system

Abstract

We provide the system. [Solution] A means of acquiring user data from an information aggregation device, A means for analyzing acquired user data and generating multiple user classifications, A means for automatically generating a human model based on each user classification, A means for creating a conversational agent that mimics a generated human model, A means for users to conduct dialogue training using a dialogue agent, A means of evaluating the results of dialogue training and providing feedback to users, A system that includes this.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of the chatbot's character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] [[ID=3,4]] In the field of online training, there is a problem that it is difficult to encourage participants to respond to unexpected questions and actively participate. To solve this problem, it is required to cultivate the practical ability to handle various end-users who appear on-site. However, it is difficult to imitate the interaction with actual users with conventional methods, so a quick and effective solution is needed.

Means for Solving the Problems

[0005] This invention provides a system that acquires user data from an information aggregation device and generates multiple user classifications by analyzing that data. Based on each user classification, it automatically generates human models and creates dialogue agents that mimic these models. Using these dialogue agents, users can conduct dialogue training in a realistic environment, and the system evaluates the results of the dialogue and provides feedback to the user. In this way, it promotes the ability to respond to unexpected situations and the engagement of participants, thereby improving practical skills.

[0006] An "information aggregation device" is a device used to collect and store vast amounts of information, such as user data.

[0007] "User data" refers to any form of information related to a user, such as their behavioral history, interests, and profile information.

[0008] "Analysis" is the process of analyzing collected data to derive useful information and patterns.

[0009] "User classification" is the result of grouping users based on specific attributes or behavioral patterns, based on analysis.

[0010] A "human model" is a virtual entity that mimics the profile of a typical user, built based on a specific user classification.

[0011] A "dialogue agent" is an artificial intelligence program built on a human model that has the ability to converse with users.

[0012] "Dialogue training" is an activity in which users improve their communication and problem-solving skills through interaction with a dialogue agent.

[0013] "Feedback" refers to information about evaluation and improvement points presented to users based on the results of dialogue training. [Brief explanation of the drawing]

[0014] [Figure 1] It is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] It is a conceptual diagram showing an example of the main functions of a data processing device and a smart device according to the first embodiment. [Figure 3] It is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] It is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] It is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] It is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] It is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] It is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] It shows an emotion map to which a plurality of emotions are mapped. [Figure 10] It shows an emotion map to which a plurality of emotions are mapped. [Figure 11] It is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Example 2 when an emotion engine is combined. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when an emotion engine is combined.

Modes for Carrying Out the Invention

[0015] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0016] First, the terms used in the following description will be explained.

[0017] In the following embodiments, the numbered processor (hereinafter simply referred to as "processor") may be one arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be one type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0018] In the following embodiments, the numbered RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0019] In the following embodiments, the numbered storage is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, etc.

[0020] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0021] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0022] [First Embodiment]

[0023] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0024] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0025] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0026] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0027] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0028] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0029] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0030] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0031] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0032] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0033] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0034] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0035] This invention is a system for improving the quality of online training and instruction. The system begins by acquiring a large amount of user data from an information aggregation device and analyzing it to identify various user classifications. Based on the analysis results, it automatically generates human models and creates conversational agents based on each model.

[0036] The server collects user data and analyzes it using machine learning algorithms. Clustering techniques are used to group users with similar behavioral patterns and attributes, and based on this, a characteristic human model is generated. This model includes name, age, interests, and typical questions and behavioral patterns.

[0037] Next, the server builds a conversational agent based on the generated human model. This agent can interact with the user using natural language processing, asking various questions and processing the user's responses.

[0038] The terminal functions as an entry point to this system, providing an interface for users to undergo dialogue training. Users interact through the terminal, responding to situations and questions provided by the agent. In this process, users' communication skills and problem-solving abilities are tested, and their skills are improved.

[0039] The server has the capability to analyze the results of dialogue training and evaluate the user's responses. It generates feedback based on various criteria, such as accuracy, speed, and creative problem-solving ability. This feedback is presented to the user through the terminal and can be used to improve future performance.

[0040] Specific example

[0041] For example, when implemented as part of a company's training program for new employees, the server can generate diverse personas from past data, such as "first-time project participants" or "users asking technical questions." Using conversational agents created based on these personas, users simulate realistic scenarios on their devices. For instance, the agent might ask a question like, "How would you handle this situation?" and the user responds. The server then analyzes the response and provides feedback, such as, "The answer is well-structured, but you need to respond more quickly." In this way, users can improve their ability to handle real-world situations in the workplace.

[0042] The following describes the processing flow.

[0043] Step 1:

[0044] The server retrieves user data from the data aggregation device. This data includes the user's behavioral history, profile, and interests. The retrieved data is anonymized to protect privacy.

[0045] Step 2:

[0046] The server performs data analysis by applying machine learning algorithms based on the acquired user data. Using clustering techniques, it classifies users with similar patterns and generates typical user classifications.

[0047] Step 3:

[0048] The server automatically generates a human model based on the generated user classification. This model includes attributes, interests, and behavioral patterns corresponding to each classification. The generated model is stored in a database.

[0049] Step 4:

[0050] The server creates a conversational agent based on a stored human model. Utilizing natural language processing technology, this agent can interact with the user in real time.

[0051] Step 5:

[0052] The terminal provides the user with an interface for dialogue training. Through this interface, the user can begin a simulation with a dialogue agent.

[0053] Step 6:

[0054] The user responds to questions and issues raised by the conversational agent. The dialogue is based on realistic scenarios and aims to improve the user's skills.

[0055] Step 7:

[0056] The server records user responses during interaction and generates logs for later detailed analysis. It evaluates the accuracy of responses, response speed, and originality of problem-solving.

[0057] Step 8:

[0058] The server generates feedback based on the analysis results and evaluates the dialogue training. This feedback includes areas for improvement, successes, and suggestions for the next training session.

[0059] Step 9:

[0060] The device presents the generated feedback to the user, highlighting areas for improvement and supporting continuous skill development.

[0061] (Example 1)

[0062] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0063] In online training and development, there is a need to provide effective systems for individualized instruction and evaluation that meet the diverse needs of users. Currently, general training systems do not adequately provide personalized feedback and interactive functions, making efficient skill improvement difficult. Therefore, the challenge is to realize training that is more flexible and tailored to the characteristics of each user.

[0064] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0065] In this invention, the server includes means for acquiring information from a data aggregation device, means for analyzing the acquired information and generating multiple information classifications, means for automatically generating a person model based on each information classification, means for grouping information using a clustering method, and means for constructing a dialogue program using natural language processing technology. This enables personalized dialogue and feedback tailored to the individual characteristics of each user, and allows for efficient skill improvement.

[0066] A "data aggregation device" is a device used to collect information from various sources and manage it centrally.

[0067] "Information classification" refers to the process of dividing collected data into groups based on specific criteria, or the result of such classification.

[0068] A "person model" is a hypothetical person who possesses typical attributes and behavioral patterns, generated based on a specific classification of information.

[0069] A "dialogue program" is software that uses natural language processing technology to communicate with users.

[0070] Clustering is a technique used in data analysis to group data with similar characteristics together.

[0071] "Natural language processing technology" is a technology that enables computers to understand and process human language.

[0072] "Feedback" is information that provides evaluations and suggestions for improvement regarding actions or responses.

[0073] This system is built to improve the quality of online training and instruction. The server begins by collecting large amounts of user data using data aggregation equipment. The collected data includes user behavior history and interaction history, which is then analyzed using machine learning algorithms. Specifically, clustering algorithms (e.g., k-means) are used to group users with similar behavioral patterns and attributes.

[0074] Next, the server generates a characteristic person model based on the analysis results. This model includes typical age groups, interests, and common question patterns, forming the basis for more personalized training. Based on the generated person model, a dialogue program is built using natural language processing techniques. This technique utilizes natural language processing libraries such as NLTK and spaCy.

[0075] The terminal provides an interface for the user to interact with the dialogue program. The user interacts with the agent through the terminal, responding to the presented situations and questions. This process tests the user's communication skills and problem-solving abilities.

[0076] The server also analyzes the results of the dialogue training and evaluates the user's responses. Evaluation criteria include the accuracy and speed of responses, as well as the presentation of creative solutions, and feedback is generated based on these criteria. This feedback is presented to the user through the terminal and contributes to further skill improvement.

[0077] For example, in a training program for new employees, the server generates diverse personas from past data, such as "first-time project participant" or "user who actively asks technical questions." Using these generated personas, a realistic simulation is conducted on the terminal, and an agent asks questions such as, "How would you handle this situation?" After the user responds, the server analyzes the response and provides feedback such as, "The response is well-structured, but speed is required."

[0078] Examples of prompts include, "Create a persona to generate a dialogue program for new employees," and "List the questions the agent will ask the user."

[0079] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0080] Step 1:

[0081] The server collects user data from data aggregation devices. This data includes user login history, training history, and interaction history. Based on this data, the server extracts characteristic information necessary for the training program. The output here is user data containing characteristic information. Specifically, this involves extracting and filtering information from the database.

[0082] Step 2:

[0083] The server analyzes the collected user data. Specifically, it uses a machine learning algorithm to perform clustering on the data. k-means clustering is used for this process, grouping users with similar behavioral patterns or common attributes. The output is the clustering result of the users. The operation includes steps such as executing the algorithm and extracting features for each cluster.

[0084] Step 3:

[0085] The server generates person models based on the clustering results. Typical age groups, interests, and common question patterns are included to create models that reflect the characteristics of each cluster. The input is the clustering results, and the output is the person model for each cluster. Specifically, the profile information is registered in a database.

[0086] Step 4:

[0087] The server builds a dialogue program based on the generated human model. A natural language processing library (e.g., NLTK or spaCy) is used to program the user interaction scenario. The input is the human model, and the output is the dialogue program. Operation includes code implementation and testing.

[0088] Step 5:

[0089] The terminal provides an interface for the user to interact with the agent. The user participates in the interaction and responds to the scenarios and questions presented. The input is the agent's questions, and the output is the user's answers. In terms of specific actions, the user interface is manipulated.

[0090] Step 6:

[0091] The server analyzes user responses to evaluate the results of dialogue training. The evaluation criteria include the accuracy, speed, and creativity of the responses. The input is a log of user responses, and the output is feedback including areas for improvement. The operation involves running an analysis algorithm and saving the feedback results to a database.

[0092] (Application Example 1)

[0093] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0094] In today's business environment, practical problem-solving skills and communication abilities are extremely important. However, traditional training methods lack sufficient training that closely resembles real-world situations, making it difficult for staff to effectively improve their skills. Therefore, there is a need for a system that enables training in environments close to actual work situations and effectively improves skills.

[0095] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0096] In this invention, the server includes means for acquiring user data from an information aggregation device, means for analyzing the acquired user data and generating multiple user classifications, and means for automatically generating a human model based on each user classification. This makes it possible to provide a simulation environment close to the real world via a smart device using a dialogue agent that mimics the generated human model, enabling users to improve their skills to suit real-world situations through dialogue training.

[0097] An "information aggregation device" is a device or platform for collecting user data and providing it to a system.

[0098] "User data analysis" is the process of using collected data to analyze the behavior and characteristics of individual users and extract information from them.

[0099] "User classification" refers to categories used to group users with similar characteristics and behavioral patterns based on analyzed data.

[0100] A "human model" is a virtual person profile automatically generated based on analysis results, and it includes typical attributes and behavioral patterns.

[0101] A "dialogue agent" is a program built on a human model and is software capable of engaging in dialogue with users.

[0102] A "simulation environment via smart devices" is a virtual training ground that reproduces situations close to the real world, provided through devices such as smartphones.

[0103] "Feedback" refers to information provided to users for evaluation and improvement based on the results of dialogue training.

[0104] To implement this invention, it is necessary to construct a system in which a server, a terminal, and a user interact with each other. The server acquires user data from an information aggregation device, analyzes it, and generates user classifications. For the analysis, it utilizes Scikit-learn, a machine learning library using Python, and identifies similar user groups by clustering user behavior patterns. Based on the obtained classification information, a human model is automatically generated. This human model mimics the user's age, interests, and typical questions and behavior patterns.

[0105] The server then creates a conversational agent based on the generated human model. This conversational agent uses NLTK, an open-source NLP library, to implement natural language processing. This agent can respond to user questions and provide various scenarios through dialogue.

[0106] The terminal provides an interface for users to interact with the system. Typically, smart devices (smartphones or tablets) fulfill this role and are operated via a browser or dedicated app. The terminal allows users to receive training in a simulated environment using conversational agents, helping staff improve their skills in situations close to real-world scenarios.

[0107] Users participate in conversational simulations using smart devices. This simulation environment replicates real-world customer service scenarios, presenting prompts such as, "Consider the following situation: A customer from overseas is unsure about their order. How can you assist them?" User responses are sent to the server sequentially and recorded as conversational training data. The server analyzes this data and provides quick and effective feedback to help improve user skills.

[0108] This system allows store staff to effectively improve their customer service skills through simulations and feedback via smart devices.

[0109] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0110] Step 1:

[0111] The server retrieves user data from the information aggregation device. This data includes the user's behavioral history and basic attribute information. The user data retrieved as input is stored within the system.

[0112] Step 2:

[0113] The server analyzes the acquired user data and performs data clustering. It uses the machine learning library Scikit-learn to perform PCA (Principal Component Analysis) and extract important features. This groups users with similar features and generates user classifications as output.

[0114] Step 3:

[0115] The server automatically generates a human model based on the generated user classification. This human model includes typical attributes and behavioral patterns. This process primarily involves generating data structures, and the output provides profile information for the conversational agent.

[0116] Step 4:

[0117] The server uses the NLTK natural language processing library to build a conversational agent. This agent is designed using profile information obtained from a human model. It receives the profile as input and generates a script simulating user interaction as output.

[0118] Step 5:

[0119] The terminal provides a user-accessible interface and connects the user with the conversational agent. It runs on a smart device via an application or browser. It receives agent scripts as input and generates visual and audio user interfaces as output.

[0120] Step 6:

[0121] The user conducts simulation training with a conversational agent via a terminal. A scenario including prompts is displayed, and the user inputs their response. The input is the user's response and is sent to the server as output.

[0122] Step 7:

[0123] The server analyzes user responses and evaluates the training. It performs Japanese text analysis to measure the accuracy and speed of responses. As a result, feedback information is generated and returned to the terminal as output.

[0124] Step 8:

[0125] The terminal presents feedback from the server to the user. It displays feedback that includes areas for improvement and effective approaches, allowing the user to utilize this information in future training sessions. It receives feedback information as input and displays it on the screen as output.

[0126] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0127] This invention is an online training system that integrates emotion recognition functionality, taking into account the user's emotions to provide more effective dialogue training. The system begins by acquiring vast amounts of user data from an information aggregation device and analyzing it to generate multiple user classifications. Based on the analysis results, it automatically generates a human model and creates a dialogue agent that mimics it.

[0128] The server processes the acquired user data using a machine learning algorithm and classifies users into multiple categories. Based on these results, it generates a human model corresponding to each category. Each model includes name, age, interests, behavioral patterns, and typical questions asked.

[0129] The emotion engine is an additional element that recognizes the user's facial expressions, tone of voice, and the emotions expressed in the input text in real time via the device. The recognized emotion information is processed on the server, and the results are reflected in the conversational agent's response content. This allows the agent to provide more human-like responses, enabling calm conversations and emotionally empathetic communication.

[0130] The terminal provides an interface for dialogue training, creating an environment where users can simulate with a dialogue agent. Users use this interface to address a variety of questions and situations based on realistic scenarios.

[0131] The server also analyzes the results of the dialogue training and generates emotional responses and feedback based on the user's emotional state. For example, if the user is feeling stressed, feedback such as recommending a more relaxed approach will be provided based on that information.

[0132] Specific example

[0133] Let's consider a scenario involving new employee training at a company. The server analyzes past data and generates diverse human models, such as "users struggling with technical questions" and "users participating in a project for the first time." Based on these models, a conversational agent is created to ask the user questions like, "What do you think about this new system?" As the user answers, the agent adjusts its response based on the emotions recognized by the emotion engine. For example, if the user shows anxiety, the agent will provide an emotionally sensitive response such as, "Don't worry, this information will be explained in more detail later." The server analyzes this conversation and provides feedback, such as, "The quick response is good, but a calmer tone is needed." In this way, users can improve their emotional awareness along with their actual conversational skills.

[0134] The following describes the processing flow.

[0135] Step 1:

[0136] The server retrieves vast amounts of user data from the data aggregation device. This data includes user behavior history, profiles, and text data. After data retrieval, it is anonymized to protect privacy.

[0137] Step 2:

[0138] The server analyzes the acquired user data and uses machine learning algorithms to divide users into multiple categories. Through clustering, it forms user groups with similar characteristics and generates a human model based on them.

[0139] Step 3:

[0140] The server creates a conversational agent based on the generated human model. Equipped with natural language processing technology, this agent enables real-time interaction with the user and constructs questions and responses tailored to specific scenarios.

[0141] Step 4:

[0142] The terminal provides the user with an interface for dialogue training. The user uses this interface to initiate dialogue simulations and experience interaction with a human-model-based agent.

[0143] Step 5:

[0144] The emotion engine built into the device analyzes the user's facial expressions, voice tone, and input text to recognize the user's emotions. This information is transmitted to the server in real time.

[0145] Step 6:

[0146] The server analyzes the received emotion recognition data and incorporates the results into the dialogue agent's response. This allows the agent to reconstruct questions to match the user's emotions and provide appropriate, empathetic responses.

[0147] Step 7:

[0148] Users interact with conversational agents and experience emotion-based feedback, which helps them enhance their realistic communication skills.

[0149] Step 8:

[0150] The server analyzes and evaluates the overall results of the dialogue training. This process includes an assessment based on the appropriateness, speed, and perceived emotions of the user's responses.

[0151] Step 9:

[0152] The server generates customized feedback based on the analysis results and provides it to the user through the terminal. The feedback includes areas for improvement in emotional response and key success factors, encouraging the user to further improve their skills.

[0153] (Example 2)

[0154] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0155] In recent years, interactive training systems have become widespread, but many of them fail to provide effective dialogue that takes user emotions into account. Conventional systems lack sufficient technology to adjust responses using real-time user emotion information, which can result in a decline in the quality of the dialogue. Therefore, there is a need for dialogue systems that allow users to gain a greater sense of satisfaction and learning effectiveness.

[0156] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0157] In this invention, the server includes means for acquiring user information from an information processing device, means for analyzing the acquired user information and generating multiple user classifications, and means for automatically generating a human-like model based on each user classification. This enables dialogue by a conversational agent that is adjusted according to the user's emotional state, providing a sophisticated communication experience that is sensitive to emotions.

[0158] An "information processing device" is a device for collecting and managing user data, and has the function of deriving new information by analyzing the obtained data.

[0159] "User information" is a general term for information related to a user that the system acquires, and includes data such as an individual's profile and past usage history.

[0160] "User classification" refers to the process of classifying and grouping users according to specific criteria based on acquired user information.

[0161] A "human-style model" is a model generated based on user classification, and serves as a foundation for mimicking user characteristics and behavioral patterns.

[0162] A "conversational agent" is a program or software designed to mimic human behavior models and engage in dialogue with users.

[0163] An "emotion recognition device" is a device that analyzes a user's emotional state in real time and provides appropriate information accordingly.

[0164] "Information retention means" refers to means that provide a function to save the characteristics and profiles of a human-like model and retrieve them later.

[0165] "Situation adjustment means" are methods used to dynamically adjust the questions and situations presented by the conversational agent during conversation training, and are used to make the dialogue more effective.

[0166] This invention describes a specific embodiment of an online training system that takes user emotions into consideration. The system consists of a server, a terminal, and a user.

[0167] The server functions as an information processing device, collecting user information. Database management software and cloud storage are used for this purpose. The server analyzes the collected user information using machine learning frameworks such as TENSORFLOW® or PyTorch, classifying users into multiple user categories. Based on the analyzed data, a human-like model is automatically generated, which is then used for further conversation simulations.

[0168] The terminal provides an interface with the user and recognizes the user's facial expressions, tone of voice, and text emotions in real time via an emotion recognition device. This uses a webcam and microphone, and OpenCV and speech processing libraries are used to analyze the emotion data. This real-time data is sent to a server, enabling a conversational agent to generate tailored responses based on the emotion data.

[0169] Users engage in conversational training with a conversational agent through this device. The conversational agent is built based on various human-like models and provides emotionally resonant responses to the user. For example, if the user feels anxious, the agent can respond with something like, "It's okay, I'll explain the details step by step later."

[0170] A concrete example is its use when new employees practice for job interviews. The server analyzes past interactions and generates models such as "stressed applicants." Based on this, a conversational agent asks typical questions such as "Why did you apply for this job?" and provides feedback tailored to the user's emotional perception, thereby enhancing the user's practice effectiveness.

[0171] An example of a prompt is, "Please tell me how to communicate in a way that takes the user's emotional state into consideration and is sensitive to their feelings." By using this prompt, the generative AI model can be given instructions to perform a more specific simulation.

[0172] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0173] Step 1:

[0174] The server collects user information through an information processing device. Inputs include user profile data and past conversation logs, which are retrieved using database management software. Outputs are structured datasets, which are used in subsequent analysis steps.

[0175] Step 2:

[0176] The server analyzes the collected user information using a machine learning framework. The input is the dataset obtained in Step 1. Based on this, the server analyzes the data using TensorFlow or PyTorch and divides each user into multiple user classifications based on their features. The output is the category information into which the users have been classified.

[0177] Step 3:

[0178] The server automatically generates human-like models based on user classification. The input is the category information created in step 2, and the process generates model parameters representative of each category. The server uses a model template and sets parameters such as name, age, and interests to create the human-like models. The output is a set of human-like models for each user classification.

[0179] Step 4:

[0180] The device performs emotion recognition through its interface with the user. Inputs include the user's facial expressions, voice tone, and text input, which are captured via a webcam and microphone. The device uses OpenCV and speech processing libraries to analyze this data in real time and extract the user's emotional information. The output is the analyzed emotional state of the user.

[0181] Step 5:

[0182] The server adjusts the conversational agent's response using emotion information received from the terminal. The inputs are the emotion state obtained in step 4 and the human-like model generated in step 3. Based on these inputs, the server applies the output response to the conversational agent, and the agent generates a response appropriate to the user's emotion. The output is the adjusted response of the conversational agent.

[0183] Step 6:

[0184] The user engages in conversational training with a conversational agent on a terminal. The input is the conversational agent's responses, which were adjusted in step 5. The user responds to questions and information from the agent and practices conversational skills through scenarios provided by the agent. The output is the user's conversational training results, and this data is stored on the server and used for subsequent analysis.

[0185] Step 7:

[0186] The server evaluates the results of the conversation training and provides feedback to the user. The input is the conversation training results obtained in step 6. The server analyzes this and makes an evaluation based on the user's skills and emotions. The output is feedback information for the user, which can be used to improve the next conversation training session. This feedback allows the user to clearly understand areas for self-improvement.

[0187] (Application Example 2)

[0188] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0189] In retail store customer service, appropriately understanding and responding to customer emotions is a challenging task. Traditional customer service methods make it difficult to instantly analyze a customer's facial expressions and tone of voice and respond accordingly, resulting in difficulties in improving customer satisfaction.

[0190] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0191] In this invention, the server includes means for acquiring personal data from an information processing device, means for analyzing the acquired personal data and generating a personal classification, and means for recognizing the user's emotional state in real time using emotion analysis means and adjusting the response of the conversational agent based on the recognized emotional state. This enables the provision of appropriate customer service in accordance with the customer's emotions and improves customer satisfaction.

[0192] An "information processing device" is a device that acquires and analyzes personal data and processes that information for a specific purpose.

[0193] "Personal data" refers to information about a user, which is used to classify and identify the user's characteristics through analysis.

[0194] "Personal classification" refers to categories created by classifying users based on specific characteristics and attributes, using analyzed personal data.

[0195] A "human characteristics model" is a model generated based on individual classification, and it imitates human characteristics and behavioral patterns.

[0196] A "dialogue agent" is a program created based on a human characteristics model to simulate conversations with users.

[0197] "Emotional analysis means" refers to technology that analyzes a user's facial expressions and tone of voice to recognize their emotional state in real time.

[0198] "Means for adjusting responses" refers to a function that modifies the content of the dialogue agent's response based on the recognized emotional state, thereby enabling dialogue that is empathetic to the user's emotions.

[0199] This invention provides an emotion recognition system aimed at improving customer service in retail stores. This system integrates and operates with an information processing device, a dialogue agent, and emotion analysis means.

[0200] First, the information processing device acquires personal data via terminals, smart glasses, and smartphones installed in retail stores. This data includes customer facial expressions and voice tone, and is used to generate personal classifications. Next, the server automatically generates a human characteristics model based on the acquired personal data. The generated model is then mimicked within the dialogue agent's program.

[0201] The conversational agent utilizes real-time sentiment analysis during conversations with customers. This allows it to analyze the customer's emotional state and adjust its response accordingly. For example, if the agent detects that the customer is irritated, it will respond in a calmer tone appropriate to that state. To achieve this functionality, APIs such as the Affectiva API can be used for sentiment analysis.

[0202] Store employees, as users of this system, can grasp customer sentiment information in real time and receive suggestions for customer service techniques based on that information. This enables more sophisticated customer service.

[0203] As a concrete example, consider a scenario where a store clerk is wearing smart glasses and a customer asks for product details. If the emotion analysis system determines that the customer is expressing dissatisfaction or doubt, a suggestion such as "Let's talk in more detail about the features of this product" will be displayed on the smart glasses' screen. At this point, the optimal response will be suggested to the generative AI model through a prompt message such as "Advise me on how to respond if the customer is showing signs of frustration."

[0204] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0205] Step 1:

[0206] The server acquires personal data via terminals in retail stores, smart glasses, and smartphones. Inputs include facial image data and voice recordings of customers, while outputs are these data. The data is collected in real time and used as material for customer sentiment analysis.

[0207] Step 2:

[0208] The server uses a machine learning model with the acquired personal data to perform emotion analysis. The input is the facial expression images and audio data obtained in step 1, and the output is the customer's emotional state (e.g., joy, anger, anxiety). The Affectiva API is used to analyze the data and instantly quantify the customer's emotions.

[0209] Step 3:

[0210] The server generates a personal classification based on the analyzed emotional state. The input is the emotional state obtained in step 2, and the output is the personal classification result. Based on this result, the server determines what response is appropriate.

[0211] Step 4:

[0212] The server generates a human characteristics model based on the personal classification results and reflects it in the conversational agent. The input is the personal classification result obtained in step 3, and the output is the human characteristics model. A prompt sentence is sent to the generated AI model to obtain the data necessary to adjust the response.

[0213] Step 5:

[0214] The terminal interacts with customers through a conversational agent. The input is the human feature model information obtained in step 4, and the output is a suggested response to the customer. The suggested response is displayed in real time on the smart glasses or terminal screen, and the store staff uses this to provide customer service.

[0215] Step 6:

[0216] The store clerk, acting as the user, provides appropriate customer service to customers based on the responses and suggestions presented by the terminal. The input is the response suggestion obtained in step 5, and the output is the actual result of the interaction with the customer. The clerk observes customer feedback and reports it to the server via the terminal as needed.

[0217] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0218] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0219] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0220] [Second Embodiment]

[0221] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0222] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0223] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0224] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0225] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0226] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0227] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0228] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0229] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0230] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0231] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0232] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0233] This invention is a system for improving the quality of online training and instruction. The system begins by acquiring a large amount of user data from an information aggregation device and analyzing it to identify various user classifications. Based on the analysis results, it automatically generates human models and creates conversational agents based on each model.

[0234] The server collects user data and analyzes it using machine learning algorithms. Clustering techniques are used to group users with similar behavioral patterns and attributes, and based on this, a characteristic human model is generated. This model includes name, age, interests, and typical questions and behavioral patterns.

[0235] Next, the server builds a conversational agent based on the generated human model. This agent can interact with the user using natural language processing, asking various questions and processing the user's responses.

[0236] The terminal functions as an entry point to this system, providing an interface for users to undergo dialogue training. Users interact through the terminal, responding to situations and questions provided by the agent. In this process, users' communication skills and problem-solving abilities are tested, and their skills are improved.

[0237] The server has the capability to analyze the results of dialogue training and evaluate the user's responses. It generates feedback based on various criteria, such as accuracy, speed, and creative problem-solving ability. This feedback is presented to the user through the terminal and can be used to improve future performance.

[0238] Specific example

[0239] For example, when implemented as part of a company's training program for new employees, the server can generate diverse personas from past data, such as "first-time project participants" or "users asking technical questions." Using conversational agents created based on these personas, users simulate realistic scenarios on their devices. For instance, the agent might ask a question like, "How would you handle this situation?" and the user responds. The server then analyzes the response and provides feedback, such as, "The answer is well-structured, but you need to respond more quickly." In this way, users can improve their ability to handle real-world situations in the workplace.

[0240] The following describes the processing flow.

[0241] Step 1:

[0242] The server retrieves user data from the data aggregation device. This data includes the user's behavioral history, profile, and interests. The retrieved data is anonymized to protect privacy.

[0243] Step 2:

[0244] The server performs data analysis by applying machine learning algorithms based on the acquired user data. Using clustering techniques, it classifies users with similar patterns and generates typical user classifications.

[0245] Step 3:

[0246] The server automatically generates a human model based on the generated user classification. This model includes attributes, interests, and behavioral patterns corresponding to each classification. The generated model is stored in a database.

[0247] Step 4:

[0248] The server creates a conversational agent based on a stored human model. Utilizing natural language processing technology, this agent can interact with the user in real time.

[0249] Step 5:

[0250] The terminal provides the user with an interface for dialogue training. Through this interface, the user can begin a simulation with a dialogue agent.

[0251] Step 6:

[0252] The user responds to questions and issues raised by the conversational agent. The dialogue is based on realistic scenarios and aims to improve the user's skills.

[0253] Step 7:

[0254] The server records user responses during interaction and generates logs for later detailed analysis. It evaluates the accuracy of responses, response speed, and originality of problem-solving.

[0255] Step 8:

[0256] The server generates feedback based on the analysis results and evaluates the dialogue training. This feedback includes areas for improvement, successes, and suggestions for the next training session.

[0257] Step 9:

[0258] The device presents the generated feedback to the user, highlighting areas for improvement and supporting continuous skill development.

[0259] (Example 1)

[0260] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0261] In online training and development, there is a need to provide effective systems for individualized instruction and evaluation that meet the diverse needs of users. Currently, general training systems do not adequately provide personalized feedback and interactive functions, making efficient skill improvement difficult. Therefore, the challenge is to realize training that is more flexible and tailored to the characteristics of each user.

[0262] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0263] In this invention, the server includes means for acquiring information from a data aggregation device, means for analyzing the acquired information and generating multiple information classifications, means for automatically generating a person model based on each information classification, means for grouping information using a clustering method, and means for constructing a dialogue program using natural language processing technology. This enables personalized dialogue and feedback tailored to the individual characteristics of each user, and allows for efficient skill improvement.

[0264] A "data aggregation device" is a device used to collect information from various sources and manage it centrally.

[0265] "Information classification" refers to the process of dividing collected data into groups based on specific criteria, or the result of such classification.

[0266] A "person model" is a hypothetical person who possesses typical attributes and behavioral patterns, generated based on a specific classification of information.

[0267] A "dialogue program" is software that uses natural language processing technology to communicate with users.

[0268] Clustering is a technique used in data analysis to group data with similar characteristics together.

[0269] "Natural language processing technology" is a technology that enables computers to understand and process human language.

[0270] "Feedback" is information that provides evaluations and suggestions for improvement regarding actions or responses.

[0271] This system is built to improve the quality of online training and instruction. The server begins by collecting large amounts of user data using data aggregation equipment. The collected data includes user behavior history and interaction history, which is then analyzed using machine learning algorithms. Specifically, clustering algorithms (e.g., k-means) are used to group users with similar behavioral patterns and attributes.

[0272] Next, the server generates a characteristic person model based on the analysis results. This model includes typical age groups, interests, and common question patterns, forming the basis for more personalized training. Based on the generated person model, a dialogue program is built using natural language processing techniques. This technique utilizes natural language processing libraries such as NLTK and spaCy.

[0273] The terminal provides an interface for the user to interact with the dialogue program. The user interacts with the agent through the terminal, responding to the presented situations and questions. This process tests the user's communication skills and problem-solving abilities.

[0274] The server also analyzes the results of the dialogue training and evaluates the user's responses. Evaluation criteria include the accuracy and speed of responses, as well as the presentation of creative solutions, and feedback is generated based on these criteria. This feedback is presented to the user through the terminal and contributes to further skill improvement.

[0275] For example, in a training program for new employees, the server generates diverse personas from past data, such as "first-time project participant" or "user who actively asks technical questions." Using these generated personas, a realistic simulation is conducted on the terminal, and an agent asks questions such as, "How would you handle this situation?" After the user responds, the server analyzes the response and provides feedback such as, "The response is well-structured, but speed is required."

[0276] Examples of prompt texts include "Please create personas for generating an interactive program for new employees." and "Please list the questions that the agent will ask the user."

[0277] The flow of the specific process in Example 1 will be described using FIG. 11.

[0278] Step 1:

[0279] The server collects user data from the data aggregation device. This data includes the user's login history, training history, and interaction history. Based on this data, the server extracts the feature information required for the training program. The output here is the user data including the feature information. Specific operations include information extraction and filtering from the database.

[0280] Step 2:

[0281] The server analyzes the collected user data. Specifically, clustering is performed on the data using a machine learning algorithm. The k-means clustering is used for this task to group users with similar behavior patterns and common attributes. The output is the clustering result of the users. The operations include executing the algorithm and extracting features for each cluster.

[0282] Step 3:

[0283] The server generates a persona model based on the clustering result. To create a model that reflects the characteristics of each cluster, it includes typical age groups, interests, and general question patterns, etc. The input is the clustering result, and the output is the persona model for each cluster. Specific operations include registering profile information in the database.

[0284] Step 4:

[0285] The server constructs a dialogue program based on the generated person model. Using a natural language processing library (e.g., NLTK or spaCy), the dialogue scenario with the user is programmed. The input is the person model, and the output is the dialogue program. The operations include code implementation and testing.

[0286] Step 5:

[0287] The terminal provides an interface for the user to interact with the agent. The user participates in the dialogue and responds to the presented scenarios and questions. The input is the question from the agent, and the output is the user's answer. Specifically, operations on the user interface are performed.

[0288] Step 6:

[0289] The server analyzes the user's responses to evaluate the results of the dialogue training. For the evaluation, the accuracy, speed, and creativity of the responses are used as criteria. The input is the user's answer log, and the output is the feedback including improvement points. The operations include running the analysis algorithm and saving the feedback results to the database.

[0290] (Application Example 1)

[0291] Next, Application Example 1 will be described. In the following description, the data processing device 12 is referred to as the "server", and the smart glasses 214 are referred to as the "terminal".

[0292] In the modern business environment, the ability to handle real-world situations and communication skills are very important. However, traditional training methods lack training in a form close to actual scenarios, making it difficult for staff to effectively improve their skills. Therefore, a mechanism that enables training in an environment close to the actual workplace and effectively improves skills is required.

[0293] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0294] In this invention, the server includes means for acquiring user data from an information aggregation device, means for analyzing the acquired user data and generating multiple user classifications, and means for automatically generating a human model based on each user classification. This makes it possible to provide a simulation environment close to the real world via a smart device using a dialogue agent that mimics the generated human model, enabling users to improve their skills to suit real-world situations through dialogue training.

[0295] An "information aggregation device" is a device or platform for collecting user data and providing it to a system.

[0296] "User data analysis" is the process of using collected data to analyze the behavior and characteristics of individual users and extract information from them.

[0297] "User classification" refers to categories used to group users with similar characteristics and behavioral patterns based on analyzed data.

[0298] A "human model" is a virtual person profile automatically generated based on analysis results, and it includes typical attributes and behavioral patterns.

[0299] A "dialogue agent" is a program built on a human model and is software capable of engaging in dialogue with users.

[0300] A "simulation environment via smart devices" is a virtual training ground that reproduces situations close to the real world, provided through devices such as smartphones.

[0301] "Feedback" refers to information provided to users for evaluation and improvement based on the results of dialogue training.

[0302] To implement this invention, it is necessary to construct a system in which a server, a terminal, and a user interact with each other. The server acquires user data from an information aggregation device, analyzes it, and generates user classifications. For the analysis, it utilizes Scikit-learn, a machine learning library using Python, and identifies similar user groups by clustering user behavior patterns. Based on the obtained classification information, a human model is automatically generated. This human model mimics the user's age, interests, and typical questions and behavior patterns.

[0303] The server then creates a conversational agent based on the generated human model. This conversational agent uses NLTK, an open-source NLP library, to implement natural language processing. This agent can respond to user questions and provide various scenarios through dialogue.

[0304] The terminal provides an interface for users to interact with the system. Typically, smart devices (smartphones or tablets) fulfill this role and are operated via a browser or dedicated app. The terminal allows users to receive training in a simulated environment using conversational agents, helping staff improve their skills in situations close to real-world scenarios.

[0305] Users participate in conversational simulations using smart devices. This simulation environment replicates real-world customer service scenarios, presenting prompts such as, "Consider the following situation: A customer from overseas is unsure about their order. How can you assist them?" User responses are sent to the server sequentially and recorded as conversational training data. The server analyzes this data and provides quick and effective feedback to help improve user skills.

[0306] With this system, staff in physical stores can effectively improve their customer service capabilities through simulations and feedback via smart devices.

[0307] The flow of the specific process in Application Example 1 will be described using FIG. 12.

[0308] Step 1:

[0309] The server acquires user data from the information aggregation device. This data includes the user's behavior history and basic attribute information. The acquired user data as input is stored in the system.

[0310] Step 2:

[0311] The server analyzes the acquired user data and performs data clustering. PCA (Principal Component Analysis) is performed using the Scikit-learn machine learning library to extract important features. Thereby, users with similar features are grouped, and user classification is generated as output.

[0312] Step 3:

[0313] The server automatically generates a human model based on the generated user classification. The human model includes typical attributes and behavior patterns. This process mainly includes the generation of a data structure, and profile information for the dialogue agent is provided as output.

[0314] Step 4:

[0315] The server constructs a dialogue agent by leveraging the NLTK natural language processing library. This agent is designed using the profile information obtained from the human model. Receiving the profile as input, a script assuming a dialogue with the user is created as output.

[0316] Step 5:

[0317] The terminal provides a user-accessible interface and connects the user with the conversational agent. It runs on a smart device via an application or browser. It receives agent scripts as input and generates visual and audio user interfaces as output.

[0318] Step 6:

[0319] The user conducts simulation training with a conversational agent via a terminal. A scenario including prompts is displayed, and the user inputs their response. The input is the user's response and is sent to the server as output.

[0320] Step 7:

[0321] The server analyzes user responses and evaluates the training. It performs Japanese text analysis to measure the accuracy and speed of responses. As a result, feedback information is generated and returned to the terminal as output.

[0322] Step 8:

[0323] The terminal presents feedback from the server to the user. It displays feedback that includes areas for improvement and effective approaches, allowing the user to utilize this information in future training sessions. It receives feedback information as input and displays it on the screen as output.

[0324] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0325] This invention is an online training system that integrates emotion recognition functionality, taking into account the user's emotions to provide more effective dialogue training. The system begins by acquiring vast amounts of user data from an information aggregation device and analyzing it to generate multiple user classifications. Based on the analysis results, it automatically generates a human model and creates a dialogue agent that mimics it.

[0326] The server processes the acquired user data using a machine learning algorithm and classifies users into multiple categories. Based on these results, it generates a human model corresponding to each category. Each model includes name, age, interests, behavioral patterns, and typical questions asked.

[0327] The emotion engine is an additional element that recognizes the user's facial expressions, tone of voice, and the emotions expressed in the input text in real time via the device. The recognized emotion information is processed on the server, and the results are reflected in the conversational agent's response content. This allows the agent to provide more human-like responses, enabling calm conversations and emotionally empathetic communication.

[0328] The terminal provides an interface for dialogue training, creating an environment where users can simulate with a dialogue agent. Users use this interface to address a variety of questions and situations based on realistic scenarios.

[0329] The server also analyzes the results of the dialogue training and generates emotional responses and feedback based on the user's emotional state. For example, if the user is feeling stressed, feedback such as recommending a more relaxed approach will be provided based on that information.

[0330] Specific example

[0331] Let's consider a scenario involving new employee training at a company. The server analyzes past data and generates diverse human models, such as "users struggling with technical questions" and "users participating in a project for the first time." Based on these models, a conversational agent is created to ask the user questions like, "What do you think about this new system?" As the user answers, the agent adjusts its response based on the emotions recognized by the emotion engine. For example, if the user shows anxiety, the agent will provide an emotionally sensitive response such as, "Don't worry, this information will be explained in more detail later." The server analyzes this conversation and provides feedback, such as, "The quick response is good, but a calmer tone is needed." In this way, users can improve their emotional awareness along with their actual conversational skills.

[0332] The following describes the processing flow.

[0333] Step 1:

[0334] The server retrieves vast amounts of user data from the data aggregation device. This data includes user behavior history, profiles, and text data. After data retrieval, it is anonymized to protect privacy.

[0335] Step 2:

[0336] The server analyzes the acquired user data and uses machine learning algorithms to divide users into multiple categories. Through clustering, it forms user groups with similar characteristics and generates a human model based on them.

[0337] Step 3:

[0338] The server creates a conversational agent based on the generated human model. Equipped with natural language processing technology, this agent enables real-time interaction with the user and constructs questions and responses tailored to specific scenarios.

[0339] Step 4:

[0340] The terminal provides the user with an interface for dialogue training. The user uses this interface to initiate dialogue simulations and experience interaction with a human-model-based agent.

[0341] Step 5:

[0342] The emotion engine built into the device analyzes the user's facial expressions, voice tone, and input text to recognize the user's emotions. This information is transmitted to the server in real time.

[0343] Step 6:

[0344] The server analyzes the received emotion recognition data and incorporates the results into the dialogue agent's response. This allows the agent to reconstruct questions to match the user's emotions and provide appropriate, empathetic responses.

[0345] Step 7:

[0346] Users interact with conversational agents and experience emotion-based feedback, which helps them enhance their realistic communication skills.

[0347] Step 8:

[0348] The server analyzes and evaluates the overall results of the dialogue training. This process includes an assessment based on the appropriateness, speed, and perceived emotions of the user's responses.

[0349] Step 9:

[0350] The server generates customized feedback based on the analysis results and provides it to the user through the terminal. The feedback includes areas for improvement in emotional response and key success factors, encouraging the user to further improve their skills.

[0351] (Example 2)

[0352] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0353] In recent years, interactive training systems have become widespread, but many of them fail to provide effective dialogue that takes user emotions into account. Conventional systems lack sufficient technology to adjust responses using real-time user emotion information, which can result in a decline in the quality of the dialogue. Therefore, there is a need for dialogue systems that allow users to gain a greater sense of satisfaction and learning effectiveness.

[0354] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0355] In this invention, the server includes means for acquiring user information from an information processing device, means for analyzing the acquired user information and generating multiple user classifications, and means for automatically generating a human-like model based on each user classification. This enables dialogue by a conversational agent that is adjusted according to the user's emotional state, providing a sophisticated communication experience that is sensitive to emotions.

[0356] An "information processing device" is a device for collecting and managing user data, and has the function of deriving new information by analyzing the obtained data.

[0357] "User information" is a general term for information related to a user that the system acquires, and includes data such as an individual's profile and past usage history.

[0358] "User classification" refers to the process of classifying and grouping users according to specific criteria based on acquired user information.

[0359] A "human-style model" is a model generated based on user classification, and serves as a foundation for mimicking user characteristics and behavioral patterns.

[0360] A "conversational agent" is a program or software designed to mimic human behavior models and engage in dialogue with users.

[0361] An "emotion recognition device" is a device that analyzes a user's emotional state in real time and provides appropriate information accordingly.

[0362] "Information retention means" refers to means that provide a function to save the characteristics and profiles of a human-like model and retrieve them later.

[0363] "Situation adjustment means" are methods used to dynamically adjust the questions and situations presented by the conversational agent during conversation training, and are used to make the dialogue more effective.

[0364] This invention describes a specific embodiment of an online training system that takes user emotions into consideration. The system consists of a server, a terminal, and a user.

[0365] The server functions as an information processing device, collecting user information. Database management software and cloud storage are used for this purpose. The server analyzes the collected user information using machine learning frameworks such as TensorFlow and PyTorch, classifying users into multiple user categories. Based on the analyzed data, a human-like model is automatically generated, which is then used for further conversation simulations.

[0366] The terminal provides an interface with the user and recognizes the user's facial expressions, tone of voice, and text emotions in real time via an emotion recognition device. This uses a webcam and microphone, and OpenCV and speech processing libraries are used to analyze the emotion data. This real-time data is sent to a server, enabling a conversational agent to generate tailored responses based on the emotion data.

[0367] Users engage in conversational training with a conversational agent through this device. The conversational agent is built based on various human-like models and provides emotionally resonant responses to the user. For example, if the user feels anxious, the agent can respond with something like, "It's okay, I'll explain the details step by step later."

[0368] A concrete example is its use when new employees practice for job interviews. The server analyzes past interactions and generates models such as "stressed applicants." Based on this, a conversational agent asks typical questions such as "Why did you apply for this job?" and provides feedback tailored to the user's emotional perception, thereby enhancing the user's practice effectiveness.

[0369] An example of a prompt is, "Please tell me how to communicate in a way that takes the user's emotional state into consideration and is sensitive to their feelings." By using this prompt, the generative AI model can be given instructions to perform a more specific simulation.

[0370] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0371] Step 1:

[0372] The server collects user information through an information processing device. Inputs include user profile data and past conversation logs, which are retrieved using database management software. Outputs are structured datasets, which are used in subsequent analysis steps.

[0373] Step 2:

[0374] The server analyzes the collected user information using a machine learning framework. The input is the dataset obtained in Step 1. Based on this, the server analyzes the data using TensorFlow or PyTorch and divides each user into multiple user classifications based on their features. The output is the category information into which the users have been classified.

[0375] Step 3:

[0376] The server automatically generates human-like models based on user classification. The input is the category information created in step 2, and the process generates model parameters representative of each category. The server uses a model template and sets parameters such as name, age, and interests to create the human-like models. The output is a set of human-like models for each user classification.

[0377] Step 4:

[0378] The device performs emotion recognition through its interface with the user. Inputs include the user's facial expressions, voice tone, and text input, which are captured via a webcam and microphone. The device uses OpenCV and speech processing libraries to analyze this data in real time and extract the user's emotional information. The output is the analyzed emotional state of the user.

[0379] Step 5:

[0380] The server adjusts the conversational agent's response using emotion information received from the terminal. The inputs are the emotion state obtained in step 4 and the human-like model generated in step 3. Based on these inputs, the server applies the output response to the conversational agent, and the agent generates a response appropriate to the user's emotion. The output is the adjusted response of the conversational agent.

[0381] Step 6:

[0382] The user engages in conversational training with a conversational agent on a terminal. The input is the conversational agent's responses, which were adjusted in step 5. The user responds to questions and information from the agent and practices conversational skills through scenarios provided by the agent. The output is the user's conversational training results, and this data is stored on the server and used for subsequent analysis.

[0383] Step 7:

[0384] The server evaluates the results of the conversation training and provides feedback to the user. The input is the conversation training results obtained in step 6. The server analyzes this and makes an evaluation based on the user's skills and emotions. The output is feedback information for the user, which can be used to improve the next conversation training session. This feedback allows the user to clearly understand areas for self-improvement.

[0385] (Application Example 2)

[0386] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0387] In retail store customer service, appropriately understanding and responding to customer emotions is a challenging task. Traditional customer service methods make it difficult to instantly analyze a customer's facial expressions and tone of voice and respond accordingly, resulting in difficulties in improving customer satisfaction.

[0388] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0389] In this invention, the server includes means for acquiring personal data from an information processing device, means for analyzing the acquired personal data and generating a personal classification, and means for recognizing the user's emotional state in real time using emotion analysis means and adjusting the response of the conversational agent based on the recognized emotional state. This enables the provision of appropriate customer service in accordance with the customer's emotions and improves customer satisfaction.

[0390] An "information processing device" is a device that acquires and analyzes personal data and processes that information for a specific purpose.

[0391] "Personal data" refers to information about a user, which is used to classify and identify the user's characteristics through analysis.

[0392] "Personal classification" refers to categories created by classifying users based on specific characteristics and attributes, using analyzed personal data.

[0393] A "human characteristics model" is a model generated based on individual classification, and it imitates human characteristics and behavioral patterns.

[0394] A "dialogue agent" is a program created based on a human characteristics model to simulate conversations with users.

[0395] "Emotional analysis means" refers to technology that analyzes a user's facial expressions and tone of voice to recognize their emotional state in real time.

[0396] "Means for adjusting responses" refers to a function that modifies the content of the dialogue agent's response based on the recognized emotional state, thereby enabling dialogue that is empathetic to the user's emotions.

[0397] This invention provides an emotion recognition system aimed at improving customer service in retail stores. This system integrates and operates with an information processing device, a dialogue agent, and emotion analysis means.

[0398] First, the information processing device acquires personal data via terminals, smart glasses, and smartphones installed in retail stores. This data includes customer facial expressions and voice tone, and is used to generate personal classifications. Next, the server automatically generates a human characteristics model based on the acquired personal data. The generated model is then mimicked within the dialogue agent's program.

[0399] The conversational agent utilizes real-time sentiment analysis during conversations with customers. This allows it to analyze the customer's emotional state and adjust its response accordingly. For example, if the agent detects that the customer is irritated, it will respond in a calmer tone appropriate to that state. To achieve this functionality, APIs such as the Affectiva API can be used for sentiment analysis.

[0400] Store employees, as users of this system, can grasp customer sentiment information in real time and receive suggestions for customer service techniques based on that information. This enables more sophisticated customer service.

[0401] As a concrete example, consider a scenario where a store clerk is wearing smart glasses and a customer asks for product details. If the emotion analysis system determines that the customer is expressing dissatisfaction or doubt, a suggestion such as "Let's talk in more detail about the features of this product" will be displayed on the smart glasses' screen. At this point, the optimal response will be suggested to the generative AI model through a prompt message such as "Advise me on how to respond if the customer is showing signs of frustration."

[0402] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0403] Step 1:

[0404] The server acquires personal data via terminals in retail stores, smart glasses, and smartphones. Inputs include facial image data and voice recordings of customers, while outputs are these data. The data is collected in real time and used as material for customer sentiment analysis.

[0405] Step 2:

[0406] The server uses a machine learning model with the acquired personal data to perform emotion analysis. The input is the facial expression images and audio data obtained in step 1, and the output is the customer's emotional state (e.g., joy, anger, anxiety). The Affectiva API is used to analyze the data and instantly quantify the customer's emotions.

[0407] Step 3:

[0408] The server generates a personal classification based on the analyzed emotional state. The input is the emotional state obtained in step 2, and the output is the personal classification result. Based on this result, the server determines what response is appropriate.

[0409] Step 4:

[0410] The server generates a human characteristics model based on the personal classification results and reflects it in the conversational agent. The input is the personal classification result obtained in step 3, and the output is the human characteristics model. A prompt sentence is sent to the generated AI model to obtain the data necessary to adjust the response.

[0411] Step 5:

[0412] The terminal interacts with customers through a conversational agent. The input is the human feature model information obtained in step 4, and the output is a suggested response to the customer. The suggested response is displayed in real time on the smart glasses or terminal screen, and the store staff uses this to provide customer service.

[0413] Step 6:

[0414] The store clerk, acting as the user, provides appropriate customer service to customers based on the responses and suggestions presented by the terminal. The input is the response suggestion obtained in step 5, and the output is the actual result of the interaction with the customer. The clerk observes customer feedback and reports it to the server via the terminal as needed.

[0415] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0416] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0417] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0418] [Third Embodiment]

[0419] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0420] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0421] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0422] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0423] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0424] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0425] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0426] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0427] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0428] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0429] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0430] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0431] This invention is a system for improving the quality of online training and instruction. The system begins by acquiring a large amount of user data from an information aggregation device and analyzing it to identify various user classifications. Based on the analysis results, it automatically generates human models and creates conversational agents based on each model.

[0432] The server collects user data and analyzes it using machine learning algorithms. Clustering techniques are used to group users with similar behavioral patterns and attributes, and based on this, a characteristic human model is generated. This model includes name, age, interests, and typical questions and behavioral patterns.

[0433] Next, the server builds a conversational agent based on the generated human model. This agent can interact with the user using natural language processing, asking various questions and processing the user's responses.

[0434] The terminal functions as an entry point to this system, providing an interface for users to undergo dialogue training. Users interact through the terminal, responding to situations and questions provided by the agent. In this process, users' communication skills and problem-solving abilities are tested, and their skills are improved.

[0435] The server has the capability to analyze the results of dialogue training and evaluate the user's responses. It generates feedback based on various criteria, such as accuracy, speed, and creative problem-solving ability. This feedback is presented to the user through the terminal and can be used to improve future performance.

[0436] Specific example

[0437] For example, when implemented as part of a company's training program for new employees, the server can generate diverse personas from past data, such as "first-time project participants" or "users asking technical questions." Using conversational agents created based on these personas, users simulate realistic scenarios on their devices. For instance, the agent might ask a question like, "How would you handle this situation?" and the user responds. The server then analyzes the response and provides feedback, such as, "The answer is well-structured, but you need to respond more quickly." In this way, users can improve their ability to handle real-world situations in the workplace.

[0438] The following describes the processing flow.

[0439] Step 1:

[0440] The server retrieves user data from the data aggregation device. This data includes the user's behavioral history, profile, and interests. The retrieved data is anonymized to protect privacy.

[0441] Step 2:

[0442] The server performs data analysis by applying machine learning algorithms based on the acquired user data. Using clustering techniques, it classifies users with similar patterns and generates typical user classifications.

[0443] Step 3:

[0444] The server automatically generates a human model based on the generated user classification. This model includes attributes, interests, and behavioral patterns corresponding to each classification. The generated model is stored in a database.

[0445] Step 4:

[0446] The server creates a conversational agent based on a stored human model. Utilizing natural language processing technology, this agent can interact with the user in real time.

[0447] Step 5:

[0448] The terminal provides the user with an interface for dialogue training. Through this interface, the user can begin a simulation with a dialogue agent.

[0449] Step 6:

[0450] The user responds to questions and issues raised by the conversational agent. The dialogue is based on realistic scenarios and aims to improve the user's skills.

[0451] Step 7:

[0452] The server records user responses during interaction and generates logs for later detailed analysis. It evaluates the accuracy of responses, response speed, and originality of problem-solving.

[0453] Step 8:

[0454] The server generates feedback based on the analysis results and evaluates the dialogue training. This feedback includes areas for improvement, successes, and suggestions for the next training session.

[0455] Step 9:

[0456] The device presents the generated feedback to the user, highlighting areas for improvement and supporting continuous skill development.

[0457] (Example 1)

[0458] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0459] In online training and development, there is a need to provide effective systems for individualized instruction and evaluation that meet the diverse needs of users. Currently, general training systems do not adequately provide personalized feedback and interactive functions, making efficient skill improvement difficult. Therefore, the challenge is to realize training that is more flexible and tailored to the characteristics of each user.

[0460] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0461] In this invention, the server includes means for acquiring information from a data aggregation device, means for analyzing the acquired information and generating multiple information classifications, means for automatically generating a person model based on each information classification, means for grouping information using a clustering method, and means for constructing a dialogue program using natural language processing technology. This enables personalized dialogue and feedback tailored to the individual characteristics of each user, and allows for efficient skill improvement.

[0462] A "data aggregation device" is a device used to collect information from various sources and manage it centrally.

[0463] "Information classification" refers to the process of dividing collected data into groups based on specific criteria, or the result of such classification.

[0464] A "person model" is a hypothetical person who possesses typical attributes and behavioral patterns, generated based on a specific classification of information.

[0465] A "dialogue program" is software that uses natural language processing technology to communicate with users.

[0466] Clustering is a technique used in data analysis to group data with similar characteristics together.

[0467] "Natural language processing technology" is a technology that enables computers to understand and process human language.

[0468] "Feedback" is information that provides evaluations and suggestions for improvement regarding actions or responses.

[0469] This system is built to improve the quality of online training and instruction. The server begins by collecting large amounts of user data using data aggregation equipment. The collected data includes user behavior history and interaction history, which is then analyzed using machine learning algorithms. Specifically, clustering algorithms (e.g., k-means) are used to group users with similar behavioral patterns and attributes.

[0470] Next, the server generates a characteristic person model based on the analysis results. This model includes typical age groups, interests, and common question patterns, forming the basis for more personalized training. Based on the generated person model, a dialogue program is built using natural language processing techniques. This technique utilizes natural language processing libraries such as NLTK and spaCy.

[0471] The terminal provides an interface for the user to interact with the dialogue program. The user interacts with the agent through the terminal, responding to the presented situations and questions. This process tests the user's communication skills and problem-solving abilities.

[0472] The server also analyzes the results of the dialogue training and evaluates the user's responses. Evaluation criteria include the accuracy and speed of responses, as well as the presentation of creative solutions, and feedback is generated based on these criteria. This feedback is presented to the user through the terminal and contributes to further skill improvement.

[0473] For example, in a training program for new employees, the server generates diverse personas from past data, such as "first-time project participant" or "user who actively asks technical questions." Using these generated personas, a realistic simulation is conducted on the terminal, and an agent asks questions such as, "How would you handle this situation?" After the user responds, the server analyzes the response and provides feedback such as, "The response is well-structured, but speed is required."

[0474] Examples of prompts include, "Create a persona to generate a dialogue program for new employees," and "List the questions the agent will ask the user."

[0475] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0476] Step 1:

[0477] The server collects user data from data aggregation devices. This data includes user login history, training history, and interaction history. Based on this data, the server extracts characteristic information necessary for the training program. The output here is user data containing characteristic information. Specifically, this involves extracting and filtering information from the database.

[0478] Step 2:

[0479] The server analyzes the collected user data. Specifically, it uses a machine learning algorithm to perform clustering on the data. k-means clustering is used for this process, grouping users with similar behavioral patterns or common attributes. The output is the clustering result of the users. The operation includes steps such as executing the algorithm and extracting features for each cluster.

[0480] Step 3:

[0481] The server generates person models based on the clustering results. Typical age groups, interests, and common question patterns are included to create models that reflect the characteristics of each cluster. The input is the clustering results, and the output is the person model for each cluster. Specifically, the profile information is registered in a database.

[0482] Step 4:

[0483] The server builds a dialogue program based on the generated human model. A natural language processing library (e.g., NLTK or spaCy) is used to program the user interaction scenario. The input is the human model, and the output is the dialogue program. Operation includes code implementation and testing.

[0484] Step 5:

[0485] The terminal provides an interface for the user to interact with the agent. The user participates in the interaction and responds to the scenarios and questions presented. The input is the agent's questions, and the output is the user's answers. In terms of specific actions, the user interface is manipulated.

[0486] Step 6:

[0487] The server analyzes user responses to evaluate the results of dialogue training. The evaluation criteria include the accuracy, speed, and creativity of the responses. The input is a log of user responses, and the output is feedback including areas for improvement. The operation involves running an analysis algorithm and saving the feedback results to a database.

[0488] (Application Example 1)

[0489] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0490] In today's business environment, practical problem-solving skills and communication abilities are extremely important. However, traditional training methods lack sufficient training that closely resembles real-world situations, making it difficult for staff to effectively improve their skills. Therefore, there is a need for a system that enables training in environments close to actual work situations and effectively improves skills.

[0491] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0492] In this invention, the server includes means for acquiring user data from an information aggregation device, means for analyzing the acquired user data and generating multiple user classifications, and means for automatically generating a human model based on each user classification. This makes it possible to provide a simulation environment close to the real world via a smart device using a dialogue agent that mimics the generated human model, enabling users to improve their skills to suit real-world situations through dialogue training.

[0493] An "information aggregation device" is a device or platform for collecting user data and providing it to a system.

[0494] "User data analysis" is the process of using collected data to analyze the behavior and characteristics of individual users and extract information from them.

[0495] "User classification" refers to categories used to group users with similar characteristics and behavioral patterns based on analyzed data.

[0496] A "human model" is a virtual person profile automatically generated based on analysis results, and it includes typical attributes and behavioral patterns.

[0497] A "dialogue agent" is a program built on a human model and is software capable of engaging in dialogue with users.

[0498] A "simulation environment via smart devices" is a virtual training ground that reproduces situations close to the real world, provided through devices such as smartphones.

[0499] "Feedback" refers to information provided to users for evaluation and improvement based on the results of dialogue training.

[0500] To implement this invention, it is necessary to construct a system in which a server, a terminal, and a user interact with each other. The server acquires user data from an information aggregation device, analyzes it, and generates user classifications. For the analysis, it utilizes Scikit-learn, a machine learning library using Python, and identifies similar user groups by clustering user behavior patterns. Based on the obtained classification information, a human model is automatically generated. This human model mimics the user's age, interests, and typical questions and behavior patterns.

[0501] The server then creates a conversational agent based on the generated human model. This conversational agent uses NLTK, an open-source NLP library, to implement natural language processing. This agent can respond to user questions and provide various scenarios through dialogue.

[0502] The terminal provides an interface for users to interact with the system. Typically, smart devices (smartphones or tablets) fulfill this role and are operated via a browser or dedicated app. The terminal allows users to receive training in a simulated environment using conversational agents, helping staff improve their skills in situations close to real-world scenarios.

[0503] Users participate in conversational simulations using smart devices. This simulation environment replicates real-world customer service scenarios, presenting prompts such as, "Consider the following situation: A customer from overseas is unsure about their order. How can you assist them?" User responses are sent to the server sequentially and recorded as conversational training data. The server analyzes this data and provides quick and effective feedback to help improve user skills.

[0504] This system allows store staff to effectively improve their customer service skills through simulations and feedback via smart devices.

[0505] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0506] Step 1:

[0507] The server retrieves user data from the information aggregation device. This data includes the user's behavioral history and basic attribute information. The user data retrieved as input is stored within the system.

[0508] Step 2:

[0509] The server analyzes the acquired user data and performs data clustering. It uses the machine learning library Scikit-learn to perform PCA (Principal Component Analysis) and extract important features. This groups users with similar features and generates user classifications as output.

[0510] Step 3:

[0511] The server automatically generates a human model based on the generated user classification. This human model includes typical attributes and behavioral patterns. This process primarily involves generating data structures, and the output provides profile information for the conversational agent.

[0512] Step 4:

[0513] The server uses the NLTK natural language processing library to build a conversational agent. This agent is designed using profile information obtained from a human model. It receives the profile as input and generates a script simulating user interaction as output.

[0514] Step 5:

[0515] The terminal provides a user-accessible interface and connects the user with the conversational agent. It runs on a smart device via an application or browser. It receives agent scripts as input and generates visual and audio user interfaces as output.

[0516] Step 6:

[0517] The user conducts simulation training with a conversational agent via a terminal. A scenario including prompts is displayed, and the user inputs their response. The input is the user's response and is sent to the server as output.

[0518] Step 7:

[0519] The server analyzes user responses and evaluates the training. It performs Japanese text analysis to measure the accuracy and speed of responses. As a result, feedback information is generated and returned to the terminal as output.

[0520] Step 8:

[0521] The terminal presents feedback from the server to the user. It displays feedback that includes areas for improvement and effective approaches, allowing the user to utilize this information in future training sessions. It receives feedback information as input and displays it on the screen as output.

[0522] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0523] This invention is an online training system that integrates emotion recognition functionality, taking into account the user's emotions to provide more effective dialogue training. The system begins by acquiring vast amounts of user data from an information aggregation device and analyzing it to generate multiple user classifications. Based on the analysis results, it automatically generates a human model and creates a dialogue agent that mimics it.

[0524] The server processes the acquired user data using a machine learning algorithm and classifies users into multiple categories. Based on these results, it generates a human model corresponding to each category. Each model includes name, age, interests, behavioral patterns, and typical questions asked.

[0525] The emotion engine is an additional element that recognizes the user's facial expressions, tone of voice, and the emotions expressed in the input text in real time via the device. The recognized emotion information is processed on the server, and the results are reflected in the conversational agent's response content. This allows the agent to provide more human-like responses, enabling calm conversations and emotionally empathetic communication.

[0526] The terminal provides an interface for dialogue training, creating an environment where users can simulate with a dialogue agent. Users use this interface to address a variety of questions and situations based on realistic scenarios.

[0527] The server also analyzes the results of the dialogue training and generates emotional responses and feedback based on the user's emotional state. For example, if the user is feeling stressed, feedback such as recommending a more relaxed approach will be provided based on that information.

[0528] Specific example

[0529] Let's consider a scenario involving new employee training at a company. The server analyzes past data and generates diverse human models, such as "users struggling with technical questions" and "users participating in a project for the first time." Based on these models, a conversational agent is created to ask the user questions like, "What do you think about this new system?" As the user answers, the agent adjusts its response based on the emotions recognized by the emotion engine. For example, if the user shows anxiety, the agent will provide an emotionally sensitive response such as, "Don't worry, this information will be explained in more detail later." The server analyzes this conversation and provides feedback, such as, "The quick response is good, but a calmer tone is needed." In this way, users can improve their emotional awareness along with their actual conversational skills.

[0530] The following describes the processing flow.

[0531] Step 1:

[0532] The server retrieves vast amounts of user data from the data aggregation device. This data includes user behavior history, profiles, and text data. After data retrieval, it is anonymized to protect privacy.

[0533] Step 2:

[0534] The server analyzes the acquired user data and uses machine learning algorithms to divide users into multiple categories. Through clustering, it forms user groups with similar characteristics and generates a human model based on them.

[0535] Step 3:

[0536] The server creates a conversational agent based on the generated human model. Equipped with natural language processing technology, this agent enables real-time interaction with the user and constructs questions and responses tailored to specific scenarios.

[0537] Step 4:

[0538] The terminal provides the user with an interface for dialogue training. The user uses this interface to initiate dialogue simulations and experience interaction with a human-model-based agent.

[0539] Step 5:

[0540] The emotion engine built into the device analyzes the user's facial expressions, voice tone, and input text to recognize the user's emotions. This information is transmitted to the server in real time.

[0541] Step 6:

[0542] The server analyzes the received emotion recognition data and incorporates the results into the dialogue agent's response. This allows the agent to reconstruct questions to match the user's emotions and provide appropriate, empathetic responses.

[0543] Step 7:

[0544] Users interact with conversational agents and experience emotion-based feedback, which helps them enhance their realistic communication skills.

[0545] Step 8:

[0546] The server analyzes and evaluates the overall results of the dialogue training. This process includes an assessment based on the appropriateness, speed, and perceived emotions of the user's responses.

[0547] Step 9:

[0548] The server generates customized feedback based on the analysis results and provides it to the user through the terminal. The feedback includes areas for improvement in emotional response and key success factors, encouraging the user to further improve their skills.

[0549] (Example 2)

[0550] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0551] In recent years, interactive training systems have become widespread, but many of them fail to provide effective dialogue that takes user emotions into account. Conventional systems lack sufficient technology to adjust responses using real-time user emotion information, which can result in a decline in the quality of the dialogue. Therefore, there is a need for dialogue systems that allow users to gain a greater sense of satisfaction and learning effectiveness.

[0552] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0553] In this invention, the server includes means for acquiring user information from an information processing device, means for analyzing the acquired user information and generating multiple user classifications, and means for automatically generating a human-like model based on each user classification. This enables dialogue by a conversational agent that is adjusted according to the user's emotional state, providing a sophisticated communication experience that is sensitive to emotions.

[0554] An "information processing device" is a device for collecting and managing user data, and has the function of deriving new information by analyzing the obtained data.

[0555] "User information" is a general term for information related to a user that the system acquires, and includes data such as an individual's profile and past usage history.

[0556] "User classification" refers to the process of classifying and grouping users according to specific criteria based on acquired user information.

[0557] A "human-style model" is a model generated based on user classification, and serves as a foundation for mimicking user characteristics and behavioral patterns.

[0558] A "conversational agent" is a program or software designed to mimic human behavior models and engage in dialogue with users.

[0559] An "emotion recognition device" is a device that analyzes a user's emotional state in real time and provides appropriate information accordingly.

[0560] "Information retention means" refers to means that provide a function to save the characteristics and profiles of a human-like model and retrieve them later.

[0561] "Situation adjustment means" are methods used to dynamically adjust the questions and situations presented by the conversational agent during conversation training, and are used to make the dialogue more effective.

[0562] This invention describes a specific embodiment of an online training system that takes user emotions into consideration. The system consists of a server, a terminal, and a user.

[0563] The server functions as an information processing device, collecting user information. Database management software and cloud storage are used for this purpose. The server analyzes the collected user information using machine learning frameworks such as TensorFlow and PyTorch, classifying users into multiple user categories. Based on the analyzed data, a human-like model is automatically generated, which is then used for further conversation simulations.

[0564] The terminal provides an interface with the user and recognizes the user's facial expressions, tone of voice, and text emotions in real time via an emotion recognition device. This uses a webcam and microphone, and OpenCV and speech processing libraries are used to analyze the emotion data. This real-time data is sent to a server, enabling a conversational agent to generate tailored responses based on the emotion data.

[0565] Users engage in conversational training with a conversational agent through this device. The conversational agent is built based on various human-like models and provides emotionally resonant responses to the user. For example, if the user feels anxious, the agent can respond with something like, "It's okay, I'll explain the details step by step later."

[0566] A concrete example is its use when new employees practice for job interviews. The server analyzes past interactions and generates models such as "stressed applicants." Based on this, a conversational agent asks typical questions such as "Why did you apply for this job?" and provides feedback tailored to the user's emotional perception, thereby enhancing the user's practice effectiveness.

[0567] An example of a prompt is, "Please tell me how to communicate in a way that takes the user's emotional state into consideration and is sensitive to their feelings." By using this prompt, the generative AI model can be given instructions to perform a more specific simulation.

[0568] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0569] Step 1:

[0570] The server collects user information through an information processing device. Inputs include user profile data and past conversation logs, which are retrieved using database management software. Outputs are structured datasets, which are used in subsequent analysis steps.

[0571] Step 2:

[0572] The server analyzes the collected user information using a machine learning framework. The input is the dataset obtained in Step 1. Based on this, the server analyzes the data using TensorFlow or PyTorch and divides each user into multiple user classifications based on their features. The output is the category information into which the users have been classified.

[0573] Step 3:

[0574] The server automatically generates human-like models based on user classification. The input is the category information created in step 2, and the process generates model parameters representative of each category. The server uses a model template and sets parameters such as name, age, and interests to create the human-like models. The output is a set of human-like models for each user classification.

[0575] Step 4:

[0576] The device performs emotion recognition through its interface with the user. Inputs include the user's facial expressions, voice tone, and text input, which are captured via a webcam and microphone. The device uses OpenCV and speech processing libraries to analyze this data in real time and extract the user's emotional information. The output is the analyzed emotional state of the user.

[0577] Step 5:

[0578] The server adjusts the conversational agent's response using emotion information received from the terminal. The inputs are the emotion state obtained in step 4 and the human-like model generated in step 3. Based on these inputs, the server applies the output response to the conversational agent, and the agent generates a response appropriate to the user's emotion. The output is the adjusted response of the conversational agent.

[0579] Step 6:

[0580] The user engages in conversational training with a conversational agent on a terminal. The input is the conversational agent's responses, which were adjusted in step 5. The user responds to questions and information from the agent and practices conversational skills through scenarios provided by the agent. The output is the user's conversational training results, and this data is stored on the server and used for subsequent analysis.

[0581] Step 7:

[0582] The server evaluates the results of the conversation training and provides feedback to the user. The input is the conversation training results obtained in step 6. The server analyzes this and makes an evaluation based on the user's skills and emotions. The output is feedback information for the user, which can be used to improve the next conversation training session. This feedback allows the user to clearly understand areas for self-improvement.

[0583] (Application Example 2)

[0584] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0585] In retail store customer service, appropriately understanding and responding to customer emotions is a challenging task. Traditional customer service methods make it difficult to instantly analyze a customer's facial expressions and tone of voice and respond accordingly, resulting in difficulties in improving customer satisfaction.

[0586] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0587] In this invention, the server includes means for acquiring personal data from an information processing device, means for analyzing the acquired personal data and generating a personal classification, and means for recognizing the user's emotional state in real time using emotion analysis means and adjusting the response of the conversational agent based on the recognized emotional state. This enables the provision of appropriate customer service in accordance with the customer's emotions and improves customer satisfaction.

[0588] An "information processing device" is a device that acquires and analyzes personal data and processes that information for a specific purpose.

[0589] "Personal data" refers to information about a user, which is used to classify and identify the user's characteristics through analysis.

[0590] "Personal classification" refers to categories created by classifying users based on specific characteristics and attributes, using analyzed personal data.

[0591] A "human characteristics model" is a model generated based on individual classification, and it imitates human characteristics and behavioral patterns.

[0592] A "dialogue agent" is a program created based on a human characteristics model to simulate conversations with users.

[0593] "Emotional analysis means" refers to technology that analyzes a user's facial expressions and tone of voice to recognize their emotional state in real time.

[0594] "Means for adjusting responses" refers to a function that modifies the content of the dialogue agent's response based on the recognized emotional state, thereby enabling dialogue that is attentive to the user's emotions.

[0595] This invention provides an emotion recognition system aimed at improving customer service in retail stores. This system integrates and operates with an information processing device, a dialogue agent, and emotion analysis means.

[0596] First, the information processing device acquires personal data via terminals, smart glasses, and smartphones installed in retail stores. This data includes customer facial expressions and voice tone, and is used to generate personal classifications. Next, the server automatically generates a human characteristics model based on the acquired personal data. The generated model is then mimicked within the dialogue agent's program.

[0597] The conversational agent utilizes real-time sentiment analysis during conversations with customers. This allows it to analyze the customer's emotional state and adjust its response accordingly. For example, if the agent detects that the customer is irritated, it will respond in a calmer tone appropriate to that state. To achieve this functionality, APIs such as the Affectiva API can be used for sentiment analysis.

[0598] Store employees, as users of this system, can grasp customer sentiment information in real time and receive suggestions for customer service techniques based on that information. This enables more sophisticated customer service.

[0599] As a concrete example, consider a scenario where a store clerk is wearing smart glasses and a customer asks for product details. If the emotion analysis system determines that the customer is expressing dissatisfaction or doubt, a suggestion such as "Let's talk in more detail about the features of this product" will be displayed on the smart glasses' screen. At this point, the optimal response will be suggested to the generative AI model through a prompt message such as "Advise me on how to respond if the customer is showing signs of frustration."

[0600] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0601] Step 1:

[0602] The server acquires personal data via terminals in retail stores, smart glasses, and smartphones. Inputs include facial image data and voice recordings of customers, while outputs are these data. The data is collected in real time and used as material for customer sentiment analysis.

[0603] Step 2:

[0604] The server uses a machine learning model with the acquired personal data to perform emotion analysis. The input is the facial expression images and audio data obtained in step 1, and the output is the customer's emotional state (e.g., joy, anger, anxiety). The Affectiva API is used to analyze the data and instantly quantify the customer's emotions.

[0605] Step 3:

[0606] The server generates a personal classification based on the analyzed emotional state. The input is the emotional state obtained in step 2, and the output is the personal classification result. Based on this result, the server determines what response is appropriate.

[0607] Step 4:

[0608] The server generates a human characteristics model based on the personal classification results and reflects it in the conversational agent. The input is the personal classification result obtained in step 3, and the output is the human characteristics model. A prompt sentence is sent to the generated AI model to obtain the data necessary to adjust the response.

[0609] Step 5:

[0610] The terminal interacts with customers through a conversational agent. The input is the human feature model information obtained in step 4, and the output is a suggested response to the customer. The suggested response is displayed in real time on the smart glasses or terminal screen, and the store staff uses this to provide customer service.

[0611] Step 6:

[0612] The store clerk, acting as the user, provides appropriate customer service to customers based on the responses and suggestions presented by the terminal. The input is the response suggestion obtained in step 5, and the output is the actual result of the interaction with the customer. The clerk observes customer feedback and reports it to the server via the terminal as needed.

[0613] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0614] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0615] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0616] [Fourth Embodiment]

[0617] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0618] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0619] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0620] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0621] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0622] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0623] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0624] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0625] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0626] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0627] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0628] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0629] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0630] This invention is a system for improving the quality of online training and instruction. The system begins by acquiring a large amount of user data from an information aggregation device and analyzing it to identify various user classifications. Based on the analysis results, it automatically generates human models and creates conversational agents based on each model.

[0631] The server collects user data and analyzes it using machine learning algorithms. Clustering techniques are used to group users with similar behavioral patterns and attributes, and based on this, a characteristic human model is generated. This model includes name, age, interests, and typical questions and behavioral patterns.

[0632] Next, the server builds a conversational agent based on the generated human model. This agent can interact with the user using natural language processing, asking various questions and processing the user's responses.

[0633] The terminal functions as an entry point to this system, providing an interface for users to undergo dialogue training. Users interact through the terminal, responding to situations and questions provided by the agent. In this process, users' communication skills and problem-solving abilities are tested, and their skills are improved.

[0634] The server has the capability to analyze the results of dialogue training and evaluate the user's responses. It generates feedback based on various criteria, such as accuracy, speed, and creative problem-solving ability. This feedback is presented to the user through the terminal and can be used to improve future performance.

[0635] Specific example

[0636] For example, when implemented as part of a company's training program for new employees, the server can generate diverse personas from past data, such as "first-time project participants" or "users asking technical questions." Using conversational agents created based on these personas, users simulate realistic scenarios on their devices. For instance, the agent might ask a question like, "How would you handle this situation?" and the user responds. The server then analyzes the response and provides feedback, such as, "The answer is well-structured, but you need to respond more quickly." In this way, users can improve their ability to handle real-world situations in the workplace.

[0637] The following describes the processing flow.

[0638] Step 1:

[0639] The server retrieves user data from the data aggregation device. This data includes the user's behavioral history, profile, and interests. The retrieved data is anonymized to protect privacy.

[0640] Step 2:

[0641] The server performs data analysis by applying machine learning algorithms based on the acquired user data. Using clustering techniques, it classifies users with similar patterns and generates typical user classifications.

[0642] Step 3:

[0643] The server automatically generates a human model based on the generated user classification. This model includes attributes, interests, and behavioral patterns corresponding to each classification. The generated model is stored in a database.

[0644] Step 4:

[0645] The server creates a conversational agent based on a stored human model. Utilizing natural language processing technology, this agent can interact with the user in real time.

[0646] Step 5:

[0647] The terminal provides the user with an interface for dialogue training. Through this interface, the user can begin a simulation with a dialogue agent.

[0648] Step 6:

[0649] The user responds to questions and issues raised by the conversational agent. The dialogue is based on realistic scenarios and aims to improve the user's skills.

[0650] Step 7:

[0651] The server records user responses during interaction and generates logs for later detailed analysis. It evaluates the accuracy of responses, response speed, and originality of problem-solving.

[0652] Step 8:

[0653] The server generates feedback based on the analysis results and evaluates the dialogue training. This feedback includes areas for improvement, successes, and suggestions for the next training session.

[0654] Step 9:

[0655] The device presents the generated feedback to the user, highlighting areas for improvement and supporting continuous skill development.

[0656] (Example 1)

[0657] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0658] In online training and development, there is a need to provide effective systems for individualized instruction and evaluation that meet the diverse needs of users. Currently, general training systems do not adequately provide personalized feedback and interactive functions, making efficient skill improvement difficult. Therefore, the challenge is to realize training that is more flexible and tailored to the characteristics of each user.

[0659] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0660] In this invention, the server includes means for acquiring information from a data aggregation device, means for analyzing the acquired information and generating multiple information classifications, means for automatically generating a person model based on each information classification, means for grouping information using a clustering method, and means for constructing a dialogue program using natural language processing technology. This enables personalized dialogue and feedback tailored to the individual characteristics of each user, and allows for efficient skill improvement.

[0661] A "data aggregation device" is a device used to collect information from various sources and manage it centrally.

[0662] "Information classification" refers to the process of dividing collected data into groups based on specific criteria, or the result of such classification.

[0663] A "person model" is a hypothetical person who possesses typical attributes and behavioral patterns, generated based on a specific classification of information.

[0664] A "dialogue program" is software that uses natural language processing technology to communicate with users.

[0665] Clustering is a technique used in data analysis to group data with similar characteristics together.

[0666] "Natural language processing technology" is a technology that enables computers to understand and process human language.

[0667] "Feedback" is information that provides evaluations and suggestions for improvement regarding actions or responses.

[0668] This system is built to improve the quality of online training and instruction. The server begins by collecting large amounts of user data using data aggregation equipment. The collected data includes user behavior history and interaction history, which is then analyzed using machine learning algorithms. Specifically, clustering algorithms (e.g., k-means) are used to group users with similar behavioral patterns and attributes.

[0669] Next, the server generates a characteristic person model based on the analysis results. This model includes typical age groups, interests, and common question patterns, forming the basis for more personalized training. Based on the generated person model, a dialogue program is built using natural language processing techniques. This technique utilizes natural language processing libraries such as NLTK and spaCy.

[0670] The terminal provides an interface for the user to interact with the dialogue program. The user interacts with the agent through the terminal, responding to the presented situations and questions. This process tests the user's communication skills and problem-solving abilities.

[0671] The server also analyzes the results of the dialogue training and evaluates the user's responses. Evaluation criteria include the accuracy and speed of responses, as well as the presentation of creative solutions, and feedback is generated based on these criteria. This feedback is presented to the user through the terminal and contributes to further skill improvement.

[0672] For example, in a training program for new employees, the server generates diverse personas from past data, such as "first-time project participant" or "user who actively asks technical questions." Using these generated personas, a realistic simulation is conducted on the terminal, and an agent asks questions such as, "How would you handle this situation?" After the user responds, the server analyzes the response and provides feedback such as, "The response is well-structured, but speed is required."

[0673] Examples of prompts include, "Create a persona to generate a dialogue program for new employees," and "List the questions the agent will ask the user."

[0674] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0675] Step 1:

[0676] The server collects user data from data aggregation devices. This data includes user login history, training history, and interaction history. Based on this data, the server extracts characteristic information necessary for the training program. The output here is user data containing characteristic information. Specifically, this involves extracting and filtering information from the database.

[0677] Step 2:

[0678] The server analyzes the collected user data. Specifically, it uses a machine learning algorithm to perform clustering on the data. k-means clustering is used for this process, grouping users with similar behavioral patterns or common attributes. The output is the clustering result of the users. The operation includes steps such as executing the algorithm and extracting features for each cluster.

[0679] Step 3:

[0680] The server generates person models based on the clustering results. Typical age groups, interests, and common question patterns are included to create models that reflect the characteristics of each cluster. The input is the clustering results, and the output is the person model for each cluster. Specifically, the profile information is registered in a database.

[0681] Step 4:

[0682] The server builds a dialogue program based on the generated human model. A natural language processing library (e.g., NLTK or spaCy) is used to program the user interaction scenario. The input is the human model, and the output is the dialogue program. Operation includes code implementation and testing.

[0683] Step 5:

[0684] The terminal provides an interface for the user to interact with the agent. The user participates in the interaction and responds to the scenarios and questions presented. The input is the agent's questions, and the output is the user's answers. In terms of specific actions, the user interface is manipulated.

[0685] Step 6:

[0686] The server analyzes user responses to evaluate the results of dialogue training. The evaluation criteria include the accuracy, speed, and creativity of the responses. The input is a log of user responses, and the output is feedback including areas for improvement. The operation involves running an analysis algorithm and saving the feedback results to a database.

[0687] (Application Example 1)

[0688] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0689] In today's business environment, practical problem-solving skills and communication abilities are extremely important. However, traditional training methods lack sufficient training that closely resembles real-world situations, making it difficult for staff to effectively improve their skills. Therefore, there is a need for a system that enables training in environments close to actual work situations and effectively improves skills.

[0690] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0691] In this invention, the server includes means for acquiring user data from an information aggregation device, means for analyzing the acquired user data and generating multiple user classifications, and means for automatically generating a human model based on each user classification. This makes it possible to provide a simulation environment close to the real world via a smart device using a dialogue agent that mimics the generated human model, enabling users to improve their skills to suit real-world situations through dialogue training.

[0692] An "information aggregation device" is a device or platform for collecting user data and providing it to a system.

[0693] "User data analysis" is the process of using collected data to analyze the behavior and characteristics of individual users and extract information from them.

[0694] "User classification" refers to categories used to group users with similar characteristics and behavioral patterns based on analyzed data.

[0695] A "human model" is a virtual person profile automatically generated based on analysis results, and it includes typical attributes and behavioral patterns.

[0696] A "dialogue agent" is a program built on a human model and is software capable of engaging in dialogue with users.

[0697] A "simulation environment via smart devices" is a virtual training ground that reproduces situations close to the real world, provided through devices such as smartphones.

[0698] "Feedback" refers to information provided to users for evaluation and improvement based on the results of dialogue training.

[0699] To implement this invention, it is necessary to construct a system in which a server, a terminal, and a user interact with each other. The server acquires user data from an information aggregation device, analyzes it, and generates user classifications. For the analysis, it utilizes Scikit-learn, a machine learning library using Python, and identifies similar user groups by clustering user behavior patterns. Based on the obtained classification information, a human model is automatically generated. This human model mimics the user's age, interests, and typical questions and behavior patterns.

[0700] The server then creates a conversational agent based on the generated human model. This conversational agent uses NLTK, an open-source NLP library, to implement natural language processing. This agent can respond to user questions and provide various scenarios through dialogue.

[0701] The terminal provides an interface for users to interact with the system. Typically, smart devices (smartphones or tablets) fulfill this role and are operated via a browser or dedicated app. The terminal allows users to receive training in a simulated environment using conversational agents, helping staff improve their skills in situations close to real-world scenarios.

[0702] Users participate in conversational simulations using smart devices. This simulation environment replicates real-world customer service scenarios, presenting prompts such as, "Consider the following situation: A customer from overseas is unsure about their order. How can you assist them?" User responses are sent to the server sequentially and recorded as conversational training data. The server analyzes this data and provides quick and effective feedback to help improve user skills.

[0703] This system allows store staff to effectively improve their customer service skills through simulations and feedback via smart devices.

[0704] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0705] Step 1:

[0706] The server retrieves user data from the information aggregation device. This data includes the user's behavioral history and basic attribute information. The user data retrieved as input is stored within the system.

[0707] Step 2:

[0708] The server analyzes the acquired user data and performs data clustering. It uses the machine learning library Scikit-learn to perform PCA (Principal Component Analysis) and extract important features. This groups users with similar features and generates user classifications as output.

[0709] Step 3:

[0710] The server automatically generates a human model based on the generated user classification. This human model includes typical attributes and behavioral patterns. This process primarily involves generating data structures, and the output provides profile information for the conversational agent.

[0711] Step 4:

[0712] The server uses the NLTK natural language processing library to build a conversational agent. This agent is designed using profile information obtained from a human model. It receives the profile as input and generates a script simulating user interaction as output.

[0713] Step 5:

[0714] The terminal provides a user-accessible interface and connects the user with the conversational agent. It runs on a smart device via an application or browser. It receives agent scripts as input and generates visual and audio user interfaces as output.

[0715] Step 6:

[0716] The user conducts simulation training with a conversational agent via a terminal. A scenario including prompts is displayed, and the user inputs their response. The input is the user's response and is sent to the server as output.

[0717] Step 7:

[0718] The server analyzes user responses and evaluates the training. It performs Japanese text analysis to measure the accuracy and speed of responses. As a result, feedback information is generated and returned to the terminal as output.

[0719] Step 8:

[0720] The terminal presents feedback from the server to the user. It displays feedback that includes areas for improvement and effective approaches, allowing the user to utilize this information in future training sessions. It receives feedback information as input and displays it on the screen as output.

[0721] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0722] This invention is an online training system that integrates emotion recognition functionality, taking into account the user's emotions to provide more effective dialogue training. The system begins by acquiring vast amounts of user data from an information aggregation device and analyzing it to generate multiple user classifications. Based on the analysis results, it automatically generates a human model and creates a dialogue agent that mimics it.

[0723] The server processes the acquired user data using a machine learning algorithm and classifies users into multiple categories. Based on these results, it generates a human model corresponding to each category. Each model includes name, age, interests, behavioral patterns, and typical questions asked.

[0724] The emotion engine is an additional element that recognizes the user's facial expressions, tone of voice, and the emotions expressed in the input text in real time via the device. The recognized emotion information is processed on the server, and the results are reflected in the conversational agent's response content. This allows the agent to provide more human-like responses, enabling calm conversations and emotionally empathetic communication.

[0725] The terminal provides an interface for dialogue training, creating an environment where users can simulate with a dialogue agent. Users use this interface to address a variety of questions and situations based on realistic scenarios.

[0726] The server also analyzes the results of the dialogue training and generates emotional responses and feedback based on the user's emotional state. For example, if the user is feeling stressed, feedback such as recommending a more relaxed approach will be provided based on that information.

[0727] Specific example

[0728] Let's consider a scenario involving new employee training at a company. The server analyzes past data and generates diverse human models, such as "users struggling with technical questions" and "users participating in a project for the first time." Based on these models, a conversational agent is created to ask the user questions like, "What do you think about this new system?" As the user answers, the agent adjusts its response based on the emotions recognized by the emotion engine. For example, if the user shows anxiety, the agent will provide an emotionally sensitive response such as, "Don't worry, this information will be explained in more detail later." The server analyzes this conversation and provides feedback, such as, "The quick response is good, but a calmer tone is needed." In this way, users can improve their emotional awareness along with their actual conversational skills.

[0729] The following describes the processing flow.

[0730] Step 1:

[0731] The server retrieves vast amounts of user data from the data aggregation device. This data includes user behavior history, profiles, and text data. After data retrieval, it is anonymized to protect privacy.

[0732] Step 2:

[0733] The server analyzes the acquired user data and uses machine learning algorithms to divide users into multiple categories. Through clustering, it forms user groups with similar characteristics and generates a human model based on them.

[0734] Step 3:

[0735] The server creates a conversational agent based on the generated human model. Equipped with natural language processing technology, this agent enables real-time interaction with the user and constructs questions and responses tailored to specific scenarios.

[0736] Step 4:

[0737] The terminal provides the user with an interface for dialogue training. The user uses this interface to initiate dialogue simulations and experience interaction with a human-model-based agent.

[0738] Step 5:

[0739] The emotion engine built into the device analyzes the user's facial expressions, voice tone, and input text to recognize the user's emotions. This information is transmitted to the server in real time.

[0740] Step 6:

[0741] The server analyzes the received emotion recognition data and incorporates the results into the conversational agent's response. This allows the agent to reconstruct questions to match the user's emotions and provide appropriate, empathetic responses.

[0742] Step 7:

[0743] Users interact with conversational agents and experience emotion-based feedback, which helps them enhance their realistic communication skills.

[0744] Step 8:

[0745] The server analyzes and evaluates the overall results of the dialogue training. This process includes an assessment based on the appropriateness, speed, and perceived emotions of the user's responses.

[0746] Step 9:

[0747] The server generates customized feedback based on the analysis results and provides it to the user through the terminal. The feedback includes areas for improvement in emotional response and key success factors, encouraging the user to further improve their skills.

[0748] (Example 2)

[0749] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0750] In recent years, interactive training systems have become widespread, but many of them fail to provide effective dialogue that takes user emotions into account. Conventional systems lack sufficient technology to adjust responses using real-time user emotion information, which can result in a decline in the quality of the dialogue. Therefore, there is a need for dialogue systems that allow users to gain a greater sense of satisfaction and learning effectiveness.

[0751] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0752] In this invention, the server includes means for acquiring user information from an information processing device, means for analyzing the acquired user information and generating multiple user classifications, and means for automatically generating a human-like model based on each user classification. This enables dialogue by a conversational agent that is adjusted according to the user's emotional state, providing a sophisticated communication experience that is sensitive to emotions.

[0753] An "information processing device" is a device for collecting and managing user data, and has the function of deriving new information by analyzing the obtained data.

[0754] "User information" is a general term for information related to a user that the system acquires, and includes data such as an individual's profile and past usage history.

[0755] "User classification" refers to the process of classifying and grouping users according to specific criteria based on acquired user information.

[0756] A "human-style model" is a model generated based on user classification, and serves as a foundation for mimicking user characteristics and behavioral patterns.

[0757] A "conversational agent" is a program or software designed to mimic human behavior models and engage in dialogue with users.

[0758] An "emotion recognition device" is a device that analyzes a user's emotional state in real time and provides appropriate information accordingly.

[0759] "Information retention means" refers to means that provide a function to save the characteristics and profiles of a human-like model and retrieve them later.

[0760] "Situation adjustment means" are methods used to dynamically adjust the questions and situations presented by the conversational agent during conversation training, and are used to make the dialogue more effective.

[0761] This invention describes a specific embodiment of an online training system that takes user emotions into consideration. The system consists of a server, a terminal, and a user.

[0762] The server functions as an information processing device, collecting user information. Database management software and cloud storage are used for this purpose. The server analyzes the collected user information using machine learning frameworks such as TensorFlow and PyTorch, classifying users into multiple user categories. Based on the analyzed data, a human-like model is automatically generated, which is then used for further conversation simulations.

[0763] The terminal provides an interface with the user and recognizes the user's facial expressions, tone of voice, and text emotions in real time via an emotion recognition device. This uses a webcam and microphone, and OpenCV and speech processing libraries are used to analyze the emotion data. This real-time data is sent to a server, enabling a conversational agent to generate tailored responses based on the emotion data.

[0764] Users engage in conversational training with a conversational agent through this device. The conversational agent is built based on various human-like models and provides emotionally resonant responses to the user. For example, if the user feels anxious, the agent can respond with something like, "It's okay, I'll explain the details step by step later."

[0765] A concrete example is its use when new employees practice for job interviews. The server analyzes past interactions and generates models such as "stressed applicants." Based on this, a conversational agent asks typical questions such as "Why did you apply for this job?" and provides feedback tailored to the user's emotional perception, thereby enhancing the user's practice effectiveness.

[0766] An example of a prompt is, "Please tell me how to communicate in a way that takes the user's emotional state into consideration and is sensitive to their feelings." By using this prompt, the generative AI model can be given instructions to perform a more specific simulation.

[0767] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0768] Step 1:

[0769] The server collects user information through an information processing device. Inputs include user profile data and past conversation logs, which are retrieved using database management software. Outputs are structured datasets, which are used in subsequent analysis steps.

[0770] Step 2:

[0771] The server analyzes the collected user information using a machine learning framework. The input is the dataset obtained in Step 1. Based on this, the server analyzes the data using TensorFlow or PyTorch and divides each user into multiple user classifications based on their features. The output is the category information into which the users have been classified.

[0772] Step 3:

[0773] The server automatically generates human-like models based on user classification. The input is the category information created in step 2, and the process generates model parameters representative of each category. The server uses a model template and sets parameters such as name, age, and interests to create the human-like models. The output is a set of human-like models for each user classification.

[0774] Step 4:

[0775] The device performs emotion recognition through its interface with the user. Inputs include the user's facial expressions, voice tone, and text input, which are captured via a webcam and microphone. The device uses OpenCV and speech processing libraries to analyze this data in real time and extract the user's emotional information. The output is the analyzed emotional state of the user.

[0776] Step 5:

[0777] The server adjusts the conversational agent's response using emotion information received from the terminal. The inputs are the emotion state obtained in step 4 and the human-like model generated in step 3. Based on these inputs, the server applies the output response to the conversational agent, and the agent generates a response appropriate to the user's emotion. The output is the adjusted response of the conversational agent.

[0778] Step 6:

[0779] The user engages in conversational training with a conversational agent on a terminal. The input is the conversational agent's responses, which were adjusted in step 5. The user responds to questions and information from the agent and practices conversational skills through scenarios provided by the agent. The output is the user's conversational training results, and this data is stored on the server and used for subsequent analysis.

[0780] Step 7:

[0781] The server evaluates the results of the conversation training and provides feedback to the user. The input is the conversation training results obtained in step 6. The server analyzes this and makes an evaluation based on the user's skills and emotions. The output is feedback information for the user, which can be used to improve the next conversation training session. This feedback allows the user to clearly understand areas for self-improvement.

[0782] (Application Example 2)

[0783] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0784] In retail store customer service, appropriately understanding and responding to customer emotions is a challenging task. Traditional customer service methods make it difficult to instantly analyze a customer's facial expressions and tone of voice and respond accordingly, resulting in difficulties in improving customer satisfaction.

[0785] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0786] In this invention, the server includes means for acquiring personal data from an information processing device, means for analyzing the acquired personal data and generating a personal classification, and means for recognizing the user's emotional state in real time using emotion analysis means and adjusting the response of the conversational agent based on the recognized emotional state. This enables the provision of appropriate customer service in accordance with the customer's emotions and improves customer satisfaction.

[0787] An "information processing device" is a device that acquires and analyzes personal data and processes that information for a specific purpose.

[0788] "Personal data" refers to information about a user, which is used to classify and identify the user's characteristics through analysis.

[0789] "Personal classification" refers to categories created by classifying users based on specific characteristics and attributes, using analyzed personal data.

[0790] A "human characteristics model" is a model generated based on individual classification, and it imitates human characteristics and behavioral patterns.

[0791] A "dialogue agent" is a program created based on a human characteristics model to simulate conversations with users.

[0792] "Emotional analysis means" refers to technology that analyzes a user's facial expressions and tone of voice to recognize their emotional state in real time.

[0793] "Means for adjusting responses" refers to a function that modifies the content of the dialogue agent's response based on the recognized emotional state, thereby enabling dialogue that is attentive to the user's emotions.

[0794] This invention provides an emotion recognition system aimed at improving customer service in retail stores. This system integrates and operates with an information processing device, a dialogue agent, and emotion analysis means.

[0795] First, the information processing device acquires personal data via terminals, smart glasses, and smartphones installed in retail stores. This data includes customer facial expressions and voice tone, and is used to generate personal classifications. Next, the server automatically generates a human characteristics model based on the acquired personal data. The generated model is then mimicked within the dialogue agent's program.

[0796] The conversational agent utilizes real-time sentiment analysis during conversations with customers. This allows it to analyze the customer's emotional state and adjust its response accordingly. For example, if the agent detects that the customer is irritated, it will respond in a calmer tone appropriate to that state. To achieve this functionality, APIs such as the Affectiva API can be used for sentiment analysis.

[0797] Store employees, as users of this system, can grasp customer sentiment information in real time and receive suggestions for customer service techniques based on that information. This enables more sophisticated customer service.

[0798] As a concrete example, consider a scenario where a store clerk is wearing smart glasses and a customer asks for product details. If the emotion analysis system determines that the customer is expressing dissatisfaction or doubt, a suggestion such as "Let's talk in more detail about the features of this product" will be displayed on the smart glasses' screen. At this point, the optimal response will be suggested to the generative AI model through a prompt message such as "Advise me on how to respond if the customer is showing signs of frustration."

[0799] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0800] Step 1:

[0801] The server acquires personal data via terminals in retail stores, smart glasses, and smartphones. Inputs include facial image data and voice recordings of customers, while outputs are these data. The data is collected in real time and used as material for customer sentiment analysis.

[0802] Step 2:

[0803] The server uses a machine learning model with the acquired personal data to perform emotion analysis. The input is the facial expression images and audio data obtained in step 1, and the output is the customer's emotional state (e.g., joy, anger, anxiety). The Affectiva API is used to analyze the data and instantly quantify the customer's emotions.

[0804] Step 3:

[0805] The server generates a personal classification based on the analyzed emotional state. The input is the emotional state obtained in step 2, and the output is the personal classification result. Based on this result, the server determines what response is appropriate.

[0806] Step 4:

[0807] The server generates a human characteristics model based on the personal classification results and reflects it in the conversational agent. The input is the personal classification result obtained in step 3, and the output is the human characteristics model. A prompt sentence is sent to the generated AI model to obtain the data necessary to adjust the response.

[0808] Step 5:

[0809] The terminal interacts with customers through a conversational agent. The input is the human feature model information obtained in step 4, and the output is a suggested response to the customer. The suggested response is displayed in real time on the smart glasses or terminal screen, and the store staff uses this to provide customer service.

[0810] Step 6:

[0811] The store clerk, acting as the user, provides appropriate customer service to customers based on the responses and suggestions presented by the terminal. The input is the response suggestion obtained in step 5, and the output is the actual result of the interaction with the customer. The clerk observes customer feedback and reports it to the server via the terminal as needed.

[0812] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0813] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0814] 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 this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0815] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0816] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0817] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0818] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0819] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0820] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0821] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0822] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0823] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0824] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0825] 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.

[0826] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0827] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0828] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0829] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0830] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0831] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0832] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[0833] The following is further disclosed regarding the embodiments described above.

[0834] (Claim 1)

[0835] A means of acquiring user data from an information aggregation device,

[0836] A means for analyzing acquired user data and generating multiple user classifications,

[0837] A means for automatically generating a human model based on each user classification,

[0838] A means for creating a conversational agent that mimics a generated human model,

[0839] A means for users to conduct dialogue training using a dialogue agent,

[0840] A means of evaluating the results of dialogue training and providing feedback to users,

[0841] A system that includes this.

[0842] (Claim 2)

[0843] The system according to claim 1, comprising data retention means for storing and retrieving profiles of each human model.

[0844] (Claim 3)

[0845] The system according to claim 1, further comprising dynamic adjustment means for changing the questions and situations presented by the dialogue agent during dialogue training.

[0846] "Example 1"

[0847] (Claim 1)

[0848] Means for obtaining information from data aggregation devices,

[0849] A means for analyzing acquired information and generating multiple information classifications,

[0850] A means for automatically generating a human model based on each information classification,

[0851] A means of creating a dialogue program that mimics a generated character model,

[0852] A means by which users conduct dialogue training using a dialogue program,

[0853] A means of evaluating the results of dialogue training and providing improvement instructions to the user,

[0854] A means of grouping information using clustering techniques,

[0855] A means of constructing a dialogue program using natural language processing technology,

[0856] A system that includes this.

[0857] (Claim 2)

[0858] The system according to claim 1, comprising data storage means for storing and retrieving information for each person model.

[0859] (Claim 3)

[0860] The system according to claim 1, further comprising dynamic adjustment means for changing the questions and situations presented by the dialogue program during dialogue training.

[0861] "Application Example 1"

[0862] (Claim 1)

[0863] A means of acquiring user data from an information aggregation device,

[0864] A means for analyzing acquired user data and generating multiple user classifications,

[0865] A means for automatically generating a human model based on each user classification,

[0866] A means for creating a conversational agent that mimics a generated human model,

[0867] A means for users to conduct dialogue training using a dialogue agent,

[0868] A means of providing a real-world simulation environment via smart devices to promote skill improvement,

[0869] A means of evaluating the results of dialogue training and providing feedback to users,

[0870] A system that includes this.

[0871] (Claim 2)

[0872] The system according to claim 1, comprising data retention means for storing and retrieving profiles of each human model.

[0873] (Claim 3)

[0874] The system according to claim 1, further comprising dynamic adjustment means for changing the questions and situations presented by the dialogue agent during dialogue training.

[0875] "Example 2 of combining an emotion engine"

[0876] (Claim 1)

[0877] A means of obtaining user information from an information processing device,

[0878] A means for analyzing acquired user information and generating multiple user classifications,

[0879] A means for automatically generating a human-like model based on each user classification,

[0880] A means for creating a conversational agent that mimics a generated human-like model,

[0881] A means for acquiring user emotional information using an emotion recognition device and adjusting the response of a conversational agent,

[0882] A means for users to conduct conversation training using a conversational agent,

[0883] A means of evaluating the results of conversation training and providing emotional feedback to users,

[0884] A system that includes this.

[0885] (Claim 2)

[0886] The system according to claim 1, comprising information holding means for storing and retrieving the characteristics of each human-type model.

[0887] (Claim 3)

[0888] The system according to claim 1, further comprising a situation adjustment means for changing the questions and situations presented by the conversational agent during conversation training.

[0889] "Application example 2 when combining with an emotional engine"

[0890] (Claim 1)

[0891] Means for obtaining personal data from information processing equipment,

[0892] A means for analyzing acquired personal data and generating multiple personal classifications,

[0893] A means for automatically generating a human characteristics model based on each individual classification,

[0894] A means for creating a conversational agent that mimics a generated human characteristics model,

[0895] A means for users to conduct dialogue training using a dialogue agent,

[0896] A means of evaluating the results of dialogue training and providing feedback to users,

[0897] A means for recognizing the user's emotional state in real time using emotion analysis means, and adjusting the response of the conversational agent based on the recognized emotional state,

[0898] A system that includes this.

[0899] (Claim 2)

[0900] The system according to claim 1, comprising data holding means for storing and retrieving attribute information for each human feature model.

[0901] (Claim 3)

[0902] The system according to claim 1, further comprising a dynamic adjustment means for changing the questions and situations presented by the dialogue agent during dialogue training, and a means for suggesting customer service methods based on the emotion analysis results. [Explanation of Symbols]

[0903] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of acquiring user data from an information aggregation device, A means for analyzing acquired user data and generating multiple user classifications, A means for automatically generating a human model based on each user classification, A means for creating a conversational agent that mimics a generated human model, A means for users to conduct dialogue training using a dialogue agent, A means of evaluating the results of dialogue training and providing feedback to users, A system that includes this.

2. The system according to claim 1, comprising data retention means for saving and retrieving profiles of each human model.

3. The system according to claim 1, further comprising dynamic adjustment means for changing the questions and situations presented by the dialogue agent during dialogue training.

Images