output, it obtains the new service version displayed to the user.

[0219] Step 6:

[0220] The device collects the user's operation log and emotional data. To collect the emotional data, an emotion engine (e.g., a camera, microphone, or biosensor) is used. The input is the user's operation and biometric data. For example, the camera captures the user's facial expression, and the microphone analyzes the tone of voice. The output is the operation log and emotional data.

[0221] Step 7:

[0222] The terminal sends the collected operation log and emotion data to the server in real time. The collected operation log and emotion data are used as input. The transmission is performed via an HTTP request. The data sent to the server is obtained as output.

[0223] Step 8:

[0224] The server stores the received operation log and emotion data in a database. The data sent from the device is used as input. Specifically, new records are added to the operation log table and emotion data table. The data stored in the database is obtained as output.

[0225] Step 9:

[0226] The server analyzes the stored data. As input, it uses operation logs and sentiment data obtained from the database. Statistical methods (e.g., mean value calculation, analysis of variance) and machine learning algorithms (e.g., supervised learning, clustering) are used for the analysis. As output, it obtains the performance evaluation results for each service version.

[0227] Step 10:

[0228] The server selects the optimal version based on the performance evaluation results and updates the service to distribute that version to all users. The evaluation results are used as input. For example, if the blue buy button version is determined to be optimal, the server updates the code for that version to distribute it to all users. The output is a service in which the optimal version is applied to all users.

[0229] (Application example 2)

[0230] 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."

[0231] In today's world, optimizing user experience is extremely important for each service provider. However, conventional methods evaluate improvement ideas based solely on user operation logs, making it difficult to fully reflect the user's emotions and intentions. Furthermore, even when A / B testing is performed, it is difficult to select the optimal improvement plan unless emotional data is taken into account. The present invention aims to solve these problems and provide a system for further improving user experience.

[0232] The specification processing by the specification 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 artificial intelligence means for generating ideas, means for creating different versions of a service based on multiple improvement ideas generated by the artificial intelligence means, means for randomly distributing the different versions of the service to user terminals, means for collecting operation logs and emotion data from the user terminals, means for analyzing the collected operation logs and emotion data and evaluating the performance of each version, and means for selecting the optimal version based on the evaluation results and updating the service. This enables optimal service improvement that takes into account not only user operations but also user emotions.

[0233] "Artificial intelligence means for idea generation" refers to a system or process that uses artificial intelligence technology to automatically generate ideas for improving services.

[0234] "Means for creating different versions of a service based on improvement ideas" refers to a process for creating multiple service versions with different designs and functions based on the multiple improvement ideas that have been generated.

[0235] The "means for randomly distributing different versions of a service to a user terminal" is a method for randomly selecting multiple created service versions and distributing them to different user terminals for testing.

[0236] "Means for collecting operation logs and emotional data from user terminals" refers to the process of acquiring and storing operation history and emotional data (facial expressions, voice, biometric signals, etc.) when a user uses a service.

[0237] "Means for analyzing collected operation logs and emotional data to evaluate the performance of each version" refers to a method for analyzing acquired operation logs and emotional data using statistical methods and machine learning algorithms to evaluate the effectiveness and performance of each service version.

[0238] "Means for selecting the optimal version based on the evaluation results and updating the service" refers to a process for selecting the most effective service version based on the performance evaluation results and automatically updating the entire service based on that version.

[0239] The present invention provides a system for optimizing user experience in electronic payment services. The system includes a server, a user terminal, and an emotion engine. Detailed embodiments of the system are described below.

[0240] 1. System Program

[0241] The server executes a program that includes the following means:

[0242] AI means for idea generation: This uses AI technology to automatically generate ideas for improving services, specifically proposing new interfaces and feature improvements based on user data and past feedback.

[0243] A means for creating different versions of a service based on improvement ideas: Based on the generated improvement ideas, multiple service versions with different designs and functions are created.

[0244] A means for randomly distributing different versions of a service to user terminals: A plurality of created service versions are randomly selected and distributed to different user terminals for testing.

[0245] Means for collecting operation logs and emotional data from user devices: Acquire and store operation history and emotional data (facial expressions, voice, biometric signals, etc.) when a user uses a service.

[0246] A means of analyzing the collected operation logs and sentiment data to evaluate the performance of each version: The acquired operation logs and sentiment data are analyzed using statistical methods and machine learning algorithms to evaluate the effectiveness and performance of each service version.

[0247] A means of selecting the optimal version based on the evaluation results and updating the service: The most effective service version is selected from the performance evaluation results, and the entire service is automatically updated based on that version.

[0248] 2. Program Processing Description

[0249] Hardware

[0250] Smartphone or tablet: Used as the user interface.

[0251] Built-in camera: Used to analyze the user's facial expressions.

[0252] Microphone: Used to collect audio data.

[0253] software

[0254] Emotion engine: Analyzes the user's facial expressions, voice, and biometric signals to generate emotion data. Examples include Affectiva and Amazon Rekognition.

[0255] Machine learning algorithms, such as TensorFlow, are used to analyze collected data and select the optimal service version.

[0256] Database: A database system for storing and managing operation logs and emotion data. Examples include MySQL and PostgreSQL.

[0257] Web server: A server system for providing services. Examples include Nginx and Apache.

[0258] 3. Specific Examples

[0259] As a specific example, the following scenario can be considered.

[0260] When a user uses the electronic payment page, facial expression and voice data collected through the camera and microphone is converted into emotional data by the emotion engine. This data, along with the operation log, is sent to the server in real time. The server stores this data in a database and evaluates the performance of each version using a machine learning algorithm based on TensorFlow. Based on the results of this evaluation, the service version that elicits the most favorable emotional response from users is selected, and that version is automatically applied to all users.

[0261] Prompt Sentence Examples

[0262] "Please advise how to automatically select the optimal interface based on facial expression data when a user completes a payment on an electronic payment page. Also, please explain the selection process."

[0263] This makes it possible to optimize the user experience by taking emotional data into account.

[0264] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0265] Step 1:

[0266] The server collects user data and past feedback and stores it in a database. Using this data as input, an artificial intelligence tool generates multiple improvement ideas. Specifically, a generative AI model is used to output ideas such as "changing the color of the purchase button" or "modifying the layout."

[0267] Step 2:

[0268] The server creates different versions of the service based on the generated improvement ideas. Specifically, it designs multiple service versions with different interfaces and functions based on each improvement idea, and outputs these versions.

[0269] Step 3:

[0270] The server randomly distributes different designed service versions to the user terminal. Specifically, it randomly selects a version based on the user ID and distributes the selected version to the user terminal. The input is the service version and the user ID, and the output is the service version distributed to the user terminal.

[0271] Step 4:

[0272] The device collects operation logs and emotional data when the user uses the service. Specifically, it uses a built-in camera and microphone to capture the user's facial expressions and voice data, analyzes them with an emotion engine (e.g., Affectiva), and generates emotional data. The operation logs and emotional data are the input, and the collected data is the output.

[0273] Step 5:

[0274] The terminal transmits the collected operation log and emotion data to the server in real time. Specifically, it transmits them as packet data. The input is the collected operation log and emotion data, and the data transmitted to the server is the output.

[0275] Step 6:

[0276] The server stores the received operation logs and emotion data in a database and analyzes them using a machine learning algorithm (e.g., TensorFlow). Based on the analysis results, the performance of each service version is evaluated. The input is the stored data, and the analysis results are the output.

[0277] Step 7:

[0278] The server selects the optimal version based on the evaluation results and updates the service with that version. Specifically, the selected optimal version is automatically applied to all users. The input is the evaluation results, and the updated service version is the output.

[0279] 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.

[0280] 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.

[0281] 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.

[0282] [Second embodiment]

[0283] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0284] 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.

[0285] 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).

[0286] 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.

[0287] 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.

[0288] 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).

[0289] 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.

[0290] 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.

[0291] 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.

[0292] 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.

[0293] 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.

[0294] 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."

[0295] The present invention is a system that uses AI to automatically generate ideas for improving a service, conducts A / B testing based on those ideas, selects the optimal improvement plan, and applies it to the service. Specific embodiments of the system are described below.

[0296] The server first collects a large amount of user data and past feedback, and then uses artificial intelligence to generate improvement ideas based on that data. The generated ideas include, for example, "changing the color of the purchase button from red to blue" or "updating the homepage layout." Based on these generated ideas, the server designs different versions of the service.

[0297] The server then randomly distributes different versions of the designed service to user devices, which then record operation logs for each version distributed and send data such as click rates and duration of visits to the server in real time.

[0298] The server collects the sent operation logs and stores them in a database for analysis. This data is then analyzed using statistical methods and machine learning algorithms to determine which version is most effective. For example, if the click rate for the version with a blue purchase button is higher than that of the red button, the blue button is deemed to be the best.

[0299] The server then selects the optimal version based on the analysis results and updates the service to apply the selected version to all users, resulting in quick and low-cost service improvements and an optimized user experience.

[0300] A specific example would be the following case: The server generates two ideas, "changing the color of the buy button" and "changing the layout," and runs an A / B test on each. The device records the user's actions and sends them to the server. After analyzing the data, the server finds that the version that changes the buy button to blue has a higher click rate, so the blue buy button is officially adopted.

[0301] By repeating this process, it is possible to keep the service in an optimal state at all times. This allows the present invention to provide a system that realizes continuous service improvement at low cost.

[0302] The processing flow will be explained below.

[0303] Step 1:

[0304] The server collects user data and past service improvement examples and stores them in a database.

[0305] Step 2:

[0306] The server uses artificial intelligence tools to generate multiple improvement ideas based on the collected data, such as "change the color of the purchase button from red to blue" or "update the homepage layout."

[0307] Step 3:

[0308] The server designs and builds different versions of the service based on the generated improvement ideas, for example, version A uses a red buy button and version B uses a blue buy button.

[0309] Step 4:

[0310] The server is configured to randomly distribute different versions of the service to user devices. The distribution ratio is determined, and version A and version B are distributed to 50% of users each.

[0311] Step 5:

[0312] Based on the version delivered to the device, user operation logs (clicks, time spent, page transitions, etc.) are recorded in real time.

[0313] Step 6:

[0314] The terminal periodically sends the recorded operation log to the server, which receives it and stores it in a database.

[0315] Step 7:

[0316] Analyze the operation logs collected by the server, using statistical algorithms and machine learning techniques to evaluate the performance of the service (e.g., click-through rate, duration of visit, user engagement).

[0317] Step 8:

[0318] The server compares the effectiveness of each version based on the analysis results. For example, if version B with a blue buy button has a higher click-through rate than version A with a red button, it determines that version B is the best.

[0319] Step 9:

[0320] The server selects the optimal version, updates the service to officially apply that version to all users, and configures the selected version of the service to be distributed to all devices.

[0321] Step 10:

[0322] The server prepares to start the next cycle of generating and testing improvement ideas, enabling a continuous cycle of service improvement.

[0323] Through these steps, the system continues to autonomously improve services and optimize the user experience.

[0324] Example 1

[0325] 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."

[0326] In recent years, formulating improvement proposals aimed at improving the user experience of services has become important, but the process requires a great deal of time and effort. Current methods also present challenges, such as the time it takes to optimize services and the inability to quickly implement effective improvements. In particular, the process of comparatively evaluating multiple improvement proposals is complex, requiring efficient data collection and analysis. Another major problem is the difficulty of collecting and applying data in real time when evaluating the effectiveness of improvement proposals.

[0327] 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.

[0328] In this invention, the server includes artificial intelligence means for generating ideas, means for creating different versions of a service based on a plurality of improvement ideas generated by the artificial intelligence means, means for randomly distributing the different versions of the service to user terminals, means for collecting operation logs from the user terminals, means for saving the collected operation logs in a database in real time, means for analyzing the saved operation logs using statistical analysis or a machine learning algorithm and evaluating the performance of each version, and means for selecting an optimal version based on the evaluation results and automatically updating the service, thereby enabling service improvement and optimization quickly and at low cost.

[0329] "Artificial intelligence means" is a computer program that generates improvement ideas based on user data and feedback.

[0330] A "means for creating a service" is a software or hardware implementation for creating and designing different versions of a service based on generated improvement ideas.

[0331] The "random distribution means" is a processing system for randomly distributing different versions of a service to user terminals.

[0332] A "means for collecting operation logs" is a process or program for collecting various data (click rate, duration of stay, etc.) when a user operates a service.

[0333] "Means for saving to a database in real time" refers to a processing system for instantly storing collected operation logs in a database.

[0334] "Statistical analysis or machine learning algorithms" are computer programs and methods for analyzing operation log data and evaluating the performance of each version.

[0335] "Means for selecting the optimal version and automatically updating the service" refers to a processing system that selects the most effective service version based on the analysis results and applies that version to all users.

[0336] A "generative AI model" is a model that uses artificial intelligence technology to analyze user data and automatically generate new improvement ideas.

[0337] A "prompt sentence" is an instruction sentence that is input to an artificial intelligence model to cause it to output specific information.

[0338] The present invention is a system that uses AI to automatically generate ideas for improving services, conducts A / B testing based on those ideas, selects the optimal improvement plan, and applies it to the service. A specific embodiment of the system is shown below.

[0339] The server collects a large amount of user data and past feedback. This data is stored using the MySQL database management system (DBMS). For example, this data includes user purchase history, page visit time, and feedback comments.

[0340] The server uses a generative AI model (e.g., GPT-3) based on the collected data to generate new service improvement ideas. At this time, the server inputs a prompt into the AI ​​model. An example of a prompt is, "Please generate new service improvement ideas based on user feedback." The AI ​​model generates a specific suggestion, such as "Change the color of the purchase button from red to blue."

[0341] The server then designs different versions of the service based on the generated improvement ideas. Specifically, it creates a version of the webpage in which the color of the purchase button is changed from red to blue. This process involves generating or modifying the HTML and CSS files of the webpage. The development environment is Visual Studio Code, and Git is used for version control.

[0342] The server randomly distributes different versions of the designed service to user devices using load balancers and content delivery networks (CDNs), which distribute specific versions to users randomly.

[0343] The user device records operation logs for each version of the service it receives. The recorded data includes click rates, scroll distances, and page visit times. The user device uses a JavaScript-based analysis library (e.g., Google Analytics) to send this log data to the server in real time.

[0344] The server stores the received operation logs in a database. The stored data is analyzed using statistical analysis software (e.g., the pandas library in R and Python) or machine learning algorithms (e.g., scikit-learn). The server compares and evaluates the click-through rate and dwell time of each version to determine the optimal version. For example, if the click-through rate of the version with a blue purchase button is higher than that of the red one, the blue version is deemed optimal.

[0345] Finally, the server selects the optimal version based on the analysis results and updates the service to apply that version to all users. The updated HTML and CSS files are deployed to the web server and are immediately reflected on the website accessed by all users.

[0346] In this way, the present invention enables rapid and low-cost service improvements, ensuring an optimal user experience at all times, and by continuously collecting, analyzing, and applying data at each stage of the process, it enables effective optimization of the service.

[0347] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0348] Step 1: Collect user data

[0349] The server collects large amounts of user data and past feedback. This data includes user purchase history, page visit time, and feedback comments. As input, it obtains real-time data and past log data when users use websites and services. It stores the data using the MySQL database management system (DBMS) and processes it by organizing it by segment and date. As output, it obtains organized user data.

[0350] Specifically, the server connects to a MySQL database and executes the SQL query "SELECT FROM user_feedbacks" to collect user feedback.

[0351] Step 2: Generate improvement ideas

[0352] The server uses a generative AI model (e.g., GPT-3) based on the collected data to generate new service improvement ideas. Organized user data and a prompt, "Generate new service improvement ideas based on user feedback," are input to the AI ​​model. The generative AI model analyzes the data and performs data calculations to generate new ideas. Specific AI-generated suggestions, such as "Change the color of the purchase button from red to blue," are obtained as output.

[0353] Specifically, the process involves inputting a prompt into the generative AI model and receiving ideas generated by the AI.

[0354] Step 3: Design different versions of the service

[0355] The server designs different versions of the service based on the generated improvement ideas. The input involves using the AI-generated improvement ideas to generate the HTML and CSS files for the web page. The data processing involves designing each version of the web page and versioning the code. The output includes different versions of the service.

[0356] Specifically, the server uses Visual Studio Code to edit the HTML and CSS files to create a version where the buy button color is changed from red to blue.

[0357] Step 4: Random distribution of services

[0358] The server randomly distributes the different versions of the designed service to user devices. As input, it uses different versions of a web page, including configuration for distribution via a load balancer or content delivery network (CDN). As data calculation, it executes an algorithm for randomly distributing the different versions to user devices. As output, it obtains the different versions of the service randomly distributed to user devices.

[0359] Specifically, a load balancer is used to randomly distribute different versions of a service page to user terminals.

[0360] Step 5: Record and send the operation log

[0361] The user terminal records operation logs for the different versions of the service received. The input includes data such as clicks, scrolls, and page visit times when the user uses the service. The data is processed by recording each operation log and sending it to the server in real time. The output is the operation log data sent to the server in real time.

[0362] Specifically, the user's device uses JavaScript to monitor click events and executes the code "document.getElementById('buy_button').addEventListener('click', function() { ...});". This data is sent to the server via Google Analytics.

[0363] Step 6: Data analysis and evaluation of the optimal version

[0364] The server stores the received operation logs in a database. The input contains the operation log data sent in real time. For data processing and data calculation, the operation log data is analyzed using statistical analysis software (e.g., R or the Python pandas library) or machine learning algorithms (e.g., scikit-learn). The output is an evaluation of the click rate and dwell time for each version.

[0365] Specifically, the server uses Python's pandas to calculate the click rate for each version using the code "df.groupby('version')['click_rate'].mean()".

[0366] Step 7: Select and apply the optimal version

[0367] The server selects the optimal version based on the analysis results and updates the service to apply that version to all users. The input includes the evaluation results of each version. The data calculation determines the optimal version and deploys the updated HTML and CSS files to the web server. The output is the optimal version of the service provided to all users.

[0368] Specifically, the server updates the HTML and CSS files and deploys them to the web server to apply the selected optimal version to all users.

[0369] (Application example 1)

[0370] 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."

[0371] In conventional service improvement systems, the generation and testing of improvement ideas is done manually, which is time-consuming and costly, making it difficult to achieve optimal improvements quickly.In addition, there are limited means of collecting and analyzing user response data in real time, making it difficult to accurately evaluate the effectiveness of improvement proposals.

[0372] 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.

[0373] In this invention, the server includes an artificial intelligence means for generating ideas, a means for creating different versions of a service based on a plurality of improvement ideas generated by the artificial intelligence means, a means for randomly distributing the different versions of the service to user terminals, a means for collecting operation logs from the user terminals, a means for analyzing the collected operation logs and evaluating the performance of each version, a means for selecting an optimal version based on the evaluation results and updating the service, a means for installing an application on a smart device and measuring a user's click rate and stay time, and a means for transmitting the measured performance data to the server and analyzing the data in real time, thereby enabling the user experience to be quickly and efficiently optimized.

[0374] "Artificial intelligence means" refers to technology that automatically generates improvement ideas based on user data and feedback.

[0375] "Different versions of a service" are multiple different forms of a service provided to users, designed based on multiple improvement ideas generated by AI.

[0376] A "user terminal" is a device, such as a smartphone or tablet, that connects to and operates the service via the Internet.

[0377] "Operation logs" refer to the operation history and behavioral data of users when using a service, including click rates and duration of stay.

[0378] "Performance" refers to the results of evaluating user reactions and usage patterns for different versions of a service, and refers to specific indicators such as click-through rate and length of stay.

[0379] The "optimal version" is the version of the service that is determined to be the most effective as a result of evaluation based on the collected and analyzed operation logs.

[0380] "Smart device" refers to a device that allows users to connect to the Internet and use services, and includes smartphones and tablets.

[0381] "Means for analyzing data in real time" refers to technology that quickly processes operation log data collected in real time and performs immediate evaluation and analysis.

[0382] "Click-through rate" is an indicator that represents the ratio of the number of times users click on a particular link or button divided by the total number of times it is displayed.

[0383] "Dwell time" refers to the time from when a user accesses a service until when they leave, and indicates how much time the user uses the service.

[0384] The present invention is a system that uses AI to automatically generate ideas for improving services, conducts A / B testing based on those ideas, selects the optimal improvement plan, and applies it to the service. Specific embodiments of the system are described below.

[0385] The server first collects a large amount of user data and past feedback, and then uses AI to generate improvement ideas based on that data. The generated ideas include, for example, changing the color of the purchase button from red to blue, or updating the homepage layout. Based on these generated ideas, the server designs different versions of the service.

[0386] The server then randomly distributes different versions of the designed service to user devices. The user devices record user actions for each distributed version and measure performance data such as click-through rate and time spent. The measured data is sent to the server in real time and collected on the server side.

[0387] The server stores the collected operation logs in a database for analysis, using statistical methods and machine learning algorithms to evaluate the performance of each version and select the most effective version.

[0388] For example, if the version with a blue buy button has a higher click-through rate than the version with a red button, the server will determine that the blue button is the best.The server will then select the best version based on this analysis and update the service to apply the selected version to all users.

[0389] This process makes it possible to keep the service in an optimal state at all times, improving the user experience. As a specific example, the server generates two ideas - "changing the color of the purchase button" and "changing the layout" - and subjects each to A / B testing. The user's device records the user's actions for each version and sends this to the server. After analyzing this data, the server determines that the change in the color of the purchase button, which has the highest click-through rate, is optimal and is officially adopted.

[0390] The technologies used include machine learning frameworks (e.g., TensorFlow, PyTorch) for implementing AI models, databases (e.g., MySQL, PostgreSQL) for data collection and analysis, APIs (e.g., RESTful APIs) for real-time data transmission, etc. Smartphones, tablets, etc. are used as user devices.

[0391] As a concrete example, consider an implementation in a mail-order app. If user A uses the blue button version and user B uses the red button version, the click rate and duration for each version are recorded and analyzed on the server. As a result, the blue button version has a higher click rate, so the blue button version is adopted. Examples of prompts include "Calculate the user click rate for the version where the color of the purchase button is changed from red to blue" and "Analyze the difference in click rates between the blue button and the red button, and select which is optimal."

[0392] The above configuration and processing provide a system that efficiently realizes optimization of the user experience.

[0393] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0394] Step 1:

[0395] The server collects large amounts of user data and past feedback. This data is obtained from user operation history, feedback forms, reviews, etc. Based on the input data, it processes the data to identify user behavior patterns and problems. The output is user data organized in a format that can be input into an AI model.

[0396] Step 2:

[0397] The server inputs the organized user data into an AI model to generate improvement ideas. This program analyzes the data using a machine learning framework (e.g., TensorFlow, PyTorch) and generates improvement ideas. The input is the processed user data, and the output is multiple improvement ideas (e.g., "change the color of the purchase button" or "change the layout of the homepage").

[0398] Step 3:

[0399] The server creates different versions of the service based on the generated improvement ideas. At this stage, different UI / UX designs are created and multiple versions are prepared for testing. The input is the improvement ideas generated by the AI ​​model, and the output is different versions of the service.

[0400] Step 4:

[0401] The server executes a means for randomly delivering different versions of a service to a user terminal. The user terminal displays the received version of the service, and the user uses the service. The input is the information of the different service versions and the user terminal, and the output is the delivery of the service version to the user terminal.

[0402] Step 5:

[0403] The user terminal records user operations for each version of the service. Specifically, it collects operation logs such as click rates and stay times. The input is the operations performed by the user, and the output is the operation log.

[0404] Step 6:

[0405] The user terminal sends the recorded operation log to the server in real time. For real-time data transmission, a RESTful API is used. The input is the operation log, and the output is the data sent to the server.

[0406] Step 7:

[0407] The server stores the received operation logs in a database and analyzes them using statistical methods and machine learning algorithms. This analysis evaluates the performance of each version of the service. The input is the operation logs sent in real time, and the output is the evaluation results.

[0408] Step 8:

[0409] The server selects the optimal version based on the analysis results. The selected version is the one with the highest performance indicators, such as click-through rate and dwell time. The input is the evaluation results of each version, and the output is the identification information of the optimal version.

[0410] Step 9:

[0411] The server automatically applies the selected optimal version to all users. This update process allows all users to enjoy the optimized service. The input is the identification information of the optimal version, and the output is the updated service.

[0412] The above is the processing flow of a system that optimizes the user experience across all steps.

[0413] 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.

[0414] The present invention is a system that uses AI to automatically generate ideas for improving services, conducts A / B testing to select the optimal improvement plan, and applies it to the service, while also taking into account user emotional data for optimization. Specific embodiments of this system are described below.

[0415] First, the server collects user data and past feedback and stores it in a database. Based on this data, it generates multiple improvement ideas using artificial intelligence. Examples of generated ideas include "changing the color of the purchase button" and "changing the layout of the homepage." Based on these ideas, the server designs different versions of the service.

[0416] The server then randomly distributes different versions of the designed service to the user's device. Along with the user's operation log using the distributed version, the device also collects the user's emotional data through an emotion engine. The emotion engine generates emotional data by analyzing the user's facial expressions, voice, or biometric signals.

[0417] The operation logs and emotional data recorded by the device are sent in real time to a server, which stores them in a database and analyzes them. Statistical methods and machine learning algorithms are used for analysis to evaluate the performance of each version. The collected emotional data is also reflected in the performance evaluation. For example, if version B, which has a blue purchase button, has a high click rate and receives a large number of positive emotional responses from users, version B is deemed optimal.

[0418] The server then selects the best version based on the analysis and updates the service to distribute that version to all users. This update is automatic and applies the best version.

[0419] As a concrete example, the server generates two ideas, "changing the color of the buy button" and "changing the layout," and conducts an A / B test. The device collects user emotional data (such as satisfaction and excitement levels read from facial expressions) through an emotion engine along with user operations. The server collects and analyzes this data, and if it determines that a blue buy button is optimal in terms of both click rate and positive emotional response, it officially adopts this version.

[0420] In this way, the present invention provides a system that improves services quickly and at low cost, optimizing the user experience.

[0421] The processing flow will be explained below.

[0422] Step 1:

[0423] The server collects user data and past improvement examples, including user behavior logs, feedback data, session information, etc. This data is then stored in a database.

[0424] Step 2:

[0425] Based on the data collected by the server, multiple improvement ideas are generated using artificial intelligence. Examples of generated ideas include "changing the color of the purchase button" and "changing the layout of the homepage."

[0426] Step 3:

[0427] The server designs and builds different versions of the service based on the generated improvement ideas, for example, version A uses a red buy button and version B uses a blue buy button.

[0428] Step 4:

[0429] The server is configured to randomly distribute different versions of the service to user devices. An expected distribution ratio is set, for example, version A is distributed to 50% of users and version B is distributed to the remaining 50%.

[0430] Step 5:

[0431] The device records operation logs for each version distributed. The recorded content includes clicks, time spent, page transitions, etc. The emotion engine also obtains emotional data from the user's facial expressions, voice, and biometric signals.

[0432] Step 6:

[0433] The device transmits the recorded operation log and emotion data to the server in real time, and the server stores this data in a database.

[0434] Step 7:

[0435] The server analyzes the operation logs and sentiment data stored in the database, and uses statistical methods and machine learning algorithms to evaluate the performance of each version and compare the data.

[0436] Step 8:

[0437] The server compares and verifies the effectiveness of each version based on the analysis results. For example, if version B with a blue purchase button has a higher click rate than version A with a red button, and emotional data indicates higher user satisfaction, version B is deemed optimal.

[0438] Step 9:

[0439] The server selects the optimal version and updates the service to apply that version to all users. The system automatically configures the selected version of the service to be distributed to all devices.

[0440] Step 10:

[0441] The server is ready to start the next idea generation and testing cycle, allowing the next improvement cycle to begin quickly.

[0442] This flow allows the system to constantly try out the latest improvement ideas and provide the optimal version to all users quickly and at low cost.

[0443] Example 2

[0444] 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."

[0445] Current service improvement systems evaluate services based solely on user operation data, making it difficult to optimize services while reflecting the user's emotions and psychological state. This limits the improvement in user experience and makes it difficult to achieve truly effective improvements. Furthermore, traditional methods often require manual data collection and analysis when comparing different versions of a service, resulting in inefficiencies.

[0446] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0447] In this invention, the server includes artificial intelligence means for generating ideas, means for creating different versions of a service based on a plurality of improvement ideas generated by the artificial intelligence means, means for randomly distributing the different versions of the service to user terminals, means for collecting operation logs from user terminals, means for collecting user emotion data using an emotion engine, means for analyzing the collected operation logs and emotion data and evaluating the performance of each version, and means for selecting an optimal version based on the evaluation results and updating the service. This enables automatic improvement and optimization of services based on both user operation logs and emotion data.

[0448] An "artificial intelligence means" is a device or method that uses an AI model to generate a particular output based on data.

[0449] "Improvement ideas" are proposed changes or new ideas to improve user experience or service performance.

[0450] "Service versions" are multiple variations of the same service that differ in specific functionality, appearance, operability, etc.

[0451] A "user terminal" is a device used by a user to access and operate a service, and includes a personal computer, smartphone, etc.

[0452] An "operation log" is data related to the actions and operations of a user when using a service.

[0453] "Emotional data" is data that indicates the user's psychological state and emotional responses, and is collected from facial expressions, tone of voice, biometric signals, etc.

[0454] An "emotion engine" is a device or method that analyzes a user's facial expressions, voice, and biological signals to generate emotion data.

[0455] "Performance" is an evaluation metric based on user actions and emotions for a particular service version.

[0456] "Server" means a computer system for providing services and collecting and analyzing data.

[0457] The present invention is a system that uses AI to automatically generate ideas for improving services, conducts A / B testing to select the optimal improvement plan, and applies it to the service, and further takes user emotional data into consideration when optimizing. A specific embodiment of this system is described below.

[0458] First, the server collects user data and past feedback and stores it in a database. This database uses a database management system such as MySQL or PostgreSQL. The collected data includes user behavior logs and feedback comments. The server efficiently organizes this data and stores it for analysis.

[0459] Next, the server uses a generative AI model (e.g., OpenAI's GPT-4) based on the stored data to generate multiple service improvement ideas. The prompt used is, "Based on past user feedback, please provide ideas for improving the color of the purchase button and the layout of the homepage." Generated ideas include, "Change the color of the purchase button to blue" and "Simplify the homepage layout."

[0460] Based on these ideas, the server designs different versions of the service: a blue buy button version and a layout-changed version, each with different HTML / CSS code and content placement.

[0461] The server then randomly distributes the designed service versions to user terminals, distributing version A to a certain group of users and version B to another group of users, allowing for a fair comparison of the performance of each version.

[0462] The device displays the received service version to the user. For example, the browser renders the page to reflect the new layout and changes to the purchase button. At the same time, the device collects the user's operation log and emotion data. Emotion data is collected using an emotion engine (e.g., camera, microphone, biosensor). The timing when the user clicks the purchase button and their reaction to the new layout are captured here.

[0463] The operation log and emotion data collected by the device are sent to the server in real time. Specifically, each time data is collected, it is sent to the server via an HTTP request. After the server receives this data, it is stored in a database.

[0464] The server analyzes the received data and uses statistical methods (e.g., calculating click-through rates) and machine learning algorithms to evaluate the performance of each version, including sentiment data. For example, it may determine that the blue buy button version has a higher click-through rate and a more positive sentiment response than other versions.

[0465] Finally, the server chooses the best version and updates the service to distribute that version to all users. The update is automatic and applies the best version.

[0466] In this way, a system can be provided in which the server and the terminal cooperate to realize automatic improvement and optimization of services and improve the user experience.

[0467] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0468] Step 1:

[0469] The server collects user data and past feedback and stores it in a database. Specifically, it uses a database management system such as MySQL or PostgreSQL. User action logs and feedback comments are used as input. By storing this data in the database, the server can use it for later analysis. For example, it can collect the number of times a user visited a particular page and feedback comments and insert them into the corresponding tables in the database.

[0470] Step 2:

[0471] The server uses a generative AI model (e.g., GPT-4) to generate multiple service improvement ideas based on the stored user data and feedback. The input is a prompt: "Based on past user feedback, please provide ideas for improving the color of the purchase button and the layout of the homepage." The generative AI model generates ideas based on this prompt and the data, and outputs improvement ideas such as "Change the color of the purchase button to blue" and "Simplify the homepage layout." These ideas are stored on the server and used in the next step.

[0472] Step 3:

[0473] The server designs different versions of the service based on the generated improvement ideas. It uses the improvement ideas generated in the previous step as input. For example, it designs a "blue purchase button version" and a "layout change version" and writes the HTML / CSS code for each. As output, different versions of the service are prepared.

[0474] Step 4:

[0475] The server randomly distributes the designed service versions to user devices. As input, it uses the designed service version and user data. The server distributes version A to a specific group of users and version B to another group of users. As output, it obtains the different service versions distributed to the user devices. This distribution is done in real time via HTTP requests.

[0476] Step 5:

[0477] The terminal displays the received service version to the user. Specifically, the browser renders the page to reflect the new layout and purchase button changes. As input, it uses the service version delivered by the server. As output, it obtains the new service version displayed to the user.

[0478] Step 6:

[0479] The device collects the user's operation log and emotional data. To collect the emotional data, an emotion engine (e.g., a camera, microphone, or biosensor) is used. The input is the user's operation and biometric data. For example, the camera captures the user's facial expression, and the microphone analyzes the tone of voice. The output is the operation log and emotional data.

[0480] Step 7:

[0481] The terminal sends the collected operation log and emotion data to the server in real time. The collected operation log and emotion data are used as input. The transmission is performed via an HTTP request. The data sent to the server is obtained as output.

[0482] Step 8:

[0483] The server stores the received operation log and emotion data in a database. The data sent from the device is used as input. Specifically, new records are added to the operation log table and emotion data table. The data stored in the database is obtained as output.

[0484] Step 9:

[0485] The server analyzes the stored data. As input, it uses operation logs and sentiment data obtained from the database. Statistical methods (e.g., mean value calculation, analysis of variance) and machine learning algorithms (e.g., supervised learning, clustering) are used for the analysis. As output, it obtains the performance evaluation results for each service version.

[0486] Step 10:

[0487] The server selects the optimal version based on the performance evaluation results and updates the service to distribute that version to all users. The evaluation results are used as input. For example, if the blue buy button version is determined to be optimal, the server updates the code for that version to distribute it to all users. The output is a service in which the optimal version is applied to all users.

[0488] (Application example 2)

[0489] 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."

[0490] In today's world, optimizing user experience is extremely important for each service provider. However, conventional methods evaluate improvement ideas based solely on user operation logs, making it difficult to fully reflect the user's emotions and intentions. Furthermore, even when A / B testing is performed, it is difficult to select the optimal improvement plan unless emotional data is taken into account. The present invention aims to solve these problems and provide a system for further improving user experience.

[0491] The specification processing by the specification 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 artificial intelligence means for generating ideas, means for creating different versions of a service based on multiple improvement ideas generated by the artificial intelligence means, means for randomly distributing the different versions of the service to user terminals, means for collecting operation logs and emotion data from the user terminals, means for analyzing the collected operation logs and emotion data and evaluating the performance of each version, and means for selecting the optimal version based on the evaluation results and updating the service. This enables optimal service improvement that takes into account not only user operations but also user emotions.

[0492] "Artificial intelligence means for idea generation" refers to a system or process that uses artificial intelligence technology to automatically generate ideas for improving services.

[0493] "Means for creating different versions of a service based on improvement ideas" refers to a process for creating multiple service versions with different designs and functions based on the multiple improvement ideas that have been generated.

[0494] The "means for randomly distributing different versions of a service to a user terminal" is a method for randomly selecting multiple created service versions and distributing them to different user terminals for testing.

[0495] "Means for collecting operation logs and emotional data from user terminals" refers to the process of acquiring and storing operation history and emotional data (facial expressions, voice, biometric signals, etc.) when a user uses a service.

[0496] "Means for analyzing collected operation logs and emotional data to evaluate the performance of each version" refers to a method for analyzing acquired operation logs and emotional data using statistical methods and machine learning algorithms to evaluate the effectiveness and performance of each service version.

[0497] "Means for selecting the optimal version based on the evaluation results and updating the service" refers to a process for selecting the most effective service version based on the performance evaluation results and automatically updating the entire service based on that version.

[0498] The present invention provides a system for optimizing user experience in electronic payment services. The system includes a server, a user terminal, and an emotion engine. Detailed embodiments of the system are described below.

[0499] 1. System Program

[0500] The server executes a program that includes the following means:

[0501] AI means for idea generation: This uses AI technology to automatically generate ideas for improving services, specifically proposing new interfaces and feature improvements based on user data and past feedback.

[0502] A means for creating different versions of a service based on improvement ideas: Based on the generated improvement ideas, multiple service versions with different designs and functions are created.

[0503] A means for randomly distributing different versions of a service to user terminals: A plurality of created service versions are randomly selected and distributed to different user terminals for testing.

[0504] Means for collecting operation logs and emotional data from user devices: Acquire and store operation history and emotional data (facial expressions, voice, biometric signals, etc.) when a user uses a service.

[0505] A means of analyzing the collected operation logs and sentiment data to evaluate the performance of each version: The acquired operation logs and sentiment data are analyzed using statistical methods and machine learning algorithms to evaluate the effectiveness and performance of each service version.

[0506] A means of selecting the optimal version based on the evaluation results and updating the service: The most effective service version is selected from the performance evaluation results, and the entire service is automatically updated based on that version.

[0507] 2. Program Processing Description

[0508] Hardware

[0509] Smartphone or tablet: Used as the user interface.

[0510] Built-in camera: Used to analyze the user's facial expressions.

[0511] Microphone: Used to collect audio data.

[0512] software

[0513] Emotion engine: Analyzes the user's facial expressions, voice, and biometric signals to generate emotion data. Examples include Affectiva and Amazon Rekognition.

[0514] Machine learning algorithms, such as TensorFlow, are used to analyze collected data and select the optimal service version.

[0515] Database: A database system for storing and managing operation logs and emotion data. Examples include MySQL and PostgreSQL.

[0516] Web server: A server system for providing services. Examples include Nginx and Apache.

[0517] 3. Specific Examples

[0518] As a specific example, the following scenario can be considered.

[0519] When a user uses the electronic payment page, facial expression and voice data collected through the camera and microphone is converted into emotional data by the emotion engine. This data, along with the operation log, is sent to the server in real time. The server stores this data in a database and evaluates the performance of each version using a machine learning algorithm based on TensorFlow. Based on the results of this evaluation, the service version that elicits the most favorable emotional response from users is selected, and that version is automatically applied to all users.

[0520] Prompt Sentence Examples

[0521] "Please advise how to automatically select the optimal interface based on facial expression data when a user completes a payment on an electronic payment page. Also, please explain the selection process."

[0522] This makes it possible to optimize the user experience by taking emotional data into account.

[0523] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0524] Step 1:

[0525] The server collects user data and past feedback and stores it in a database. Using this data as input, an artificial intelligence tool generates multiple improvement ideas. Specifically, a generative AI model is used to output ideas such as "changing the color of the purchase button" or "modifying the layout."

[0526] Step 2:

[0527] The server creates different versions of the service based on the generated improvement ideas. Specifically, it designs multiple service versions with different interfaces and functions based on each improvement idea, and outputs these versions.

[0528] Step 3:

[0529] The server randomly distributes different designed service versions to the user terminal. Specifically, it randomly selects a version based on the user ID and distributes the selected version to the user terminal. The input is the service version and the user ID, and the output is the service version distributed to the user terminal.

[0530] Step 4:

[0531] The device collects operation logs and emotional data when the user uses the service. Specifically, it uses a built-in camera and microphone to capture the user's facial expressions and voice data, analyzes them with an emotion engine (e.g., Affectiva), and generates emotional data. The operation logs and emotional data are the input, and the collected data is the output.

[0532] Step 5:

[0533] The terminal transmits the collected operation log and emotion data to the server in real time. Specifically, it transmits them as packet data. The input is the collected operation log and emotion data, and the data transmitted to the server is the output.

[0534] Step 6:

[0535] The server stores the received operation logs and emotion data in a database and analyzes them using a machine learning algorithm (e.g., TensorFlow). Based on the analysis results, the performance of each service version is evaluated. The input is the stored data, and the analysis results are the output.

[0536] Step 7:

[0537] The server selects the optimal version based on the evaluation results and updates the service with that version. Specifically, the selected optimal version is automatically applied to all users. The input is the evaluation results, and the updated service version is the output.

[0538] 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.

[0539] 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.

[0540] 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.

[0541] [Third embodiment]

[0542] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0543] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0544] 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).

[0545] 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.

[0546] 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.

[0547] 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).

[0548] 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.

[0549] 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.

[0550] 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.

[0551] 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.

[0552] 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.

[0553] 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."

[0554] The present invention is a system that uses AI to automatically generate ideas for improving a service, conducts A / B testing based on those ideas, selects the optimal improvement plan, and applies it to the service. Specific embodiments of the system are described below.

[0555] The server first collects a large amount of user data and past feedback, and then uses artificial intelligence to generate improvement ideas based on that data. The generated ideas include, for example, "changing the color of the purchase button from red to blue" or "updating the homepage layout." Based on these generated ideas, the server designs different versions of the service.

[0556] The server then randomly distributes different versions of the designed service to user devices, which then record operation logs for each version distributed and send data such as click rates and duration of visits to the server in real time.

[0557] The server collects the sent operation logs and stores them in a database for analysis. This data is then analyzed using statistical methods and machine learning algorithms to determine which version is most effective. For example, if the click rate for the version with a blue purchase button is higher than that of the red button, the blue button is deemed to be the best.

[0558] The server then selects the optimal version based on the analysis results and updates the service to apply the selected version to all users, resulting in quick and low-cost service improvements and an optimized user experience.

[0559] A specific example would be the following case: The server generates two ideas, "changing the color of the buy button" and "changing the layout," and runs an A / B test on each. The device records the user's actions and sends them to the server. After analyzing the data, the server finds that the version that changes the buy button to blue has a higher click rate, so the blue buy button is officially adopted.

[0560] By repeating this process, it is possible to keep the service in an optimal state at all times. This allows the present invention to provide a system that realizes continuous service improvement at low cost.

[0561] The processing flow will be explained below.

[0562] Step 1:

[0563] The server collects user data and past service improvement examples and stores them in a database.

[0564] Step 2:

[0565] The server uses artificial intelligence tools to generate multiple improvement ideas based on the collected data, such as "change the color of the purchase button from red to blue" or "update the homepage layout."

[0566] Step 3:

[0567] The server designs and builds different versions of the service based on the generated improvement ideas, for example, version A uses a red buy button and version B uses a blue buy button.

[0568] Step 4:

[0569] The server is configured to randomly distribute different versions of the service to user devices. The distribution ratio is determined, and version A and version B are distributed to 50% of users each.

[0570] Step 5:

[0571] Based on the version delivered to the device, user operation logs (clicks, time spent, page transitions, etc.) are recorded in real time.

[0572] Step 6:

[0573] The terminal periodically sends the recorded operation log to the server, which receives it and stores it in a database.

[0574] Step 7:

[0575] Analyze the operation logs collected by the server, using statistical algorithms and machine learning techniques to evaluate the performance of the service (e.g., click-through rate, duration of visit, user engagement).

[0576] Step 8:

[0577] The server compares the effectiveness of each version based on the analysis results. For example, if version B with a blue buy button has a higher click-through rate than version A with a red button, it determines that version B is the best.

[0578] Step 9:

[0579] The server selects the optimal version, updates the service to officially apply that version to all users, and configures the selected version of the service to be distributed to all devices.

[0580] Step 10:

[0581] The server prepares to start the next cycle of generating and testing improvement ideas, enabling a continuous cycle of service improvement.

[0582] Through these steps, the system continues to autonomously improve services and optimize the user experience.

[0583] Example 1

[0584] 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."

[0585] In recent years, formulating improvement proposals aimed at improving the user experience of services has become important, but the process requires a great deal of time and effort. Current methods also present challenges, such as the time it takes to optimize services and the inability to quickly implement effective improvements. In particular, the process of comparatively evaluating multiple improvement proposals is complex, requiring efficient data collection and analysis. Another major problem is the difficulty of collecting and applying data in real time when evaluating the effectiveness of improvement proposals.

[0586] 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.

[0587] In this invention, the server includes artificial intelligence means for generating ideas, means for creating different versions of a service based on a plurality of improvement ideas generated by the artificial intelligence means, means for randomly distributing the different versions of the service to user terminals, means for collecting operation logs from the user terminals, means for saving the collected operation logs in a database in real time, means for analyzing the saved operation logs using statistical analysis or a machine learning algorithm and evaluating the performance of each version, and means for selecting an optimal version based on the evaluation results and automatically updating the service, thereby enabling service improvement and optimization quickly and at low cost.

[0588] "Artificial intelligence means" is a computer program that generates improvement ideas based on user data and feedback.

[0589] A "means for creating a service" is a software or hardware implementation for creating and designing different versions of a service based on generated improvement ideas.

[0590] The "random distribution means" is a processing system for randomly distributing different versions of a service to user terminals.

[0591] A "means for collecting operation logs" is a process or program for collecting various data (click rate, duration of stay, etc.) when a user operates a service.

[0592] "Means for saving to a database in real time" refers to a processing system for instantly storing collected operation logs in a database.

[0593] "Statistical analysis or machine learning algorithms" are computer programs and methods for analyzing operation log data and evaluating the performance of each version.

[0594] "Means for selecting the optimal version and automatically updating the service" refers to a processing system that selects the most effective service version based on the analysis results and applies that version to all users.

[0595] A "generative AI model" is a model that uses artificial intelligence technology to analyze user data and automatically generate new improvement ideas.

[0596] A "prompt sentence" is an instruction sentence that is input to an artificial intelligence model to cause it to output specific information.

[0597] The present invention is a system that uses AI to automatically generate ideas for improving services, conducts A / B testing based on those ideas, selects the optimal improvement plan, and applies it to the service. A specific embodiment of the system is shown below.

[0598] The server collects a large amount of user data and past feedback. This data is stored using the MySQL database management system (DBMS). For example, this data includes user purchase history, page visit time, and feedback comments.

[0599] The server uses a generative AI model (e.g., GPT-3) based on the collected data to generate new service improvement ideas. At this time, the server inputs a prompt into the AI ​​model. An example of a prompt is, "Please generate new service improvement ideas based on user feedback." The AI ​​model generates a specific suggestion, such as "Change the color of the purchase button from red to blue."

[0600] The server then designs different versions of the service based on the generated improvement ideas. Specifically, it creates a version of the webpage in which the color of the purchase button is changed from red to blue. This process involves generating or modifying the HTML and CSS files of the webpage. The development environment is Visual Studio Code, and Git is used for version control.

[0601] The server randomly distributes different versions of the designed service to user devices using load balancers and content delivery networks (CDNs), which distribute specific versions to users randomly.

[0602] The user device records operation logs for each version of the service it receives. The recorded data includes click rates, scroll distances, and page visit times. The user device uses a JavaScript-based analysis library (e.g., Google Analytics) to send this log data to the server in real time.

[0603] The server stores the received operation logs in a database. The stored data is analyzed using statistical analysis software (e.g., the pandas library in R and Python) or machine learning algorithms (e.g., scikit-learn). The server compares and evaluates the click-through rate and dwell time of each version to determine the optimal version. For example, if the click-through rate of the version with a blue purchase button is higher than that of the red one, the blue version is deemed optimal.

[0604] Finally, the server selects the optimal version based on the analysis results and updates the service to apply that version to all users. The updated HTML and CSS files are deployed to the web server and are immediately reflected on the website accessed by all users.

[0605] In this way, the present invention enables rapid and low-cost service improvements, ensuring an optimal user experience at all times, and by continuously collecting, analyzing, and applying data at each stage of the process, it enables effective optimization of the service.

[0606] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0607] Step 1: Collect user data

[0608] The server collects large amounts of user data and past feedback. This data includes user purchase history, page visit time, and feedback comments. As input, it obtains real-time data and past log data when users use websites and services. It stores the data using the MySQL database management system (DBMS) and processes it by organizing it by segment and date. As output, it obtains organized user data.

[0609] Specifically, the server connects to a MySQL database and executes the SQL query "SELECT FROM user_feedbacks" to collect user feedback.

[0610] Step 2: Generate improvement ideas

[0611] The server uses a generative AI model (e.g., GPT-3) based on the collected data to generate new service improvement ideas. Organized user data and a prompt, "Generate new service improvement ideas based on user feedback," are input to the AI ​​model. The generative AI model analyzes the data and performs data calculations to generate new ideas. Specific AI-generated suggestions, such as "Change the color of the purchase button from red to blue," are obtained as output.

[0612] Specifically, the process involves inputting a prompt into the generative AI model and receiving ideas generated by the AI.

[0613] Step 3: Design different versions of the service

[0614] The server designs different versions of the service based on the generated improvement ideas. The input involves using the AI-generated improvement ideas to generate the HTML and CSS files for the web page. The data processing involves designing each version of the web page and versioning the code. The output includes different versions of the service.

[0615] Specifically, the server uses Visual Studio Code to edit the HTML and CSS files to create a version where the buy button color is changed from red to blue.

[0616] Step 4: Random distribution of services

[0617] The server randomly distributes the different versions of the designed service to user devices. As input, it uses different versions of a web page, including configuration for distribution via a load balancer or content delivery network (CDN). As data calculation, it executes an algorithm for randomly distributing the different versions to user devices. As output, it obtains the different versions of the service randomly distributed to user devices.

[0618] Specifically, a load balancer is used to randomly distribute different versions of a service page to user terminals.

[0619] Step 5: Record and send the operation log

[0620] The user terminal records operation logs for the different versions of the service received. The input includes data such as clicks, scrolls, and page visit times when the user uses the service. The data is processed by recording each operation log and sending it to the server in real time. The output is the operation log data sent to the server in real time.

[0621] Specifically, the user's device uses JavaScript to monitor click events and executes the code "document.getElementById('buy_button').addEventListener('click', function() { ...});". This data is sent to the server via Google Analytics.

[0622] Step 6: Data analysis and evaluation of the optimal version

[0623] The server stores the received operation logs in a database. The input contains the operation log data sent in real time. For data processing and data calculation, the operation log data is analyzed using statistical analysis software (e.g., R or the Python pandas library) or machine learning algorithms (e.g., scikit-learn). The output is an evaluation of the click rate and dwell time for each version.

[0624] Specifically, the server uses Python's pandas to calculate the click rate for each version using the code "df.groupby('version')['click_rate'].mean()".

[0625] Step 7: Select and apply the optimal version

[0626] The server selects the optimal version based on the analysis results and updates the service to apply that version to all users. The input includes the evaluation results of each version. The data calculation determines the optimal version and deploys the updated HTML and CSS files to the web server. The output is the optimal version of the service provided to all users.

[0627] Specifically, the server updates the HTML and CSS files and deploys them to the web server to apply the selected optimal version to all users.

[0628] (Application example 1)

[0629] 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."

[0630] In conventional service improvement systems, the generation and testing of improvement ideas is done manually, which is time-consuming and costly, making it difficult to achieve optimal improvements quickly.In addition, there are limited means of collecting and analyzing user response data in real time, making it difficult to accurately evaluate the effectiveness of improvement proposals.

[0631] 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.

[0632] In this invention, the server includes an artificial intelligence means for generating ideas, a means for creating different versions of a service based on a plurality of improvement ideas generated by the artificial intelligence means, a means for randomly distributing the different versions of the service to user terminals, a means for collecting operation logs from the user terminals, a means for analyzing the collected operation logs and evaluating the performance of each version, a means for selecting an optimal version based on the evaluation results and updating the service, a means for installing an application on a smart device and measuring a user's click rate and stay time, and a means for transmitting the measured performance data to the server and analyzing the data in real time, thereby enabling the user experience to be quickly and efficiently optimized.

[0633] "Artificial intelligence means" refers to technology that automatically generates improvement ideas based on user data and feedback.

[0634] "Different versions of a service" are multiple different forms of a service provided to users, designed based on multiple improvement ideas generated by AI.

[0635] A "user terminal" is a device, such as a smartphone or tablet, that connects to and operates the service via the Internet.

[0636] "Operation logs" refer to the operation history and behavioral data of users when using a service, including click rates and duration of stay.

[0637] "Performance" refers to the results of evaluating user reactions and usage patterns for different versions of a service, and refers to specific indicators such as click-through rate and length of stay.

[0638] The "optimal version" is the version of the service that is determined to be the most effective as a result of evaluation based on the collected and analyzed operation logs.

[0639] "Smart device" refers to a device that allows users to connect to the Internet and use services, and includes smartphones and tablets.

[0640] "Means for analyzing data in real time" refers to technology that quickly processes operation log data collected in real time and performs immediate evaluation and analysis.

[0641] "Click-through rate" is an indicator that represents the ratio of the number of times users click on a particular link or button divided by the total number of times it is displayed.

[0642] "Dwell time" refers to the time from when a user accesses a service until when they leave, and indicates how much time the user uses the service.

[0643] The present invention is a system that uses AI to automatically generate ideas for improving services, conducts A / B testing based on those ideas, selects the optimal improvement plan, and applies it to the service. Specific embodiments of the system are described below.

[0644] The server first collects a large amount of user data and past feedback, and then uses AI to generate improvement ideas based on that data. The generated ideas include, for example, changing the color of the purchase button from red to blue, or updating the homepage layout. Based on these generated ideas, the server designs different versions of the service.

[0645] The server then randomly distributes different versions of the designed service to user devices. The user devices record user actions for each distributed version and measure performance data such as click-through rate and time spent. The measured data is sent to the server in real time and collected on the server side.

[0646] The server stores the collected operation logs in a database for analysis, using statistical methods and machine learning algorithms to evaluate the performance of each version and select the most effective version.

[0647] For example, if the version with a blue buy button has a higher click-through rate than the version with a red button, the server will determine that the blue button is the best.The server will then select the best version based on this analysis and update the service to apply the selected version to all users.

[0648] This process makes it possible to keep the service in an optimal state at all times, improving the user experience. As a specific example, the server generates two ideas - "changing the color of the purchase button" and "changing the layout" - and subjects each to A / B testing. The user's device records the user's actions for each version and sends this to the server. After analyzing this data, the server determines that the change in the color of the purchase button, which has the highest click-through rate, is optimal and is officially adopted.

[0649] The technologies used include machine learning frameworks (e.g., TensorFlow, PyTorch) for implementing AI models, databases (e.g., MySQL, PostgreSQL) for data collection and analysis, APIs (e.g., RESTful APIs) for real-time data transmission, etc. Smartphones, tablets, etc. are used as user devices.

[0650] As a concrete example, consider an implementation in a mail-order app. If user A uses the blue button version and user B uses the red button version, the click rate and duration for each version are recorded and analyzed on the server. As a result, the blue button version has a higher click rate, so the blue button version is adopted. Examples of prompts include "Calculate the user click rate for the version where the color of the purchase button is changed from red to blue" and "Analyze the difference in click rates between the blue button and the red button, and select which is optimal."

[0651] The above configuration and processing provide a system that efficiently realizes optimization of the user experience.

[0652] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0653] Step 1:

[0654] The server collects large amounts of user data and past feedback. This data is obtained from user operation history, feedback forms, reviews, etc. Based on the input data, it processes the data to identify user behavior patterns and problems. The output is user data organized in a format that can be input into an AI model.

[0655] Step 2:

[0656] The server inputs the organized user data into an AI model to generate improvement ideas. This program analyzes the data using a machine learning framework (e.g., TensorFlow, PyTorch) and generates improvement ideas. The input is the processed user data, and the output is multiple improvement ideas (e.g., "change the color of the purchase button" or "change the layout of the homepage").

[0657] Step 3:

[0658] The server creates different versions of the service based on the generated improvement ideas. At this stage, different UI / UX designs are created and multiple versions are prepared for testing. The input is the improvement ideas generated by the AI ​​model, and the output is different versions of the service.

[0659] Step 4:

[0660] The server executes a means for randomly delivering different versions of a service to a user terminal. The user terminal displays the received version of the service, and the user uses the service. The input is the information of the different service versions and the user terminal, and the output is the delivery of the service version to the user terminal.

[0661] Step 5:

[0662] The user terminal records user operations for each version of the service. Specifically, it collects operation logs such as click rates and stay times. The input is the operations performed by the user, and the output is the operation log.

[0663] Step 6:

[0664] The user terminal sends the recorded operation log to the server in real time. For real-time data transmission, a RESTful API is used. The input is the operation log, and the output is the data sent to the server.

[0665] Step 7:

[0666] The server stores the received operation logs in a database and analyzes them using statistical methods and machine learning algorithms. This analysis evaluates the performance of each version of the service. The input is the operation logs sent in real time, and the output is the evaluation results.

[0667] Step 8:

[0668] The server selects the optimal version based on the analysis results. The selected version is the one with the highest performance indicators, such as click-through rate and dwell time. The input is the evaluation results of each version, and the output is the identification information of the optimal version.

[0669] Step 9:

[0670] The server automatically applies the selected optimal version to all users. This update process allows all users to enjoy the optimized service. The input is the identification information of the optimal version, and the output is the updated service.

[0671] The above is the processing flow of a system that optimizes the user experience across all steps.

[0672] 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.

[0673] The present invention is a system that uses AI to automatically generate ideas for improving services, conducts A / B testing to select the optimal improvement plan, and applies it to the service, while also taking into account user emotional data for optimization. Specific embodiments of this system are described below.

[0674] First, the server collects user data and past feedback and stores it in a database. Based on this data, it generates multiple improvement ideas using artificial intelligence. Examples of generated ideas include "changing the color of the purchase button" and "changing the layout of the homepage." Based on these ideas, the server designs different versions of the service.

[0675] The server then randomly distributes different versions of the designed service to the user's device. Along with the user's operation log using the distributed version, the device also collects the user's emotional data through an emotion engine. The emotion engine generates emotional data by analyzing the user's facial expressions, voice, or biometric signals.

[0676] The operation logs and emotional data recorded by the device are sent in real time to a server, which stores them in a database and analyzes them. Statistical methods and machine learning algorithms are used for analysis to evaluate the performance of each version. The collected emotional data is also reflected in the performance evaluation. For example, if version B, which has a blue purchase button, has a high click rate and receives a large number of positive emotional responses from users, version B is deemed optimal.

[0677] The server then selects the best version based on the analysis and updates the service to distribute that version to all users. This update is automatic and applies the best version.

[0678] As a concrete example, the server generates two ideas, "changing the color of the buy button" and "changing the layout," and conducts an A / B test. The device collects user emotional data (such as satisfaction and excitement levels read from facial expressions) through an emotion engine along with user operations. The server collects and analyzes this data, and if it determines that a blue buy button is optimal in terms of both click rate and positive emotional response, it officially adopts this version.

[0679] In this way, the present invention provides a system that improves services quickly and at low cost, optimizing the user experience.

[0680] The processing flow will be explained below.

[0681] Step 1:

[0682] The server collects user data and past improvement examples, including user behavior logs, feedback data, session information, etc. This data is then stored in a database.

[0683] Step 2:

[0684] Based on the data collected by the server, multiple improvement ideas are generated using artificial intelligence. Examples of generated ideas include "changing the color of the purchase button" and "changing the layout of the homepage."

[0685] Step 3:

[0686] The server designs and builds different versions of the service based on the generated improvement ideas, for example, version A uses a red buy button and version B uses a blue buy button.

[0687] Step 4:

[0688] The server is configured to randomly distribute different versions of the service to user devices. An expected distribution ratio is set, for example, version A is distributed to 50% of users and version B is distributed to the remaining 50%.

[0689] Step 5:

[0690] The device records operation logs for each version distributed. The recorded content includes clicks, time spent, page transitions, etc. The emotion engine also obtains emotional data from the user's facial expressions, voice, and biometric signals.

[0691] Step 6:

[0692] The device transmits the recorded operation log and emotion data to the server in real time, and the server stores this data in a database.

[0693] Step 7:

[0694] The server analyzes the operation logs and sentiment data stored in the database, and uses statistical methods and machine learning algorithms to evaluate the performance of each version and compare the data.

[0695] Step 8:

[0696] The server compares and verifies the effectiveness of each version based on the analysis results. For example, if version B with a blue purchase button has a higher click rate than version A with a red button, and emotional data indicates higher user satisfaction, version B is deemed optimal.

[0697] Step 9:

[0698] The server selects the optimal version and updates the service to apply that version to all users. The system automatically configures the selected version of the service to be distributed to all devices.

[0699] Step 10:

[0700] The server is ready to start the next idea generation and testing cycle, allowing the next improvement cycle to begin quickly.

[0701] This flow allows the system to constantly try out the latest improvement ideas and provide the optimal version to all users quickly and at low cost.

[0702] Example 2

[0703] 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."

[0704] Current service improvement systems evaluate services based solely on user operation data, making it difficult to optimize services while reflecting the user's emotions and psychological state. This limits the improvement in user experience and makes it difficult to achieve truly effective improvements. Furthermore, traditional methods often require manual data collection and analysis when comparing different versions of a service, resulting in inefficiencies.

[0705] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0706] In this invention, the server includes artificial intelligence means for generating ideas, means for creating different versions of a service based on a plurality of improvement ideas generated by the artificial intelligence means, means for randomly distributing the different versions of the service to user terminals, means for collecting operation logs from user terminals, means for collecting user emotion data using an emotion engine, means for analyzing the collected operation logs and emotion data and evaluating the performance of each version, and means for selecting an optimal version based on the evaluation results and updating the service. This enables automatic improvement and optimization of services based on both user operation logs and emotion data.

[0707] An "artificial intelligence means" is a device or method that uses an AI model to generate a particular output based on data.

[0708] "Improvement ideas" are proposed changes or new ideas to improve user experience or service performance.

[0709] "Service versions" are multiple variations of the same service that differ in specific functionality, appearance, operability, etc.

[0710] A "user terminal" is a device used by a user to access and operate a service, and includes a personal computer, smartphone, etc.

[0711] An "operation log" is data related to the actions and operations of a user when using a service.

[0712] "Emotional data" is data that indicates the user's psychological state and emotional responses, and is collected from facial expressions, tone of voice, biometric signals, etc.

[0713] An "emotion engine" is a device or method that analyzes a user's facial expressions, voice, and biological signals to generate emotion data.

[0714] "Performance" is an evaluation metric based on user actions and emotions for a particular service version.

[0715] "Server" means a computer system for providing services and collecting and analyzing data.

[0716] The present invention is a system that uses AI to automatically generate ideas for improving services, conducts A / B testing to select the optimal improvement plan, and applies it to the service, and further takes user emotional data into consideration when optimizing. A specific embodiment of this system is described below.

[0717] First, the server collects user data and past feedback and stores it in a database. This database uses a database management system such as MySQL or PostgreSQL. The collected data includes user behavior logs and feedback comments. The server efficiently organizes this data and stores it for analysis.

[0718] Next, the server uses a generative AI model (e.g., OpenAI's GPT-4) based on the stored data to generate multiple service improvement ideas. The prompt used is, "Based on past user feedback, please provide ideas for improving the color of the purchase button and the layout of the homepage." Generated ideas include, "Change the color of the purchase button to blue" and "Simplify the homepage layout."

[0719] Based on these ideas, the server designs different versions of the service: a blue buy button version and a layout-changed version, each with different HTML / CSS code and content placement.

[0720] The server then randomly distributes the designed service versions to user terminals, distributing version A to a certain group of users and version B to another group of users, allowing for a fair comparison of the performance of each version.

[0721] The device displays the received service version to the user. For example, the browser renders the page to reflect the new layout and changes to the purchase button. At the same time, the device collects the user's operation log and emotion data. Emotion data is collected using an emotion engine (e.g., camera, microphone, biosensor). The timing when the user clicks the purchase button and their reaction to the new layout are captured here.

[0722] The operation log and emotion data collected by the device are sent to the server in real time. Specifically, each time data is collected, it is sent to the server via an HTTP request. After the server receives this data, it is stored in a database.

[0723] The server analyzes the received data and uses statistical methods (e.g., calculating click-through rates) and machine learning algorithms to evaluate the performance of each version, including sentiment data. For example, it may determine that the blue buy button version has a higher click-through rate and a more positive sentiment response than other versions.

[0724] Finally, the server chooses the best version and updates the service to distribute that version to all users. The update is automatic and applies the best version.

[0725] In this way, a system can be provided in which the server and the terminal cooperate to realize automatic improvement and optimization of services and improve the user experience.

[0726] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0727] Step 1:

[0728] The server collects user data and past feedback and stores it in a database. Specifically, it uses a database management system such as MySQL or PostgreSQL. User action logs and feedback comments are used as input. By storing this data in the database, the server can use it for later analysis. For example, it can collect the number of times a user visited a particular page and feedback comments and insert them into the corresponding tables in the database.

[0729] Step 2:

[0730] The server uses a generative AI model (e.g., GPT-4) to generate multiple service improvement ideas based on the stored user data and feedback. The input is a prompt: "Based on past user feedback, please provide ideas for improving the color of the purchase button and the layout of the homepage." The generative AI model generates ideas based on this prompt and the data, and outputs improvement ideas such as "Change the color of the purchase button to blue" and "Simplify the homepage layout." These ideas are stored on the server and used in the next step.

[0731] Step 3:

[0732] The server designs different versions of the service based on the generated improvement ideas. It uses the improvement ideas generated in the previous step as input. For example, it designs a "blue purchase button version" and a "layout change version" and writes the HTML / CSS code for each. As output, different versions of the service are prepared.

[0733] Step 4:

[0734] The server randomly distributes the designed service versions to user devices. As input, it uses the designed service version and user data. The server distributes version A to a specific group of users and version B to another group of users. As output, it obtains the different service versions distributed to the user devices. This distribution is done in real time via HTTP requests.

[0735] Step 5:

[0736] The terminal displays the received service version to the user. Specifically, the browser renders the page to reflect the new layout and purchase button changes. As input, it uses the service version delivered by the server. As output, it obtains the new service version displayed to the user.

[0737] Step 6:

[0738] The device collects the user's operation log and emotional data. To collect the emotional data, an emotion engine (e.g., a camera, microphone, or biosensor) is used. The input is the user's operation and biometric data. For example, the camera captures the user's facial expression, and the microphone analyzes the tone of voice. The output is the operation log and emotional data.

[0739] Step 7:

[0740] The terminal sends the collected operation log and emotion data to the server in real time. The collected operation log and emotion data are used as input. The transmission is performed via an HTTP request. The data sent to the server is obtained as output.

[0741] Step 8:

[0742] The server stores the received operation log and emotion data in a database. The data sent from the device is used as input. Specifically, new records are added to the operation log table and emotion data table. The data stored in the database is obtained as output.

[0743] Step 9:

[0744] The server analyzes the stored data. As input, it uses operation logs and sentiment data obtained from the database. Statistical methods (e.g., mean value calculation, analysis of variance) and machine learning algorithms (e.g., supervised learning, clustering) are used for the analysis. As output, it obtains the performance evaluation results for each service version.

[0745] Step 10:

[0746] The server selects the optimal version based on the performance evaluation results and updates the service to distribute that version to all users. The evaluation results are used as input. For example, if the blue buy button version is determined to be optimal, the server updates the code for that version to distribute it to all users. The output is a service in which the optimal version is applied to all users.

[0747] (Application example 2)

[0748] 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."

[0749] In today's world, optimizing user experience is extremely important for each service provider. However, conventional methods evaluate improvement ideas based solely on user operation logs, making it difficult to fully reflect the user's emotions and intentions. Furthermore, even when A / B testing is performed, it is difficult to select the optimal improvement plan unless emotional data is taken into account. The present invention aims to solve these problems and provide a system for further improving user experience.

[0750] The specification processing by the specification 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 artificial intelligence means for generating ideas, means for creating different versions of a service based on multiple improvement ideas generated by the artificial intelligence means, means for randomly distributing the different versions of the service to user terminals, means for collecting operation logs and emotion data from the user terminals, means for analyzing the collected operation logs and emotion data and evaluating the performance of each version, and means for selecting the optimal version based on the evaluation results and updating the service. This enables optimal service improvement that takes into account not only user operations but also user emotions.

[0751] "Artificial intelligence means for idea generation" refers to a system or process that uses artificial intelligence technology to automatically generate ideas for improving services.

[0752] "Means for creating different versions of a service based on improvement ideas" refers to a process for creating multiple service versions with different designs and functions based on the multiple improvement ideas that have been generated.

[0753] The "means for randomly distributing different versions of a service to a user terminal" is a method for randomly selecting multiple created service versions and distributing them to different user terminals for testing.

[0754] "Means for collecting operation logs and emotional data from user terminals" refers to the process of acquiring and storing operation history and emotional data (facial expressions, voice, biometric signals, etc.) when a user uses a service.

[0755] "Means for analyzing collected operation logs and emotional data to evaluate the performance of each version" refers to a method for analyzing acquired operation logs and emotional data using statistical methods and machine learning algorithms to evaluate the effectiveness and performance of each service version.

[0756] "Means for selecting the optimal version based on the evaluation results and updating the service" refers to a process for selecting the most effective service version based on the performance evaluation results and automatically updating the entire service based on that version.

[0757] The present invention provides a system for optimizing user experience in electronic payment services. The system includes a server, a user terminal, and an emotion engine. Detailed embodiments of the system are described below.

[0758] 1. System Program

[0759] The server executes a program that includes the following means:

[0760] AI means for idea generation: This uses AI technology to automatically generate ideas for improving services, specifically proposing new interfaces and feature improvements based on user data and past feedback.

[0761] A means for creating different versions of a service based on improvement ideas: Based on the generated improvement ideas, multiple service versions with different designs and functions are created.

[0762] A means for randomly distributing different versions of a service to user terminals: A plurality of created service versions are randomly selected and distributed to different user terminals for testing.

[0763] Means for collecting operation logs and emotional data from user devices: Acquire and store operation history and emotional data (facial expressions, voice, biometric signals, etc.) when a user uses a service.

[0764] A means of analyzing the collected operation logs and sentiment data to evaluate the performance of each version: The acquired operation logs and sentiment data are analyzed using statistical methods and machine learning algorithms to evaluate the effectiveness and performance of each service version.

[0765] A means of selecting the optimal version based on the evaluation results and updating the service: The most effective service version is selected from the performance evaluation results, and the entire service is automatically updated based on that version.

[0766] 2. Program Processing Description

[0767] Hardware

[0768] Smartphone or tablet: Used as the user interface.

[0769] Built-in camera: Used to analyze the user's facial expressions.

[0770] Microphone: Used to collect audio data.

[0771] software

[0772] Emotion engine: Analyzes the user's facial expressions, voice, and biometric signals to generate emotion data. Examples include Affectiva and Amazon Rekognition.

[0773] Machine learning algorithms, such as TensorFlow, are used to analyze collected data and select the optimal service version.

[0774] Database: A database system for storing and managing operation logs and emotion data. Examples include MySQL and PostgreSQL.

[0775] Web server: A server system for providing services. Examples include Nginx and Apache.

[0776] 3. Specific Examples

[0777] As a specific example, the following scenario can be considered.

[0778] When a user uses the electronic payment page, facial expression and voice data collected through the camera and microphone is converted into emotional data by the emotion engine. This data, along with the operation log, is sent to the server in real time. The server stores this data in a database and evaluates the performance of each version using a machine learning algorithm based on TensorFlow. Based on the results of this evaluation, the service version that elicits the most favorable emotional response from users is selected, and that version is automatically applied to all users.

[0779] Prompt Sentence Examples

[0780] "Please advise how to automatically select the optimal interface based on facial expression data when a user completes a payment on an electronic payment page. Also, please explain the selection process."

[0781] This makes it possible to optimize the user experience by taking emotional data into account.

[0782] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0783] Step 1:

[0784] The server collects user data and past feedback and stores it in a database. Using this data as input, an artificial intelligence tool generates multiple improvement ideas. Specifically, a generative AI model is used to output ideas such as "changing the color of the purchase button" or "modifying the layout."

[0785] Step 2:

[0786] The server creates different versions of the service based on the generated improvement ideas. Specifically, it designs multiple service versions with different interfaces and functions based on each improvement idea, and outputs these versions.

[0787] Step 3:

[0788] The server randomly distributes different designed service versions to the user terminal. Specifically, it randomly selects a version based on the user ID and distributes the selected version to the user terminal. The input is the service version and the user ID, and the output is the service version distributed to the user terminal.

[0789] Step 4:

[0790] The device collects operation logs and emotional data when the user uses the service. Specifically, it uses a built-in camera and microphone to capture the user's facial expressions and voice data, analyzes them with an emotion engine (e.g., Affectiva), and generates emotional data. The operation logs and emotional data are the input, and the collected data is the output.

[0791] Step 5:

[0792] The terminal transmits the collected operation log and emotion data to the server in real time. Specifically, it transmits them as packet data. The input is the collected operation log and emotion data, and the data transmitted to the server is the output.

[0793] Step 6:

[0794] The server stores the received operation logs and emotion data in a database and analyzes them using a machine learning algorithm (e.g., TensorFlow). Based on the analysis results, the performance of each service version is evaluated. The input is the stored data, and the analysis results are the output.

[0795] Step 7:

[0796] The server selects the optimal version based on the evaluation results and updates the service with that version. Specifically, the selected optimal version is automatically applied to all users. The input is the evaluation results, and the updated service version is the output.

[0797] 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.

[0798] 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.

[0799] 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.

[0800] [Fourth embodiment]

[0801] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0802] 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.

[0803] 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).

[0804] 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.

[0805] 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.

[0806] 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).

[0807] 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.

[0808] 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.

[0809] 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.

[0810] 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.

[0811] 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.

[0812] 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.

[0813] 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."

[0814] The present invention is a system that uses AI to automatically generate ideas for improving a service, conducts A / B testing based on those ideas, selects the optimal improvement plan, and applies it to the service. Specific embodiments of the system are described below.

[0815] The server first collects a large amount of user data and past feedback, and then uses artificial intelligence to generate improvement ideas based on that data. The generated ideas include, for example, "changing the color of the purchase button from red to blue" or "updating the homepage layout." Based on these generated ideas, the server designs different versions of the service.

[0816] The server then randomly distributes different versions of the designed service to user devices, which then record operation logs for each version distributed and send data such as click rates and duration of visits to the server in real time.

[0817] The server collects the sent operation logs and stores them in a database for analysis. This data is then analyzed using statistical methods and machine learning algorithms to determine which version is most effective. For example, if the click rate for the version with a blue purchase button is higher than that of the red button, the blue button is deemed to be the best.

[0818] The server then selects the optimal version based on the analysis results and updates the service to apply the selected version to all users, resulting in quick and low-cost service improvements and an optimized user experience.

[0819] A specific example would be the following case: The server generates two ideas, "changing the color of the buy button" and "changing the layout," and runs an A / B test on each. The device records the user's actions and sends them to the server. After analyzing the data, the server finds that the version that changes the buy button to blue has a higher click rate, so the blue buy button is officially adopted.

[0820] By repeating this process, it is possible to keep the service in an optimal state at all times. This allows the present invention to provide a system that realizes continuous service improvement at low cost.

[0821] The processing flow will be explained below.

[0822] Step 1:

[0823] The server collects user data and past service improvement examples and stores them in a database.

[0824] Step 2:

[0825] The server uses artificial intelligence tools to generate multiple improvement ideas based on the collected data, such as "change the color of the purchase button from red to blue" or "update the homepage layout."

[0826] Step 3:

[0827] The server designs and builds different versions of the service based on the generated improvement ideas, for example, version A uses a red buy button and version B uses a blue buy button.

[0828] Step 4:

[0829] The server is configured to randomly distribute different versions of the service to user devices. The distribution ratio is determined, and version A and version B are distributed to 50% of users each.

[0830] Step 5:

[0831] Based on the version delivered to the device, user operation logs (clicks, time spent, page transitions, etc.) are recorded in real time.

[0832] Step 6:

[0833] The terminal periodically sends the recorded operation log to the server, which receives it and stores it in a database.

[0834] Step 7:

[0835] Analyze the operation logs collected by the server, using statistical algorithms and machine learning techniques to evaluate the performance of the service (e.g., click-through rate, duration of visit, user engagement).

[0836] Step 8:

[0837] The server compares the effectiveness of each version based on the analysis results. For example, if version B with a blue buy button has a higher click-through rate than version A with a red button, it determines that version B is the best.

[0838] Step 9:

[0839] The server selects the optimal version, updates the service to officially apply that version to all users, and configures the selected version of the service to be distributed to all devices.

[0840] Step 10:

[0841] The server prepares to start the next cycle of generating and testing improvement ideas, enabling a continuous cycle of service improvement.

[0842] Through these steps, the system continues to autonomously improve services and optimize the user experience.

[0843] Example 1

[0844] 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."

[0845] In recent years, formulating improvement proposals aimed at improving the user experience of services has become important, but the process requires a great deal of time and effort. Current methods also present challenges, such as the time it takes to optimize services and the inability to quickly implement effective improvements. In particular, the process of comparatively evaluating multiple improvement proposals is complex, requiring efficient data collection and analysis. Another major problem is the difficulty of collecting and applying data in real time when evaluating the effectiveness of improvement proposals.

[0846] 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.

[0847] In this invention, the server includes artificial intelligence means for generating ideas, means for creating different versions of a service based on a plurality of improvement ideas generated by the artificial intelligence means, means for randomly distributing the different versions of the service to user terminals, means for collecting operation logs from the user terminals, means for saving the collected operation logs in a database in real time, means for analyzing the saved operation logs using statistical analysis or a machine learning algorithm and evaluating the performance of each version, and means for selecting an optimal version based on the evaluation results and automatically updating the service, thereby enabling service improvement and optimization quickly and at low cost.

[0848] "Artificial intelligence means" is a computer program that generates improvement ideas based on user data and feedback.

[0849] A "means for creating a service" is a software or hardware implementation for creating and designing different versions of a service based on generated improvement ideas.

[0850] The "random distribution means" is a processing system for randomly distributing different versions of a service to user terminals.

[0851] A "means for collecting operation logs" is a process or program for collecting various data (click rate, duration of stay, etc.) when a user operates a service.

[0852] "Means for saving to a database in real time" refers to a processing system for instantly storing collected operation logs in a database.

[0853] "Statistical analysis or machine learning algorithms" are computer programs and methods for analyzing operation log data and evaluating the performance of each version.

[0854] "Means for selecting the optimal version and automatically updating the service" refers to a processing system that selects the most effective service version based on the analysis results and applies that version to all users.

[0855] A "generative AI model" is a model that uses artificial intelligence technology to analyze user data and automatically generate new improvement ideas.

[0856] A "prompt sentence" is an instruction sentence that is input to an artificial intelligence model to cause it to output specific information.

[0857] The present invention is a system that uses AI to automatically generate ideas for improving services, conducts A / B testing based on those ideas, selects the optimal improvement plan, and applies it to the service. A specific embodiment of the system is shown below.

[0858] The server collects a large amount of user data and past feedback. This data is stored using the MySQL database management system (DBMS). For example, this data includes user purchase history, page visit time, and feedback comments.

[0859] The server uses a generative AI model (e.g., GPT-3) based on the collected data to generate new service improvement ideas. At this time, the server inputs a prompt into the AI ​​model. An example of a prompt is, "Please generate new service improvement ideas based on user feedback." The AI ​​model generates a specific suggestion, such as "Change the color of the purchase button from red to blue."

[0860] The server then designs different versions of the service based on the generated improvement ideas. Specifically, it creates a version of the webpage in which the color of the purchase button is changed from red to blue. This process involves generating or modifying the HTML and CSS files of the webpage. The development environment is Visual Studio Code, and Git is used for version control.

[0861] The server randomly distributes different versions of the designed service to user devices using load balancers and content delivery networks (CDNs), which distribute specific versions to users randomly.

[0862] The user device records operation logs for each version of the service it receives. The recorded data includes click rates, scroll distances, and page visit times. The user device uses a JavaScript-based analysis library (e.g., Google Analytics) to send this log data to the server in real time.

[0863] The server stores the received operation logs in a database. The stored data is analyzed using statistical analysis software (e.g., the pandas library in R and Python) or machine learning algorithms (e.g., scikit-learn). The server compares and evaluates the click-through rate and dwell time of each version to determine the optimal version. For example, if the click-through rate of the version with a blue purchase button is higher than that of the red one, the blue version is deemed optimal.

[0864] Finally, the server selects the optimal version based on the analysis results and updates the service to apply that version to all users. The updated HTML and CSS files are deployed to the web server and are immediately reflected on the website accessed by all users.

[0865] In this way, the present invention enables rapid and low-cost service improvements, ensuring an optimal user experience at all times, and by continuously collecting, analyzing, and applying data at each stage of the process, it enables effective optimization of the service.

[0866] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0867] Step 1: Collect user data

[0868] The server collects large amounts of user data and past feedback. This data includes user purchase history, page visit time, and feedback comments. As input, it obtains real-time data and past log data when users use websites and services. It stores the data using the MySQL database management system (DBMS) and processes it by organizing it by segment and date. As output, it obtains organized user data.

[0869] Specifically, the server connects to a MySQL database and executes the SQL query "SELECT FROM user_feedbacks" to collect user feedback.

[0870] Step 2: Generate improvement ideas

[0871] The server uses a generative AI model (e.g., GPT-3) based on the collected data to generate new service improvement ideas. Organized user data and a prompt, "Generate new service improvement ideas based on user feedback," are input to the AI ​​model. The generative AI model analyzes the data and performs data calculations to generate new ideas. Specific AI-generated suggestions, such as "Change the color of the purchase button from red to blue," are obtained as output.

[0872] Specifically, the process involves inputting a prompt into the generative AI model and receiving ideas generated by the AI.

[0873] Step 3: Design different versions of the service

[0874] The server designs different versions of the service based on the generated improvement ideas. The input involves using the AI-generated improvement ideas to generate the HTML and CSS files for the web page. The data processing involves designing each version of the web page and versioning the code. The output includes different versions of the service.

[0875] Specifically, the server uses Visual Studio Code to edit the HTML and CSS files to create a version where the buy button color is changed from red to blue.

[0876] Step 4: Random distribution of services

[0877] The server randomly distributes the different versions of the designed service to user devices. As input, it uses different versions of a web page, including configuration for distribution via a load balancer or content delivery network (CDN). As data calculation, it executes an algorithm for randomly distributing the different versions to user devices. As output, it obtains the different versions of the service randomly distributed to user devices.

[0878] Specifically, a load balancer is used to randomly distribute different versions of a service page to user terminals.

[0879] Step 5: Record and send the operation log

[0880] The user terminal records operation logs for the different versions of the service received. The input includes data such as clicks, scrolls, and page visit times when the user uses the service. The data is processed by recording each operation log and sending it to the server in real time. The output is the operation log data sent to the server in real time.

[0881] Specifically, the user's device uses JavaScript to monitor click events and executes the code "document.getElementById('buy_button').addEventListener('click', function() { ...});". This data is sent to the server via Google Analytics.

[0882] Step 6: Data analysis and evaluation of the optimal version

[0883] The server stores the received operation logs in a database. The input contains the operation log data sent in real time. For data processing and data calculation, the operation log data is analyzed using statistical analysis software (e.g., R or the Python pandas library) or machine learning algorithms (e.g., scikit-learn). The output is an evaluation of the click rate and dwell time for each version.

[0884] Specifically, the server uses Python's pandas to calculate the click rate for each version using the code "df.groupby('version')['click_rate'].mean()".

[0885] Step 7: Select and apply the optimal version

[0886] The server selects the optimal version based on the analysis results and updates the service to apply that version to all users. The input includes the evaluation results of each version. The data calculation determines the optimal version and deploys the updated HTML and CSS files to the web server. The output is the optimal version of the service provided to all users.

[0887] Specifically, the server updates the HTML and CSS files and deploys them to the web server to apply the selected optimal version to all users.

[0888] (Application example 1)

[0889] 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."

[0890] In conventional service improvement systems, the generation and testing of improvement ideas is done manually, which is time-consuming and costly, making it difficult to achieve optimal improvements quickly.In addition, there are limited means of collecting and analyzing user response data in real time, making it difficult to accurately evaluate the effectiveness of improvement proposals.

[0891] 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.

[0892] In this invention, the server includes an artificial intelligence means for generating ideas, a means for creating different versions of a service based on a plurality of improvement ideas generated by the artificial intelligence means, a means for randomly distributing the different versions of the service to user terminals, a means for collecting operation logs from the user terminals, a means for analyzing the collected operation logs and evaluating the performance of each version, a means for selecting an optimal version based on the evaluation results and updating the service, a means for installing an application on a smart device and measuring a user's click rate and stay time, and a means for transmitting the measured performance data to the server and analyzing the data in real time, thereby enabling the user experience to be quickly and efficiently optimized.

[0893] "Artificial intelligence means" refers to technology that automatically generates improvement ideas based on user data and feedback.

[0894] "Different versions of a service" are multiple different forms of a service provided to users, designed based on multiple improvement ideas generated by AI.

[0895] A "user terminal" is a device, such as a smartphone or tablet, that connects to and operates the service via the Internet.

[0896] "Operation logs" refer to the operation history and behavioral data of users when using a service, including click rates and duration of stay.

[0897] "Performance" refers to the results of evaluating user reactions and usage patterns for different versions of a service, and refers to specific indicators such as click-through rate and length of stay.

[0898] The "optimal version" is the version of the service that is determined to be the most effective as a result of evaluation based on the collected and analyzed operation logs.

[0899] "Smart device" refers to a device that allows users to connect to the Internet and use services, and includes smartphones and tablets.

[0900] "Means for analyzing data in real time" refers to technology that quickly processes operation log data collected in real time and performs immediate evaluation and analysis.

[0901] "Click-through rate" is an indicator that represents the ratio of the number of times users click on a particular link or button divided by the total number of times it is displayed.

[0902] "Dwell time" refers to the time from when a user accesses a service until when they leave, and indicates how much time the user uses the service.

[0903] The present invention is a system that uses AI to automatically generate ideas for improving services, conducts A / B testing based on those ideas, selects the optimal improvement plan, and applies it to the service. Specific embodiments of the system are described below.

[0904] The server first collects a large amount of user data and past feedback, and then uses AI to generate improvement ideas based on that data. The generated ideas include, for example, changing the color of the purchase button from red to blue, or updating the homepage layout. Based on these generated ideas, the server designs different versions of the service.

[0905] The server then randomly distributes different versions of the designed service to user devices. The user devices record user actions for each distributed version and measure performance data such as click-through rate and time spent. The measured data is sent to the server in real time and collected on the server side.

[0906] The server stores the collected operation logs in a database for analysis, using statistical methods and machine learning algorithms to evaluate the performance of each version and select the most effective version.

[0907] For example, if the version with a blue buy button has a higher click-through rate than the version with a red button, the server will determine that the blue button is the best.The server will then select the best version based on this analysis and update the service to apply the selected version to all users.

[0908] This process makes it possible to keep the service in an optimal state at all times, improving the user experience. As a specific example, the server generates two ideas - "changing the color of the purchase button" and "changing the layout" - and subjects each to A / B testing. The user's device records the user's actions for each version and sends this to the server. After analyzing this data, the server determines that the change in the color of the purchase button, which has the highest click-through rate, is optimal and is officially adopted.

[0909] The technologies used include machine learning frameworks (e.g., TensorFlow, PyTorch) for implementing AI models, databases (e.g., MySQL, PostgreSQL) for data collection and analysis, APIs (e.g., RESTful APIs) for real-time data transmission, etc. Smartphones, tablets, etc. are used as user devices.

[0910] As a concrete example, consider an implementation in a mail-order app. If user A uses the blue button version and user B uses the red button version, the click rate and duration for each version are recorded and analyzed on the server. As a result, the blue button version has a higher click rate, so the blue button version is adopted. Examples of prompts include "Calculate the user click rate for the version where the color of the purchase button is changed from red to blue" and "Analyze the difference in click rates between the blue button and the red button, and select which is optimal."

[0911] The above configuration and processing provide a system that efficiently realizes optimization of the user experience.

[0912] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0913] Step 1:

[0914] The server collects large amounts of user data and past feedback. This data is obtained from user operation history, feedback forms, reviews, etc. Based on the input data, it processes the data to identify user behavior patterns and problems. The output is user data organized in a format that can be input into an AI model.

[0915] Step 2:

[0916] The server inputs the organized user data into an AI model to generate improvement ideas. This program analyzes the data using a machine learning framework (e.g., TensorFlow, PyTorch) and generates improvement ideas. The input is the processed user data, and the output is multiple improvement ideas (e.g., "change the color of the purchase button" or "change the layout of the homepage").

[0917] Step 3:

[0918] The server creates different versions of the service based on the generated improvement ideas. At this stage, different UI / UX designs are created and multiple versions are prepared for testing. The input is the improvement ideas generated by the AI ​​model, and the output is different versions of the service.

[0919] Step 4:

[0920] The server executes a means for randomly delivering different versions of a service to a user terminal. The user terminal displays the received version of the service, and the user uses the service. The input is the information of the different service versions and the user terminal, and the output is the delivery of the service version to the user terminal.

[0921] Step 5:

[0922] The user terminal records user operations for each version of the service. Specifically, it collects operation logs such as click rates and stay times. The input is the operations performed by the user, and the output is the operation log.

[0923] Step 6:

[0924] The user terminal sends the recorded operation log to the server in real time. For real-time data transmission, a RESTful API is used. The input is the operation log, and the output is the data sent to the server.

[0925] Step 7:

[0926] The server stores the received operation logs in a database and analyzes them using statistical methods and machine learning algorithms. This analysis evaluates the performance of each version of the service. The input is the operation logs sent in real time, and the output is the evaluation results.

[0927] Step 8:

[0928] The server selects the optimal version based on the analysis results. The selected version is the one with the highest performance indicators, such as click-through rate and dwell time. The input is the evaluation results of each version, and the output is the identification information of the optimal version.

[0929] Step 9:

[0930] The server automatically applies the selected optimal version to all users. This update process allows all users to enjoy the optimized service. The input is the identification information of the optimal version, and the output is the updated service.

[0931] The above is the processing flow of a system that optimizes the user experience across all steps.

[0932] 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.

[0933] The present invention is a system that uses AI to automatically generate ideas for improving services, conducts A / B testing to select the optimal improvement plan, and applies it to the service, while also taking into account user emotional data for optimization. Specific embodiments of this system are described below.

[0934] First, the server collects user data and past feedback and stores it in a database. Based on this data, it generates multiple improvement ideas using artificial intelligence. Examples of generated ideas include "changing the color of the purchase button" and "changing the layout of the homepage." Based on these ideas, the server designs different versions of the service.

[0935] The server then randomly distributes different versions of the designed service to the user's device. Along with the user's operation log using the distributed version, the device also collects the user's emotional data through an emotion engine. The emotion engine generates emotional data by analyzing the user's facial expressions, voice, or biometric signals.

[0936] The operation logs and emotional data recorded by the device are sent in real time to a server, which stores them in a database and analyzes them. Statistical methods and machine learning algorithms are used for analysis to evaluate the performance of each version. The collected emotional data is also reflected in the performance evaluation. For example, if version B, which has a blue purchase button, has a high click rate and receives a large number of positive emotional responses from users, version B is deemed optimal.

[0937] The server then selects the best version based on the analysis and updates the service to distribute that version to all users. This update is automatic and applies the best version.

[0938] As a concrete example, the server generates two ideas, "changing the color of the buy button" and "changing the layout," and conducts an A / B test. The device collects user emotional data (such as satisfaction and excitement levels read from facial expressions) through an emotion engine along with user operations. The server collects and analyzes this data, and if it determines that a blue buy button is optimal in terms of both click rate and positive emotional response, it officially adopts this version.

[0939] In this way, the present invention provides a system that improves services quickly and at low cost, optimizing the user experience.

[0940] The processing flow will be explained below.

[0941] Step 1:

[0942] The server collects user data and past improvement examples, including user behavior logs, feedback data, session information, etc. This data is then stored in a database.

[0943] Step 2:

[0944] Based on the data collected by the server, multiple improvement ideas are generated using artificial intelligence. Examples of generated ideas include "changing the color of the purchase button" and "changing the layout of the homepage."

[0945] Step 3:

[0946] The server designs and builds different versions of the service based on the generated improvement ideas, for example, version A uses a red buy button and version B uses a blue buy button.

[0947] Step 4:

[0948] The server is configured to randomly distribute different versions of the service to user devices. An expected distribution ratio is set, for example, version A is distributed to 50% of users and version B is distributed to the remaining 50%.

[0949] Step 5:

[0950] The device records operation logs for each version distributed. The recorded content includes clicks, time spent, page transitions, etc. The emotion engine also obtains emotional data from the user's facial expressions, voice, and biometric signals.

[0951] Step 6:

[0952] The device transmits the recorded operation log and emotion data to the server in real time, and the server stores this data in a database.

[0953] Step 7:

[0954] The server analyzes the operation logs and sentiment data stored in the database, and uses statistical methods and machine learning algorithms to evaluate the performance of each version and compare the data.

[0955] Step 8:

[0956] The server compares and verifies the effectiveness of each version based on the analysis results. For example, if version B with a blue purchase button has a higher click rate than version A with a red button, and emotional data indicates higher user satisfaction, version B is deemed optimal.

[0957] Step 9:

[0958] The server selects the optimal version and updates the service to apply that version to all users. The system automatically configures the selected version of the service to be distributed to all devices.

[0959] Step 10:

[0960] The server is ready to start the next idea generation and testing cycle, allowing the next improvement cycle to begin quickly.

[0961] This flow allows the system to constantly try out the latest improvement ideas and provide the optimal version to all users quickly and at low cost.

[0962] Example 2

[0963] 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."

[0964] Current service improvement systems evaluate services based solely on user operation data, making it difficult to optimize services while reflecting the user's emotions and psychological state. This limits the improvement in user experience and makes it difficult to achieve truly effective improvements. Furthermore, traditional methods often require manual data collection and analysis when comparing different versions of a service, resulting in inefficiencies.

[0965] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0966] In this invention, the server includes artificial intelligence means for generating ideas, means for creating different versions of a service based on a plurality of improvement ideas generated by the artificial intelligence means, means for randomly distributing the different versions of the service to user terminals, means for collecting operation logs from user terminals, means for collecting user emotion data using an emotion engine, means for analyzing the collected operation logs and emotion data and evaluating the performance of each version, and means for selecting an optimal version based on the evaluation results and updating the service. This enables automatic improvement and optimization of services based on both user operation logs and emotion data.

[0967] An "artificial intelligence means" is a device or method that uses an AI model to generate a particular output based on data.

[0968] "Improvement ideas" are proposed changes or new ideas to improve user experience or service performance.

[0969] "Service versions" are multiple variations of the same service that differ in specific functionality, appearance, operability, etc.

[0970] A "user terminal" is a device used by a user to access and operate a service, and includes a personal computer, smartphone, etc.

[0971] An "operation log" is data related to the actions and operations of a user when using a service.

[0972] "Emotional data" is data that indicates the user's psychological state and emotional responses, and is collected from facial expressions, tone of voice, biometric signals, etc.

[0973] An "emotion engine" is a device or method that analyzes a user's facial expressions, voice, and biological signals to generate emotion data.

[0974] "Performance" is an evaluation metric based on user actions and emotions for a particular service version.

[0975] "Server" means a computer system for providing services and collecting and analyzing data.

[0976] The present invention is a system that uses AI to automatically generate ideas for improving services, conducts A / B testing to select the optimal improvement plan, and applies it to the service, and further takes user emotional data into consideration when optimizing. A specific embodiment of this system is described below.

[0977] First, the server collects user data and past feedback and stores it in a database. This database uses a database management system such as MySQL or PostgreSQL. The collected data includes user behavior logs and feedback comments. The server efficiently organizes this data and stores it for analysis.

[0978] Next, the server uses a generative AI model (e.g., OpenAI's GPT-4) based on the stored data to generate multiple service improvement ideas. The prompt used is, "Based on past user feedback, please provide ideas for improving the color of the purchase button and the layout of the homepage." Generated ideas include, "Change the color of the purchase button to blue" and "Simplify the homepage layout."

[0979] Based on these ideas, the server designs different versions of the service: a blue buy button version and a layout-changed version, each with different HTML / CSS code and content placement.

[0980] The server then randomly distributes the designed service versions to user terminals, distributing version A to a certain group of users and version B to another group of users, allowing for a fair comparison of the performance of each version.

[0981] The device displays the received service version to the user. For example, the browser renders the page to reflect the new layout and changes to the purchase button. At the same time, the device collects the user's operation log and emotion data. Emotion data is collected using an emotion engine (e.g., camera, microphone, biosensor). The timing when the user clicks the purchase button and their reaction to the new layout are captured here.

[0982] The operation log and emotion data collected by the device are sent to the server in real time. Specifically, each time data is collected, it is sent to the server via an HTTP request. After the server receives this data, it is stored in a database.

[0983] The server analyzes the received data and uses statistical methods (e.g., calculating click-through rates) and machine learning algorithms to evaluate the performance of each version, including sentiment data. For example, it may determine that the blue buy button version has a higher click-through rate and a more positive sentiment response than other versions.

[0984] Finally, the server chooses the best version and updates the service to distribute that version to all users. The update is automatic and applies the best version.

[0985] In this way, a system can be provided in which the server and the terminal cooperate to realize automatic improvement and optimization of services and improve the user experience.

[0986] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0987] Step 1:

[0988] The server collects user data and past feedback and stores it in a database. Specifically, it uses a database management system such as MySQL or PostgreSQL. User action logs and feedback comments are used as input. By storing this data in the database, the server can use it for later analysis. For example, it can collect the number of times a user visited a particular page and feedback comments and insert them into the corresponding tables in the database.

[0989] Step 2:

[0990] The server uses a generative AI model (e.g., GPT-4) to generate multiple service improvement ideas based on the stored user data and feedback. The input is a prompt: "Based on past user feedback, please provide ideas for improving the color of the purchase button and the layout of the homepage." The generative AI model generates ideas based on this prompt and the data, and outputs improvement ideas such as "Change the color of the purchase button to blue" and "Simplify the homepage layout." These ideas are stored on the server and used in the next step.

[0991] Step 3:

[0992] The server designs different versions of the service based on the generated improvement ideas. It uses the improvement ideas generated in the previous step as input. For example, it designs a "blue purchase button version" and a "layout change version" and writes the HTML / CSS code for each. As output, different versions of the service are prepared.

[0993] Step 4:

[0994] The server randomly distributes the designed service versions to user devices. As input, it uses the designed service version and user data. The server distributes version A to a specific group of users and version B to another group of users. As output, it obtains the different service versions distributed to the user devices. This distribution is done in real time via HTTP requests.

[0995] Step 5:

[0996] The terminal displays the received service version to the user. Specifically, the browser renders the page to reflect the new layout and purchase button changes. As input, it uses the service version delivered by the server. As output, it obtains the new service version displayed to the user.

[0997] Step 6:

[0998] The device collects the user's operation log and emotional data. To collect the emotional data, an emotion engine (e.g., a camera, microphone, or biosensor) is used. The input is the user's operation and biometric data. For example, the camera captures the user's facial expression, and the microphone analyzes the tone of voice. The output is the operation log and emotional data.

[0999] Step 7:

[1000] The terminal sends the collected operation log and emotion data to the server in real time. The collected operation log and emotion data are used as input. The transmission is performed via an HTTP request. The data sent to the server is obtained as output.

[1001] Step 8:

[1002] The server stores the received operation log and emotion data in a database. The data sent from the device is used as input. Specifically, new records are added to the operation log table and emotion data table. The data stored in the database is obtained as output.

[1003] Step 9:

[1004] The server analyzes the stored data. As input, it uses operation logs and sentiment data obtained from the database. Statistical methods (e.g., mean value calculation, analysis of variance) and machine learning algorithms (e.g., supervised learning, clustering) are used for the analysis. As output, it obtains the performance evaluation results for each service version.

[1005] Step 10:

[1006] The server selects the optimal version based on the performance evaluation results and updates the service to distribute that version to all users. The evaluation results are used as input. For example, if the blue buy button version is determined to be optimal, the server updates the code for that version to distribute it to all users. The output is a service in which the optimal version is applied to all users.

[1007] (Application example 2)

[1008] 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."

[1009] In today's world, optimizing user experience is extremely important for each service provider. However, conventional methods evaluate improvement ideas based solely on user operation logs, making it difficult to fully reflect the user's emotions and intentions. Furthermore, even when A / B testing is performed, it is difficult to select the optimal improvement plan unless emotional data is taken into account. The present invention aims to solve these problems and provide a system for further improving user experience.

[1010] The specification processing by the specification 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 artificial intelligence means for generating ideas, means for creating different versions of a service based on multiple improvement ideas generated by the artificial intelligence means, means for randomly distributing the different versions of the service to user terminals, means for collecting operation logs and emotion data from the user terminals, means for analyzing the collected operation logs and emotion data and evaluating the performance of each version, and means for selecting the optimal version based on the evaluation results and updating the service. This enables optimal service improvement that takes into account not only user operations but also user emotions.

[1011] "Artificial intelligence means for idea generation" refers to a system or process that uses artificial intelligence technology to automatically generate ideas for improving services.

[1012] "Means for creating different versions of a service based on improvement ideas" refers to a process for creating multiple service versions with different designs and functions based on the multiple improvement ideas that have been generated.

[1013] The "means for randomly distributing different versions of a service to a user terminal" is a method for randomly selecting multiple created service versions and distributing them to different user terminals for testing.

[1014] "Means for collecting operation logs and emotional data from user terminals" refers to the process of acquiring and storing operation history and emotional data (facial expressions, voice, biometric signals, etc.) when a user uses a service.

[1015] "Means for analyzing collected operation logs and emotional data to evaluate the performance of each version" refers to a method for analyzing acquired operation logs and emotional data using statistical methods and machine learning algorithms to evaluate the effectiveness and performance of each service version.

[1016] "Means for selecting the optimal version based on the evaluation results and updating the service" refers to a process for selecting the most effective service version based on the performance evaluation results and automatically updating the entire service based on that version.

[1017] The present invention provides a system for optimizing user experience in electronic payment services. The system includes a server, a user terminal, and an emotion engine. Detailed embodiments of the system are described below.

[1018] 1. System Program

[1019] The server executes a program that includes the following means:

[1020] AI means for idea generation: This uses AI technology to automatically generate ideas for improving services, specifically proposing new interfaces and feature improvements based on user data and past feedback.

[1021] A means for creating different versions of a service based on improvement ideas: Based on the generated improvement ideas, multiple service versions with different designs and functions are created.

[1022] A means for randomly distributing different versions of a service to user terminals: A plurality of created service versions are randomly selected and distributed to different user terminals for testing.

[1023] Means for collecting operation logs and emotional data from user devices: Acquire and store operation history and emotional data (facial expressions, voice, biometric signals, etc.) when a user uses a service.

[1024] A means of analyzing the collected operation logs and sentiment data to evaluate the performance of each version: The acquired operation logs and sentiment data are analyzed using statistical methods and machine learning algorithms to evaluate the effectiveness and performance of each service version.

[1025] A means of selecting the optimal version based on the evaluation results and updating the service: The most effective service version is selected from the performance evaluation results, and the entire service is automatically updated based on that version.

[1026] 2. Program Processing Description

[1027] Hardware

[1028] Smartphone or tablet: Used as the user interface.

[1029] Built-in camera: Used to analyze the user's facial expressions.

[1030] Microphone: Used to collect audio data.

[1031] software

[1032] Emotion engine: Analyzes the user's facial expressions, voice, and biometric signals to generate emotion data. Examples include Affectiva and Amazon Rekognition.

[1033] Machine learning algorithms, such as TensorFlow, are used to analyze collected data and select the optimal service version.

[1034] Database: A database system for storing and managing operation logs and emotion data. Examples include MySQL and PostgreSQL.

[1035] Web server: A server system for providing services. Examples include Nginx and Apache.

[1036] 3. Specific Examples

[1037] As a specific example, the following scenario can be considered.

[1038] When a user uses the electronic payment page, facial expression and voice data collected through the camera and microphone is converted into emotional data by the emotion engine. This data, along with the operation log, is sent to the server in real time. The server stores this data in a database and evaluates the performance of each version using a machine learning algorithm based on TensorFlow. Based on the results of this evaluation, the service version that elicits the most favorable emotional response from users is selected, and that version is automatically applied to all users.

[1039] Prompt Sentence Examples

[1040] "Please advise how to automatically select the optimal interface based on facial expression data when a user completes a payment on an electronic payment page. Also, please explain the selection process."

[1041] This makes it possible to optimize the user experience by taking emotional data into account.

[1042] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1043] Step 1:

[1044] The server collects user data and past feedback and stores it in a database. Using this data as input, an artificial intelligence tool generates multiple improvement ideas. Specifically, a generative AI model is used to output ideas such as "changing the color of the purchase button" or "modifying the layout."

[1045] Step 2:

[1046] The server creates different versions of the service based on the generated improvement ideas. Specifically, it designs multiple service versions with different interfaces and functions based on each improvement idea, and outputs these versions.

[1047] Step 3:

[1048] The server randomly distributes different designed service versions to the user terminal. Specifically, it randomly selects a version based on the user ID and distributes the selected version to the user terminal. The input is the service version and the user ID, and the output is the service version distributed to the user terminal.

[1049] Step 4:

[1050] The device collects operation logs and emotional data when the user uses the service. Specifically, it uses a built-in camera and microphone to capture the user's facial expressions and voice data, analyzes them with an emotion engine (e.g., Affectiva), and generates emotional data. The operation logs and emotional data are the input, and the collected data is the output.

[1051] Step 5:

[1052] The terminal transmits the collected operation log and emotion data to the server in real time. Specifically, it transmits them as packet data. The input is the collected operation log and emotion data, and the data transmitted to the server is the output.

[1053] Step 6:

[1054] The server stores the received operation logs and emotion data in a database and analyzes them using a machine learning algorithm (e.g., TensorFlow). Based on the analysis results, the performance of each service version is evaluated. The input is the stored data, and the analysis results are the output.

[1055] Step 7:

[1056] The server selects the optimal version based on the evaluation results and updates the service with that version. Specifically, the selected optimal version is automatically applied to all users. The input is the evaluation results, and the updated service version is the output.

[1057] 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.

[1058] 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.

[1059] 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.

[1060] 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.

[1061] 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.

[1062] 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.

[1063] 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).

[1064] 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.

[1065] 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."

[1066] 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.

[1067] 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).

[1068] 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.

[1069] 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.

[1070] 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.

[1071] 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.

[1072] 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.

[1073] 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.

[1074] 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.

[1075] 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.

[1076] 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.

[1077] 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.

[1078] The following is further disclosed regarding the above embodiment.

[1079] (Claim 1)

[1080] an artificial intelligence means for idea generation;

[1081] means for creating different versions of the service based on a plurality of improvement ideas generated by the artificial intelligence means;

[1082] means for randomly distributing the different versions of the service to a user terminal;

[1083] A means for collecting operation logs from user terminals;

[1084] A means to analyze the collected operation logs and evaluate the performance of each version;

[1085] A means for selecting the optimal version based on the evaluation results and updating the service;

[1086] A system including:

[1087] (Claim 2)

[1088] 2. The system according to claim 1, wherein the collected operation log is transmitted to a server in real time.

[1089] (Claim 3)

[1090] 2. The system of claim 1, wherein the selected optimal version is automatically applied to all users.

[1091] "Example 1"

[1092] (Claim 1)

[1093] an artificial intelligence means for idea generation;

[1094] means for creating different versions of the service based on a plurality of improvement ideas generated by the artificial intelligence means;

[1095] means for randomly distributing the different versions of the service to a user terminal;

[1096] A means for collecting operation logs from user terminals;

[1097] A means to save the collected operation logs in a database in real time,

[1098] A means for analyzing the stored operation logs using statistical analysis or machine learning algorithms to evaluate the performance of each version;

[1099] A means for selecting the optimal version based on the evaluation results and automatically updating the service;

[1100] A system including:

[1101] (Claim 2)

[1102] The system of claim 1 generates improvement ideas by inputting a prompt sentence and using a generative AI model.

[1103] (Claim 3)

[1104] 10. The system of claim 1, wherein the service is updated to apply the selected optimal version to all users.

[1105] "Application Example 1"

[1106] (Claim 1)

[1107] an artificial intelligence means for idea generation;

[1108] means for creating different versions of the service based on a plurality of improvement ideas generated by the artificial intelligence means;

[1109] means for randomly distributing the different versions of the service to a user terminal;

[1110] A means for collecting operation logs from user terminals;

[1111] A means to analyze the collected operation logs and evaluate the performance of each version;

[1112] A means for selecting the optimal version based on the evaluation results and updating the service;

[1113] A means of installing an application on a smart device to measure user click rates and time spent on the site.

[1114] a means for transmitting the measured performance data to a server and analyzing the data in real time;

[1115] A system including:

[1116] (Claim 2)

[1117] The system according to claim 1, wherein the collected operation logs are transmitted to a server in real time and analyzed using statistical methods and machine learning algorithms.

[1118] (Claim 3)

[1119] 2. The system of claim 1, wherein data collection and analysis are automatically performed to select the optimal version, and the selected optimal version is applied to all users.

[1120] "Example 2: Combining Emotion Engines"

[1121] (Claim 1)

[1122] an artificial intelligence means for idea generation;

[1123] means for creating different versions of the service based on a plurality of improvement ideas generated by the artificial intelligence means;

[1124] means for randomly distributing the different versions of the service to a user terminal;

[1125] A means for collecting operation logs from user terminals;

[1126] means for collecting user emotion data by an emotion engine;

[1127] A means for analyzing the collected operation logs and sentiment data to evaluate the performance of each version;

[1128] A means for selecting the optimal version based on the evaluation results and updating the service;

[1129] A system including:

[1130] (Claim 2)

[1131] 2. The system according to claim 1, wherein the collected operation log and emotion data are transmitted to a server in real time.

[1132] (Claim 3)

[1133] 2. The system of claim 1, wherein the selected optimal version is automatically applied to all users.

[1134] "Application example 2 when combining emotion engines"

[1135] (Claim 1)

[1136] an artificial intelligence means for idea generation;

[1137] means for creating different versions of the service based on a plurality of improvement ideas generated by the artificial intelligence means;

[1138] means for randomly distributing the different versions of the service to a user terminal;

[1139] means for collecting operation logs and emotion data from user terminals;

[1140] A means to evaluate the performance of each version by analyzing the collected operation logs and sentiment data;

[1141] A means for selecting the optimal version based on the evaluation results and updating the service;

[1142] A system including:

[1143] (Claim 2)

[1144] 2. The system according to claim 1, wherein the collected operation log and emotion data are transmitted to a server in real time.

[1145] (Claim 3)

[1146] 2. The system of claim 1, wherein the selected optimal version is automatically applied to all users. [Explanation of symbols]

[1147] 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. an artificial intelligence means for idea generation; means for creating different versions of the service based on a plurality of improvement ideas generated by the artificial intelligence means; means for randomly distributing the different versions of the service to a user terminal; A means for collecting operation logs from user terminals; A means to analyze the collected operation logs and evaluate the performance of each version; A means for selecting the optimal version based on the evaluation results and updating the service; A system including:

2. 2. The system according to claim 1, wherein the collected operation log is transmitted to a server in real time.

3. 2. The system of claim 1, wherein the selected optimal version is automatically applied to all users.

Images