system

Abstract

We provide the system. [Solution] A means for collecting user contract information, A means of analyzing the collected contract information, A means of preparing individual customized scenarios for each user based on the analysis results, A method for generating a user-friendly facial image for each user, A means for transmitting the generated facial image and corresponding scenario to the user terminal, A means for displaying a face image and corresponding scenario generated on the user terminal, 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 a character of the chatbot, 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] An object of the present invention is to improve customer satisfaction while reducing the cost of providing individual dedicated customer support in a mobile service provider having a large user base. In a conventional customer support system, the same response is often given to all customers, making it difficult to respond to individual needs. Also, the cost of allocating dedicated staff is high, and the problem is that effective support cannot be provided.

Means for Solving the Problems

[0005] The present invention solves the above problems by providing a system that includes the following means: means for collecting user contract information; means for analyzing the collected contract information; means for preparing individual customized response scenarios for each user based on the analysis results; means for generating a user-friendly facial image for each user; means for transmitting the generated facial image and response scenario to the user terminal; and means for displaying the generated facial image and response scenario on the user terminal. The system also includes means for analyzing the content of user inquiries and generating the optimal response, and means for collecting user feedback and reflecting it in the analysis results. This makes it possible to respond to the individual needs of users and to provide high-quality customer support while further reducing costs.

[0006] "User" refers to an individual or legal entity that uses the system.

[0007] "Contract information" refers to the content and terms of the service contract provided by the user to the system.

[0008] "Means of collection" refers to the mechanism by which the system retrieves user contract information from a database.

[0009] "Means of analysis" refers to the process of analyzing collected user contract information to identify individual needs and trends.

[0010] A "customized support scenario" refers to an individual support scenario prepared for each user based on the analysis results.

[0011] "Generation method" refers to the technology or algorithm that creates a facial image of a dedicated representative based on the user's basic information.

[0012] "Face image" refers to an image generated to provide users with a visual representation of their dedicated representative that is easy to relate to.

[0013] "Means of transmission" refers to the means of communication used to send the generated data (face image and corresponding scenario) to the user's terminal.

[0014] A "terminal" refers to a device that serves as an interface with the user (e.g., smartphones, tablets, personal computers, etc.).

[0015] "Means of display" refers to an interface that shows the facial image generated on the user's terminal and the customized response scenario.

[0016] "Inquiry content" refers to questions or requests that users make to customer support.

[0017] "Means of generating answers" refers to the technologies and processes used to create the most appropriate answers to user inquiries.

[0018] "Feedback" refers to the opinions and evaluations that users provide regarding customer support responses.

[0019] "Means of collection and analysis" refers to the technology and processes of storing user feedback in a database and analyzing it to improve the quality of future responses. [Brief explanation of the drawing]

[0020] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This 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] This 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.

MODE FOR CARRYING OUT THE INVENTION

[0021] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described according to the accompanying drawings.

[0022] First, the language used in the following description will be explained.

[0023] In the following embodiments, the signed processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Furthermore, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include CPU (Central Processing Unit), GPU (Graphics Processing Unit), GPGPU (General-Purpose computing on Graphics Processing Units), and APU (Accelerated Processing Unit).

[0024] In the following embodiments, signed RAM (Random Access Memory) is a memory that temporarily stores information and is used as work memory by the processor.

[0025] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0026] 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).

[0027] 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."

[0028] [First Embodiment]

[0029] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0030] 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.

[0031] 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).

[0032] 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.

[0033] 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.

[0034] 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.

[0035] 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.

[0036] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0037] 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.

[0038] 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.

[0039] 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.

[0040] 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".

[0041] This invention relates to a system that provides mobile service providers with personalized, dedicated customer support for a large user base. To reduce costs while improving customer satisfaction, this system comprises the means and processing steps described below.

[0042] 1. Data Collection

[0043] The server collects user contract information. This contract information includes contract details, contract period, personal attributes, and past inquiry history. This data is retrieved from the database using SQL queries.

[0044] 2. Data Analysis

[0045] The server analyzes the collected contract information. Data analysis algorithms such as clustering and categorization are used for the analysis. Based on the analysis results, the needs and inquiry trends of the contract holders are identified.

[0046] 3. Image generation for the person in charge

[0047] The server generates a personalized facial image for each user based on the analysis results. Image generation algorithms such as GANs (Generative Adversarial Networks) are used to generate the facial images. The generated facial images are customized based on the user's basic information.

[0048] 4. Preparation for individualized support

[0049] The server prepares individual customized response scenarios based on the analysis results. This includes customizing a pre-prepared FAQ database and generating answer templates based on past inquiries.

[0050] 5. Interface preparation

[0051] The server sends the generated facial image and customized scenario to the user's terminal. An HTTP request is used for this transmission. The terminal then prepares to display the transmitted facial image and scenario in its user interface.

[0052] 6. Start of response

[0053] When a user contacts customer support, the device receives the user's inquiry input and sends it to the server. The server analyzes the inquiry and generates the most appropriate response based on pre-prepared response templates. The device then displays the response to the user, along with a profile picture of the assigned support representative.

[0054] 7. Feedback on the response

[0055] After the service is completed, the device displays an interface for collecting user feedback. This feedback includes a form and a simple rating button. The user enters their feedback, the server collects the feedback information, stores it in a database, and analyzes it to improve future services.

[0056] Specific example

[0057] For new customers

[0058] 1. The server retrieves and analyzes information about newly contracted users.

[0059] 2. The server uses a GAN to generate a facial image of the user's dedicated representative.

[0060] 3. The server sends the generated facial image and customized scenario to the terminal, which then displays it.

[0061] 4. When a user makes an inquiry, the device displays "Hello, I am your personal support representative. How can I help you today?"

[0062] Measures to prevent cancellations

[0063] 1. The server retrieves and analyzes data from users who show signs of canceling their subscription.

[0064] 2. The server determines a response scenario to prevent cancellation and generates a facial image using a GAN.

[0065] 3. The server sends the generated image and corresponding scenario to the terminal, and the terminal displays it.

[0066] 4. When a user inquires about canceling their service, the device displays the message, "Hello, this is your dedicated representative. We understand you are considering canceling your service. Could you tell us what your concerns are?" and asks for the reason for cancellation.

[0067] In this way, the present invention can provide dedicated customer support tailored to the individual needs of users, thereby improving customer satisfaction while keeping costs down.

[0068] The following describes the processing flow.

[0069] Step 1:

[0070] The server retrieves the user's contract information from the database. This contract information includes contract details, contract period, personal attributes, and past inquiry history. This information is retrieved from the database using SQL queries.

[0071] Step 2:

[0072] The server analyzes the collected contract information. It uses data analysis algorithms (e.g., clustering and categorization) to analyze user needs and inquiry trends. Based on the analysis results, it determines individual response strategies.

[0073] Step 3:

[0074] The server generates a personalized facial image for each user based on the analysis results. Image generation algorithms such as GANs (Generative Adversarial Networks) are used to generate these facial images. The generated facial images are customized based on the user's basic information (e.g., age, gender, region).

[0075] Step 4:

[0076] The server prepares individual customized response scenarios based on the analysis results. These scenarios include customizing a pre-prepared FAQ database and generating answer templates based on the user's past inquiries.

[0077] Step 5:

[0078] The server sends the generated facial image and customized scenario to the user's device. This transmission uses an HTTP request. The device prepares an interface to display the transmitted facial image and corresponding scenario.

[0079] Step 6:

[0080] A user contacts customer support. The device receives this inquiry and sends the inquiry details to the server.

[0081] Step 7:

[0082] The server analyzes the user's inquiry. Based on pre-prepared response templates, it generates the most appropriate response and sends it to the terminal.

[0083] Step 8:

[0084] The terminal displays the response received from the server to the user. At the same time, a generated image of the person in charge's face is also displayed.

[0085] Step 9:

[0086] After the support is complete, the device will display an interface for collecting user feedback. This includes a feedback form and a simple rating button.

[0087] Step 10:

[0088] Users provide feedback on the response they receive. This includes evaluations and opinions about the response.

[0089] Step 11:

[0090] The server collects user feedback and stores it in a database. The collected feedback is then analyzed and used as data to improve the quality of service.

[0091] (Example 1)

[0092] 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."

[0093] Mobile service providers are required to provide individual, dedicated customer support to a large number of users, but effectively and cost-effectively serving such a large user base is difficult. Traditional systems suffer from problems such as inconsistent quality of support, long response times, and excessive resource requirements.

[0094] 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.

[0095] In this invention, the server includes means for collecting user contract information, means for analyzing the collected contract information, means for generating a user-friendly facial image based on the analysis results, means for preparing individual customized response scenarios based on the analysis results, means for transmitting the generated facial image and customized response scenario to the user terminal, and means for receiving, analyzing, and generating an optimal response to inquiries via the user terminal. This makes it possible to provide efficient and cost-effective dedicated customer support to many users and improve customer satisfaction.

[0096] "Means for collecting user contract information" refers to methods or systems for obtaining data related to services contracted by users, and includes means of collecting information from a database that includes contract details, contract period, personal attributes, past inquiries, etc.

[0097] "Means for analyzing collected contract information" refers to methods and systems for analyzing collected user contract data and extracting specific patterns or trends, and includes data analysis algorithms such as clustering and categorization.

[0098] "Means for generating friendly facial images based on analysis results" refers to methods or systems for generating different friendly facial images for each user based on the results of data analysis, and these methods utilize generative adversarial networks (GANs).

[0099] "Means for preparing individual customized response scenarios based on analysis results" refers to methods or systems for preparing different customized response scenarios for each user using the results of data analysis, and includes means for generating response templates based on a pre-prepared FAQ database or past inquiry content.

[0100] "Means for sending generated facial images and customized response scenarios to the user's terminal" refers to a method or system for sending generated facial images and customized response scenarios to the user's terminal, and is a means that uses HTTP requests.

[0101] "Means for displaying face images and customized scenarios generated on a user terminal" refers to methods or systems for displaying face images and corresponding scenarios sent to a user terminal on a user interface.

[0102] "Means for receiving, analyzing, and generating optimal responses via user terminals" refers to methods or systems for receiving user inquiries via user terminals, analyzing their content, and generating optimal responses, and which utilize natural language processing (NLP) algorithms.

[0103] "Means for collecting user feedback and improving analysis results" refers to methods and systems for collecting and storing user feedback information and using it to improve the quality of analysis results and corresponding scenarios.

[0104] This invention relates to a system that provides mobile service providers with personalized, dedicated customer support for a large user base. This system consists of multiple means and processing steps to improve customer satisfaction while keeping costs down. Each means and its associated processing steps are as follows:

[0105] First, the server collects user contract information. This information includes contract details, contract period, personal attributes, and past inquiry history. This data is retrieved from the database using SQL queries. For example, the server connects to the user database and collects data by executing the SQL query "SELECT FROM user_contracts WHERE user_id = [User ID]".

[0106] Next, the server uses K-means clustering and categorization algorithms to analyze the collected contract information. For example, it performs clustering using K-means(n_clusters=5).fit(contract information) and extracts features to identify the needs and inquiry trends of the contract holders.

[0107] Based on the analysis results, the server generates face images using a GAN (Generative Adversarial Network). In this process, it selects a GAN model with parameters optimized for a specific user category and generates friendly face images using prompts. For example, if you select StyleGAN2 as the GAN model to use and input prompts such as "Attributes for the newly generated face image: friendly, smiling, adult male," the server will generate the image. The generated face image is then temporarily stored in the server's storage.

[0108] Furthermore, the server prepares individual customized response scenarios based on the analysis results. To do this, it extracts relevant questions using a pre-prepared FAQ database and past inquiries, and generates individual answer templates. For example, it executes an SQL query such as "SELECT faq_answer FROM FAQ WHERE question LIKE '%contract renewal%'" to customize the response scenario.

[0109] The generated facial image and customized response scenario are sent from the server to the user's terminal. The HTTP protocol is used for this transmission. For example, the data is sent in the format "POST / api / send_userdata HTTP / 1.1".

[0110] The user terminal stores the received facial image and corresponding scenario in memory and prepares to display them on the user interface. Specifically, the facial image and text are appropriately positioned.

[0111] When a user submits an inquiry, the terminal receives the inquiry details. The received content is sent to the server using an HTTP request such as "POST / api / send_query HTTP / 1.1". The server analyzes the received inquiry using a natural language processing (NLP) algorithm and generates the optimal response. An example of an NLP model used is GPT-3®. Based on the inquiry, the server generates the optimal response to a question such as "Please tell me about contract renewal." The generated response is displayed on the user's terminal, along with a photo of the representative's face.

[0112] Once the interaction is complete, the user's terminal displays a feedback interface. This feedback includes a form and a simple rating button. When the user enters their feedback and presses the submit button, the server receives the feedback and saves it to the database. For example, an SQL query such as "INSERT INTO user_feedback (user_id, rating, comments) VALUES ([User ID], 5, 'I was satisfied with the interaction.')" is executed. The server analyzes the collected feedback and uses it to improve future interactions.

[0113] Example of a prompt

[0114] "Please generate a friendly, dedicated representative profile picture and support scenarios for new mobile service subscribers. Please also refer to their contract details and past inquiries."

[0115] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0116] Step 1:

[0117] The server collects user contract information from the database.

[0118] Input: User ID

[0119] Process: Execute an SQL query against the database and retrieve the corresponding row data. Example: "SELECT FROM user_contracts WHERE user_id = [User ID]"

[0120] Output: User contract information including contract details, contract period, personal attributes, and past inquiries.

[0121] Step 2:

[0122] The server analyzes the contract information it has collected.

[0123] Input: Collected contract information

[0124] Processing: The contract information is divided into clusters using the K-means clustering algorithm. Example: "K-means(n_clusters=5).fit(contract information)"

[0125] Output: Classification results and features for each user cluster

[0126] Step 3:

[0127] The server generates a friendly-looking facial image based on the analysis results.

[0128] Input: User classification results and features for each cluster

[0129] Processing: Generate face images using a GAN (Generative Adversarial Network) model. Example: "GAN model to use: StyleGAN2" "Attributes of the newly generated face image: Friendly, smiling, adult male"

[0130] Output: Generated facial image

[0131] Step 4:

[0132] The server prepares individual customized scenarios based on the analysis results.

[0133] Input: Cluster classification results, past inquiry details

[0134] Processing: Extract relevant questions from the FAQ database and generate individual answer templates. Example: "SELECT faq_answer FROM FAQ WHERE question LIKE '%contract renewal%'"

[0135] Output: Customized response scenarios and answer templates

[0136] Step 5:

[0137] The server sends the generated facial image and customized scenario to the user's terminal.

[0138] Input: Generated facial image and customizable scenario

[0139] Processing: Sends data to the user's terminal using the HTTP protocol. Example: "POST / api / send_userdata HTTP / 1.1"

[0140] Output: The transmitted data is stored on the user's terminal.

[0141] Step 6:

[0142] The user terminal displays the generated facial image and the customized scenario.

[0143] Input: Received facial image and corresponding scenario

[0144] Processing: Place and display face images and scenarios on the user interface.

[0145] Output: Face image and corresponding scenario displayed to the user

[0146] Step 7:

[0147] When a user submits an inquiry, the user's device receives the inquiry details and sends them to the server.

[0148] Input: User inquiry details

[0149] Processing: Receives the query content and sends it to the server as an HTTP request. Example: "POST / api / send_query HTTP / 1.1"

[0150] Output: Sent inquiry content

[0151] Step 8:

[0152] The server analyzes the inquiry and generates the most appropriate response.

[0153] Input: User inquiry details

[0154] Processing: Analyze the query using a natural language processing (NLP) algorithm and generate the optimal response. NLP model used: GPT-3

[0155] Output: Generated answer

[0156] Step 9:

[0157] The user's terminal displays the generated response to the user, along with the face image of the person in charge.

[0158] Input: Generated response and facial image

[0159] Processing: Display the answer on the user interface, along with the facial image.

[0160] Output: Responses and facial images displayed to the user

[0161] Step 10:

[0162] Once the issue is resolved, the user's device will display a feedback interface.

[0163] Input: Trigger for completion of response

[0164] Processing: Display a feedback form and a simple rating button on the interface.

[0165] Output: User-inputtable feedback interface

[0166] Step 11:

[0167] When a user enters feedback and presses the submit button, the server receives the feedback and saves it to the database.

[0168] Input: User feedback content

[0169] Process: Receive the feedback and save it to the database. Example: "INSERT INTO user_feedback (user_id, rating, comments) VALUES ([User ID], 5, 'I was satisfied with the service.')"

[0170] Output: Saved feedback data

[0171] Step 12:

[0172] The server will analyze the collected feedback and use it to improve future responses.

[0173] Input: Saved feedback data

[0174] Processing: Analyze feedback data and use it to improve the service.

[0175] Output: Insights for improved response scenarios and system updates

[0176] (Application Example 1)

[0177] 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."

[0178] In food service provision, a challenge is responding quickly and individually to customers' specific requests and inquiries. Traditional customer support systems often provide uniform responses, making it difficult to improve customer satisfaction and provide efficient support. In particular, in food service, responses must be based on customer preferences and past order history, and traditional systems often lacked flexibility and personalization.

[0179] 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.

[0180] In this invention, the server includes means for collecting user contract information, means for analyzing the collected contract information, means for preparing individual customized response scenarios for each user based on the analysis results, means for generating a user-friendly facial image for each user, means for transmitting the generated facial image and response scenario to the user terminal, means for displaying the generated facial image and response scenario on the user terminal, and means for functioning as a dedicated support person for food services via a smart device. This enables prompt and individual responses to customers' special requests and inquiries, improving customer satisfaction and providing efficient support.

[0181] "User contract information" refers to information such as the contract details, contract period, personal attributes, and past inquiries of food service users.

[0182] "Means of collection" refers to the technical means of obtaining user contract information from a database.

[0183] "Means of analysis" refers to technical means of analyzing collected information using data analysis algorithms such as clustering and categorization.

[0184] "Individualized customized response scenarios" refer to specific response plans or scenarios prepared for individual users based on the analysis results.

[0185] A "friendly face image" refers to a customized face image that users can find appealing.

[0186] "Generative means" refers to technical methods for generating friendly facial images using GANs (Generative Adversarial Networks).

[0187] "Means of transmission" refers to the technical means for transmitting the generated facial image and corresponding scenario to the user's terminal.

[0188] "Means of display" refers to the technical means for displaying face images and corresponding scenarios generated on the user's terminal.

[0189] A "smart device" refers to a portable information terminal with internet connectivity, such as a smartphone or tablet.

[0190] A "dedicated support representative" refers to a virtual support representative who is assigned to each individual user.

[0191] This invention relates to a dedicated support system for users in the food service industry. This system is configured to provide food service providers with personalized, dedicated customer support to a large user base and is implemented through the following elements.

[0192] The server first collects user contract information. This contract information includes the user's contract details, contract period, personal attributes, and past inquiry history. This data is retrieved from an SQL database using SQL queries.

[0193] Next, the server analyzes the collected contract information. Data analysis algorithms such as clustering and categorization are used for this analysis. Python's scikit-learn library is used for data analysis. Based on the analysis results, the needs and inquiry trends of the contract holders are identified.

[0194] Based on the analysis results, the server generates a user-friendly facial image for each user. Image generation algorithms such as GANs (Generative Adversarial Networks) are used to generate the facial images. The generated facial images are customized based on the user's basic information. In this process, deep learning frameworks such as TENSORFLOW® and PyTorch are used to implement the GANs.

[0195] The server prepares individual customized response scenarios based on the analysis results. This includes customizing the pre-prepared FAQ database and generating answer templates based on past inquiries. This improves the quality and efficiency of responses.

[0196] The prepared facial image and customized scenario are sent from the server to the user's terminal. HTTP requests are used for this transmission. The smart device (smartphone or tablet) prepares to display the transmitted facial image and corresponding scenario in the user interface.

[0197] When a user contacts customer support, the device receives the user's inquiry input and sends it to the server. The server analyzes the inquiry and generates the most appropriate response based on pre-prepared response templates. The device then displays the response to the user, along with a profile picture of the assigned support representative.

[0198] After the service is completed, the device displays an interface for collecting user feedback. This feedback includes a form and a simple rating button. The user enters their feedback, the server collects the feedback information, stores it in a database, and analyzes it to improve future services.

[0199] Specific example

[0200] For example, if user A has a habit of ordering a specific pizza every Friday, the server will use this information to create a dedicated support representative for user A and prepare support scenarios to address any special requests regarding the pizza. When user A requests support, they will be shown a message such as, "Hello, I am your dedicated support representative. Do you have any questions regarding your pizza order?"

[0201] Examples of input prompts for a generative AI model

[0202] "Based on User A's past order history, generate a facial image of a dedicated support representative and a customized scenario for ordering pizza."

[0203] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0204] Step 1:

[0205] The server collects user contract information.

[0206] Input: User ID

[0207] Operation: The server queries the SQL database to retrieve the user's contract details, contract period, personal attributes, and past query history.

[0208] Output: User contract information

[0209] Step 2:

[0210] The server analyzes the collected contract information.

[0211] Input: User contract information

[0212] Operation: The server analyzes contract information using data analysis algorithms such as clustering and categorization. Specifically, it uses the Python scikit-learn library to cluster the data.

[0213] Output: Analysis results (user needs and inquiry trends)

[0214] Step 3:

[0215] The server generates a user-friendly facial image based on the analysis results.

[0216] Input: Analysis results

[0217] Operation: The server uses a GAN (Generative Adversarial Network) to generate user-friendly facial images. TensorFlow and PyTorch are used as deep learning frameworks for this purpose.

[0218] Output: Friendly facial image

[0219] Step 4:

[0220] The server prepares individual customized response scenarios based on the analysis results.

[0221] Input: Analysis results

[0222] Operation: The server customizes a pre-prepared FAQ database and generates answer templates based on past inquiries.

[0223] Output: Customizable scenarios

[0224] Step 5:

[0225] The server sends the generated facial image and customized scenario to the user's terminal.

[0226] Input: Friendly facial image, customizable scenarios

[0227] Operation: The server uses an HTTP request to send the generated facial image and customizable scenario to the user's smart device.

[0228] Output: Face image and corresponding scenario sent to the user's terminal

[0229] Step 6:

[0230] The user's terminal displays the generated facial image and the customizable scenario.

[0231] Input: Face images sent from the server and customizable scenarios.

[0232] Operation: The user terminal displays the transmitted facial image and customizable scenario on the user interface.

[0233] Output: Displayed facial image and corresponding scenario

[0234] Step 7:

[0235] The user contacts customer support.

[0236] Input: User inquiry

[0237] Operation: The user contacts customer support via a smart device. The device sends this inquiry to the server.

[0238] Output: User queries sent to the server

[0239] Step 8:

[0240] The server analyzes the user's inquiry and generates the most appropriate response.

[0241] Input: User inquiry

[0242] Operation: The server analyzes the query text and generates the most suitable response based on pre-prepared response templates.

[0243] Output: Best Answer

[0244] Step 9:

[0245] The device displays the answer to the user, along with a generated image of the person in charge's face.

[0246] Input: Best response sent from the server, face image of the person in charge

[0247] Operation: The device displays the answer and the face image of the person in charge on the screen.

[0248] Output: Displayed response and the respondent's face image

[0249] Step 10:

[0250] The user enters feedback, and the server collects the feedback information and stores it in a database.

[0251] Input: User feedback

[0252] Operation: The terminal displays a feedback collection interface, and the user enters feedback. The entered feedback is sent to the server and stored in the database.

[0253] Output: Saved feedback information

[0254] Examples of input prompts for a generative AI model

[0255] "Based on User A's past order history, generate a facial image of a dedicated support representative and a customized scenario for ordering pizza."

[0256] 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.

[0257] This invention aims to further improve customer satisfaction by combining a system that provides mobile service providers with personalized, dedicated customer support for a large user base with an emotion engine that recognizes user emotions. This system consists of the means and processing steps described below.

[0258] 1. Data Collection

[0259] The server retrieves user contract information from the database. This contract information includes contract details, contract period, personal attributes, and past inquiry history. This data is retrieved from the database using SQL queries.

[0260] 2. Data Analysis

[0261] The server analyzes the collected contract information. It uses data analysis algorithms (e.g., clustering and categorization) to analyze user needs and inquiry trends. Based on the analysis results, it determines individual response strategies.

[0262] 3. Image generation for the person in charge

[0263] The server generates a personalized facial image for each user based on the analysis results. Image generation algorithms such as GANs (Generative Adversarial Networks) are used to generate these facial images. The generated facial images are customized based on the user's basic information (e.g., age, gender, region).

[0264] 4. Preparation for individualized support

[0265] The server prepares individual customized response scenarios based on the analysis results. These scenarios include customizing a pre-prepared FAQ database and generating answer templates based on the user's past inquiries.

[0266] 5. Interface preparation

[0267] The server sends the generated facial image and customized scenario to the user's device. An HTTP request is used for this transmission. The device then prepares an interface to display the transmitted facial image and corresponding scenario.

[0268] 6. Emotion recognition

[0269] The device uses an emotion engine to recognize emotions from the user's facial expressions and voice. The emotion engine identifies the emotions the user is feeling in real time through facial expression recognition and voice analysis.

[0270] 7. Start of response

[0271] A user contacts customer support. The device receives this inquiry and sends the inquiry details to the server.

[0272] 8. Emotion-based dynamic adjustment

[0273] The server analyzes the content of the user's inquiry and the emotion data transmitted from the terminal. Based on the analysis results, it dynamically adjusts the customization response scenarios and answer templates to generate an optimal answer corresponding to the user's current emotion.

[0274] 9. Display of Answer

[0275] The terminal displays the answer content received from the server to the user. At this time, the generated face image of the person in charge is also displayed together, and the output of voice and text is adjusted according to the user's emotion.

[0276] 10. Feedback on Response Content

[0277] The terminal displays an interface for collecting feedback from the user. This includes a feedback form and a simple evaluation button.

[0278] 11. Collection and Analysis of Feedback

[0279] The user inputs feedback on the response. This includes evaluations and opinions on the response content. The server collects the user's feedback and saves it in the database. Then, the collected feedback is analyzed and utilized as data for improving the quality of the response.

[0280] Specific Example

[0281] For New Contract Customers

[0282] 1. The server acquires and analyzes the information of the newly contracted user.

[0283] 2. The server uses a GAN to generate a face image of the person in charge exclusive to the user.

[0284] 3. The server transmits the generated face image and the customization response scenario to the terminal, and the terminal displays them.

[0285] 4. The terminal analyzes the user's expression and voice with an emotion engine to recognize the user's emotion.

[0286] 5. When the user makes an inquiry, the terminal displays "Hello, I am your dedicated support staff. How can I assist you today?"

[0287] Measures to prevent contract termination

[0288] 1. The server acquires and analyzes the data of users who show signs of contract termination.

[0289] 2. The server determines a response scenario for preventing contract termination and generates a face image using a GAN.

[0290] 3. The server transmits the generated image and the response scenario to the terminal, and the terminal displays them.

[0291] 4. The terminal analyzes the user's expression and voice with an emotion engine to recognize the user's emotion.

[0292] 5. When the user makes an inquiry about contract termination, the terminal displays "Hello, I am your dedicated staff. I heard that you are considering contract termination. May I ask if you have any dissatisfaction?" and makes an appropriate response according to the user's emotion.

[0293] In this way, the present invention provides dedicated customer support according to individual needs based on the user's emotion, and further improves customer satisfaction.

[0294] The following describes the processing flow.

[0295] Step 1:

[0296] The server acquires the user's contract information from the database. The contract information includes contract content, contract period, personal attributes, and past inquiry content. These data are acquired from the database using an SQL query.

[0297] Step 2:

[0298] The server analyzes the collected contract information. Using data analysis algorithms such as clustering and categorization, it analyzes the needs and inquiry trends of users. Based on the analysis results, it determines individual response policies for users.

[0299] Step 3:

[0300] Based on the analysis results, the server generates a friendly face image for each user. Using an image generation algorithm such as GAN (Generative Adversarial Network), it customizes the face image based on the user's basic information (age, gender, region, etc.).

[0301] Step 4:

[0302] The server prepares an individual customization response scenario. It customizes the pre-prepared FAQ database and generates a response template based on the user's past inquiry content.

[0303] Step 5:

[0304] The server sends the generated face image and the customization response scenario to the user terminal. Using an HTTP request, it sends the face image and scenario data to the terminal.

[0305] Step 6:

[0306] The terminal displays the generated face image and the customization response scenario. Here, a user interface is prepared, and the scenario and face image are visually displayed to the user.

[0307] Step 7:

[0308] The device analyzes the user's facial expressions and voice, and recognizes emotions using an emotion engine. The emotion engine utilizes facial recognition and voice analysis algorithms to identify the user's emotions (e.g., joy, anger, sadness, etc.) in real time.

[0309] Step 8:

[0310] A user contacts customer support. The device receives the user's inquiry and sends it to the server.

[0311] Step 9:

[0312] The server analyzes the user's inquiry and the sentiment data recognized by the sentiment engine. Based on the analysis results, it generates the optimal response corresponding to the user's emotions and dynamically adjusts the customized response scenario.

[0313] Step 10:

[0314] The server sends the generated response to the terminal. The response data is sent to the terminal as an HTTP response.

[0315] Step 11:

[0316] The terminal displays the response from the server to the user. Along with the generated image of the person in charge, it displays the response in voice or text format, adjusted to the user's emotions.

[0317] Step 12:

[0318] After the support is complete, the device will display an interface for collecting feedback from the user. It will provide a feedback form and a simple rating button to receive opinions and ratings from the user.

[0319] Step 13:

[0320] Users enter feedback. This feedback includes evaluations and opinions about the actions taken.

[0321] Step 14:

[0322] The server collects user feedback and stores it in a database. The collected feedback is analyzed and used as data to improve the quality of future responses.

[0323] (Example 2)

[0324] 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".

[0325] Traditional customer support systems can provide personalized support based on user contract information, but they lack the ability to recognize user emotions and dynamically adjust responses, limiting their potential for improving customer satisfaction. Furthermore, they lack mechanisms to analyze user feedback and improve responses, resulting in insufficient improvement in service quality.

[0326] 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.

[0327] In this invention, the server includes means for collecting user contract information, means for analyzing the collected contract information, means for preparing individual customized response scenarios for each user based on the analysis results, means for generating a user-friendly facial image for each user, means for transmitting the generated facial image and response scenario to the user terminal, means for collecting emotional data from the user, means for analyzing the user's emotional data and dynamically adjusting the customized response scenario, and means for displaying the generated facial image and dynamically adjusted response scenario on the user terminal. This enables individualized responses based on the user's emotions, thereby improving customer satisfaction.

[0328] "Contract information" refers to data such as the details of the contract a mobile service user has made, the contract period, personal attributes, and past inquiries.

[0329] A "database" is a system used to store and manage user contract information, inquiry details, and other data.

[0330] "Data analysis" is the process of analyzing user needs and inquiry trends based on collected contract information, and determining individual response strategies.

[0331] Clustering is a technique for grouping and classifying user data based on specific criteria.

[0332] A "face image" is an image of a human face that is generated to make the user feel a sense of familiarity.

[0333] GAN (Generative Adversarial Network) is a type of image generation algorithm that generates high-quality images by pitting two neural networks against each other.

[0334] A "customized scenario" is a scenario created for each user based on the analysis results, designed to address their specific needs.

[0335] An "emotion engine" is software that recognizes and analyzes emotions from a user's facial expressions and voice.

[0336] An "HTTP request" is a protocol used to send and receive data between a server and a terminal.

[0337] "Dynamic adjustment" is a process that optimizes response scenarios in real time based on user emotion data.

[0338] "Feedback" refers to the evaluations and opinions that users provide regarding the response.

[0339] Modes for carrying out the invention

[0340] This invention aims to further improve customer satisfaction by combining a system that provides mobile service providers with personalized, dedicated customer support for large user bases with an emotion engine that recognizes user emotions. This system links a server and a terminal and provides a function to dynamically adjust individual responses using user contract information and emotion data.

[0341] Data collection

[0342] The server retrieves user contract information from the database. This contract information includes contract details, contract period, personal attributes, and past inquiry history. This data is retrieved from the database using SQL queries. For example, an SQL query such as "SELECT FROM user_contracts WHERE user_id = '<user ID>'" is used.

[0343] Data Analysis

[0344] The server analyzes the collected contract information. It uses data analysis algorithms (e.g., clustering and categorization) to analyze user needs and inquiry trends. For clustering, for example, the KMeans algorithm is used.

[0345] Image generation for the person in charge

[0346] The server generates a personalized facial image for each user based on the analysis results. This facial image generation uses a Generative Adversarial Network (GAN). For example, a GAN model can be used to generate a customized facial image based on the user's age, gender, and location.

[0347] Preparation for individual support

[0348] The server prepares individual customized response scenarios based on the analysis results. These scenarios include customizing a pre-prepared FAQ database and generating answer templates based on the user's past inquiries. Appropriate content is extracted from the FAQ database and the answer templates are customized.

[0349] Interface preparation

[0350] The server sends the generated facial image and customizable scenario to the user's device. This communication uses HTTP requests. Based on the transmitted data, the device prepares a display interface and presents it to the user.

[0351] emotion recognition

[0352] The device uses an emotion engine to recognize emotions from the user's facial expressions and voice. This emotion engine identifies the emotions the user is feeling in real time through facial expression recognition and voice analysis. For example, the EmotionEngine is used to analyze the user's emotions.

[0353] Start of response

[0354] A user contacts customer support. The device receives this inquiry and sends the inquiry details to the server. For example, the user types "I can't connect to the internet" and sends it from the device to the server.

[0355] Emotion-based dynamic adjustment

[0356] The server analyzes the user's inquiry and sentiment data sent from the device. Based on the analysis results, it dynamically adjusts customized response scenarios and answer templates to generate the optimal response that corresponds to the user's current sentiment. It generates a dynamically adjusted response based on sentiment data and the inquiry content.

[0357] Display the answer

[0358] The terminal displays the response received from the server to the user. At the same time, a generated image of the support representative's face is also displayed, and the output of voice and text is adjusted according to the user's emotions. For example, it might display, "Hello, I'm your personal support representative. How can I help you today?"

[0359] Feedback on the response

[0360] The device displays an interface for collecting user feedback. This includes a feedback form and a quick rating button. It receives user feedback and sends it to the server.

[0361] Feedback collection and analysis

[0362] The server collects user feedback and stores it in a database. The collected feedback is then analyzed and used as data to improve the quality of service. For example, the feedback can be analyzed to improve the service.

[0363] Specific example

[0364] For new customers

[0365] 1. The server retrieves and analyzes information about newly contracted users.

[0366] 2. The server uses a GAN to generate a facial image of the user's dedicated representative.

[0367] 3. The server sends the generated facial image and customized scenario to the terminal, which then displays it.

[0368] 4. The device analyzes the user's facial expressions and voice using an emotion engine to recognize the user's emotions.

[0369] 5. When a user makes an inquiry, the device displays "Hello, I am your personal support representative. How can I help you today?"

[0370] Measures to prevent cancellations

[0371] 1. The server retrieves and analyzes data from users who show signs of canceling their subscription.

[0372] 2. The server determines a response scenario to prevent cancellation and generates a facial image using a GAN.

[0373] 3. The server sends the generated image and corresponding scenario to the terminal, and the terminal displays it.

[0374] 4. The device analyzes the user's facial expressions and voice using an emotion engine to recognize the user's emotions.

[0375] 5. When a user inquires about canceling their service, the device will display a message saying, "Hello, this is your dedicated representative. We understand you are considering canceling your service. May we hear about any points of dissatisfaction?" and will respond appropriately based on the user's feelings.

[0376] Examples of prompts include, "Create customer support scenarios tailored to the emotions of newly signed-up users," and "Generate response scenarios that include emotion recognition for users considering cancellation."

[0377] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0378] Step 1:

[0379] The server retrieves the user's contract information from the database.

[0380] Specifically, the server connects to the database and executes an SQL query to extract the user's contract information.

[0381] Input: User ID

[0382] Output: User contract information (contract details, contract period, personal attributes, past inquiries)

[0383] The server executes the SQL query "SELECT FROM user_contracts WHERE user_id = '<User ID>'" to retrieve relevant information from the database.

[0384] Step 2:

[0385] The server analyzes the collected contract information.

[0386] In terms of specific operations, the server performs analysis using data analysis algorithms (clustering and categorization). Python's KMeans algorithm, for example, can be used.

[0387] Input: User's contract information

[0388] Output: Analysis results (user needs and inquiry trends)

[0389] Based on the contract information collected by the server, users are classified using the KMeans algorithm, and analysis results are obtained.

[0390] Step 3:

[0391] The server generates a user-friendly facial image based on the analysis results.

[0392] Specifically, the server uses a GAN to generate a facial image based on the user's basic information.

[0393] Input: Analysis results, user's basic information (age, gender, region)

[0394] Output: Generated facial image

[0395] The server uses a GAN model to generate facial images based on, for example, the user's age and gender.

[0396] Step 4:

[0397] The server prepares individual customized response scenarios based on the analysis results.

[0398] Specifically, the server customizes a pre-prepared FAQ database and generates answer templates based on past inquiries.

[0399] Input: Analysis results, FAQ database, past inquiry content

[0400] Output: Customizable scenarios, response templates

[0401] The server creates customized response scenarios and answer templates based on the FAQ database and past inquiries.

[0402] Step 5:

[0403] The server sends the generated facial image and customized scenario to the user's device.

[0404] In terms of specific operations, the server sends data to the terminal using an HTTP request.

[0405] Input: Generated facial image, customizable scenario

[0406] Output: Transmission status

[0407] The server uses an HTTP request to send the generated facial image and corresponding scenario to the terminal.

[0408] Step 6:

[0409] The device uses an emotion engine to recognize emotions from the user's facial expressions and voice.

[0410] Specifically, the device performs facial recognition and voice analysis, and identifies emotions through an emotion engine.

[0411] Input: User facial expression data, voice data

[0412] Output: User sentiment data

[0413] The device uses EmotionEngine to analyze, for example, the user's facial expressions and voice in real time, and obtain emotional data.

[0414] Step 7:

[0415] The user contacts customer support.

[0416] In terms of specific actions, the user enters their question using a contact form on their device.

[0417] Input: User's inquiry

[0418] Output: Query data

[0419] The user enters their inquiry into the device, and that information is sent to the device.

[0420] Step 8:

[0421] The terminal receives this query and sends the query details to the server.

[0422] Specifically, the terminal sends an HTTP request to forward the inquiry details to the server.

[0423] Input: Inquiry data

[0424] Output: Transmission status

[0425] The device sends an HTTP request containing the inquiry details to the server.

[0426] Step 9:

[0427] The server analyzes the user's inquiry and sentiment data sent from the device.

[0428] Specifically, the server analyzes sentiment data and query content, and dynamically adjusts the customized response scenario.

[0429] Input: Inquiry details, sentiment data

[0430] Output: Dynamically adjusted response scenarios, response templates

[0431] The server integrates inquiry content and sentiment data to generate optimized response scenarios and answers.

[0432] Step 10:

[0433] The terminal displays the response received from the server to the user.

[0434] In terms of specific operation, the terminal displays the received response and the face image of the person in charge on the display interface.

[0435] Input: Response content from the server, generated facial image

[0436] Output: Content displayed to the user

[0437] Based on the information received by the terminal from the server, the response content and facial image are displayed on the screen.

[0438] Step 11:

[0439] The device displays an interface for collecting feedback from the user.

[0440] In terms of specific actions, the device will present the user with a feedback form and rating buttons.

[0441] Input: None

[0442] Output: Feedback input interface

[0443] The device displays a form or button for collecting feedback.

[0444] Step 12:

[0445] Users provide feedback on the response.

[0446] In terms of specific actions, users enter their ratings and opinions into a feedback form and submit it.

[0447] Input: User feedback content

[0448] Output: Feedback data

[0449] The user enters feedback and sends it to their device.

[0450] Step 13:

[0451] The server collects user feedback and stores it in a database.

[0452] Specifically, the server adds the received feedback data to the database.

[0453] Input: User feedback content

[0454] Output: Saved status

[0455] The server receives the feedback data and saves it to the database.

[0456] Step 14:

[0457] The server analyzes the collected feedback and uses it as data to improve the quality of service.

[0458] Specifically, the server analyzes feedback data to gain insights that can be used to improve the quality of the service.

[0459] Input: Feedback data

[0460] Output: Improvement suggestions, analysis results

[0461] The server analyzes the feedback data and creates specific suggestions for service improvement.

[0462] (Application Example 2)

[0463] 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".

[0464] Traditional content delivery services have not adequately addressed individual user needs and emotions, making it difficult to improve user satisfaction. In particular, the lack of real-time content recommendations based on emotional changes during viewing, and the absence of dynamic interface adjustments, sometimes led to user stress and decreased satisfaction.

[0465] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for collecting user contract information, means for analyzing the collected contract information, means for preparing individual customized response scenarios for each user based on the analysis results, means for generating a user-friendly face image for each user, means for transmitting the generated face image and response scenario to the user terminal, means for displaying the generated face image and response scenario on the user terminal, means for collecting and analyzing user emotion data, means for dynamically recommending content based on the user's emotions, means for dynamically adjusting the interface according to the user's emotions, and means for collecting and improving feedback from the user. This enables real-time personalized content recommendation and interface adjustment in accordance with the user's emotions and needs, thereby improving user satisfaction.

[0466] "User contract information" refers to information such as the user's contract details, contract period, personal attributes, and past inquiries.

[0467] "Analysis results" refer to conclusions and insights derived by the server based on the collected data.

[0468] A "customized response scenario" refers to a pre-prepared plan and flow of individual responses based on user needs and inquiry trends.

[0469] A "face image" is a friendly, customized image of a face created using a Generative Adversarial Network (GAN).

[0470] A "user terminal" refers to a device used by a user, such as a smartphone, tablet, or personal computer.

[0471] "Emotional data" refers to information about a user's emotions, obtained in real time from their facial expressions and voice.

[0472] "A means of dynamically recommending content" refers to a technology that recommends appropriate content in real time based on user sentiment data and viewing history.

[0473] "Methods for dynamically adjusting the interface" refer to techniques that change the design and layout of an application's user interface (UI) in response to the user's emotions.

[0474] "Methods for collecting feedback and making improvements" refers to technologies that collect evaluations and opinions from users and use that information to perform data analysis and implement improvement measures to enhance future services.

[0475] This invention is a system for delivering dynamic content based on user emotions. First, a server collects user contract information and analyzes it. The analysis uses data mining techniques such as clustering and categorization. Next, based on the analysis results, a customized response scenario is prepared for each user, and a friendly facial image is generated. A Generative Adversarial Network (GAN) is used to generate this facial image. The generated facial image and response scenario are sent to the user's terminal.

[0476] On the user's terminal, the generated facial image and corresponding scenario are displayed, and the user's emotional data is collected in real time. Facial expression recognition and voice analysis technologies using a camera and microphone are employed to collect emotional data. Based on the emotional analysis results, the server dynamically recommends specific content. For example, a user experiencing stress might be recommended relaxing music or a comedy movie.

[0477] Furthermore, the application's user interface (UI) is dynamically adjusted according to the user's emotions. This means, for example, changing the UI design to a simpler and calmer one if the user is feeling stressed.

[0478] In addition, user feedback is collected and incorporated into subsequent analyses and scenario generation. This ensures continuous improvement of the service.

[0479] As a concrete example, when a user opens the application while feeling stressed, the server recommends a relaxing music playlist. The UI also automatically changes to calming colors to help reduce user stress. Afterwards, the user enters feedback on the content they viewed into a form within the application, and this data is sent to the server and used to improve the recommendation algorithm for the next time.

[0480] The program to realize this invention will be implemented using a general-purpose programming language such as Python. The libraries used will be "some_emotion_recognition_library" for emotion recognition and "some_content_recommendation_library" for content recommendation. Libraries such as "requests" and "SQLAlchemy" will be used for HTTP requests and database access.

[0481] Examples of prompt statements to input into the generative AI model are as follows:

[0482] "Design an algorithm that recommends relaxing music and movies based on the user's viewing history data and real-time sentiment analysis data. This algorithm should also consider past rating data to recommend content that best suits the user's current mood in real time. Additionally, the UI theme should dynamically change according to the user's mood."

[0483] In this way, a system is realized that dynamically recommends content based on user emotions, adjusts the interface, and incorporates feedback.

[0484] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0485] Step 1:

[0486] The server collects user contract information from the database. Input is identification information such as the user ID, and output is user contract information (contract details, contract period, personal attributes, past inquiry history). Specifically, it executes SQL queries to extract the necessary data.

[0487] Step 2:

[0488] The server analyzes the collected contract information. The input is the contract information obtained in step 1, and the output is the analysis results regarding user needs and inquiry trends. Data analysis algorithms (e.g., clustering and categorization) are used to group the data and derive insights.

[0489] Step 3:

[0490] The server generates user-specific, customized response scenarios and friendly facial images based on the analysis results. The input is the analysis results from step 2, and the output is the customized scenario and facial image corresponding to the user. A Generative Adversarial Network (GAN) is used to generate facial images and customize pre-defined response scenarios.

[0491] Step 4:

[0492] The server sends the generated face image and corresponding scenario to the user terminal. The input is the face image and corresponding scenario generated in step 3, and the output is the face image and corresponding scenario sent to the user terminal. Data is sent using an HTTP request.

[0493] Step 5:

[0494] The user terminal displays the generated face image and a customizable scenario. The input is the face image and corresponding scenario sent from the server, and the output is a visual display that the user can confirm on their terminal. The interface displays the face image and text.

[0495] Step 6:

[0496] The user terminal collects and analyzes the user's emotional data (facial expressions and voice) in real time. The input is the user's facial and voice data, and the output is the analysis results regarding the user's emotions. Specifically, it uses a camera and microphone and an emotion recognition engine to identify emotions.

[0497] Step 7:

[0498] The server dynamically recommends the most suitable content to the user based on sentiment data. The input is the sentiment data from step 6 and the user's viewing history, and the output is the recommended content. A content recommendation algorithm is used to select the content best suited to the user's current sentiment.

[0499] Step 8:

[0500] The user's device dynamically adjusts the interface according to the user's emotions. The input is the content and emotion data recommended in step 7, and the output is the UI adjusted according to the emotion. For example, a user feeling stressed will be shown a UI with calming colors.

[0501] Step 9:

[0502] Users view content and then provide feedback. The input is the user's feedback, and the output is feedback data. By filling out the feedback form on their device, the data is sent to the server.

[0503] Step 10:

[0504] The server analyzes the collected feedback data to improve response scenarios and recommendation algorithms. The input is the feedback collected in step 9, and the output is the improved response scenarios and recommendation algorithms. Analysis based on the feedback data improves system performance.

[0505] The above steps enable a dynamic content delivery system based on user emotions.

[0506] 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.

[0507] 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.

[0508] 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.

[0509] [Second Embodiment]

[0510] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0511] 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.

[0512] 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).

[0513] 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.

[0514] 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.

[0515] 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).

[0516] 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.

[0517] 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.

[0518] 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.

[0519] 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.

[0520] 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.

[0521] 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".

[0522] This invention relates to a system that provides mobile service providers with personalized, dedicated customer support for a large user base. To reduce costs while improving customer satisfaction, this system comprises the means and processing steps described below.

[0523] 1. Data Collection

[0524] The server collects user contract information. This contract information includes contract details, contract period, personal attributes, and past inquiry history. This data is retrieved from the database using SQL queries.

[0525] 2. Data Analysis

[0526] The server analyzes the collected contract information. Data analysis algorithms such as clustering and categorization are used for the analysis. Based on the analysis results, the needs and inquiry trends of the contract holders are identified.

[0527] 3. Image generation for the person in charge

[0528] The server generates a personalized facial image for each user based on the analysis results. Image generation algorithms such as GANs (Generative Adversarial Networks) are used to generate the facial images. The generated facial images are customized based on the user's basic information.

[0529] 4. Preparation for individualized support

[0530] The server prepares individual customized response scenarios based on the analysis results. This includes customizing a pre-prepared FAQ database and generating answer templates based on past inquiries.

[0531] 5. Interface preparation

[0532] The server sends the generated facial image and customized scenario to the user's terminal. An HTTP request is used for this transmission. The terminal then prepares to display the transmitted facial image and scenario in its user interface.

[0533] 6. Start of response

[0534] When a user contacts customer support, the device receives the user's inquiry input and sends it to the server. The server analyzes the inquiry and generates the most appropriate response based on pre-prepared response templates. The device then displays the response to the user, along with a profile picture of the assigned support representative.

[0535] 7. Feedback on the response

[0536] After the service is completed, the device displays an interface for collecting user feedback. This feedback includes a form and a simple rating button. The user enters their feedback, the server collects the feedback information, stores it in a database, and analyzes it to improve future services.

[0537] Specific example

[0538] For new customers

[0539] 1. The server retrieves and analyzes information about newly contracted users.

[0540] 2. The server uses a GAN to generate a facial image of the user's dedicated representative.

[0541] 3. The server sends the generated facial image and customized scenario to the terminal, which then displays it.

[0542] 4. When a user makes an inquiry, the device displays "Hello, I am your personal support representative. How can I help you today?"

[0543] Measures to prevent cancellations

[0544] 1. The server retrieves and analyzes data from users who show signs of canceling their subscription.

[0545] 2. The server determines a response scenario to prevent cancellation and generates a facial image using a GAN.

[0546] 3. The server sends the generated image and corresponding scenario to the terminal, and the terminal displays it.

[0547] 4. When a user inquires about canceling their service, the device displays the message, "Hello, this is your dedicated representative. We understand you are considering canceling your service. Could you tell us what your concerns are?" and asks for the reason for cancellation.

[0548] In this way, the present invention can provide dedicated customer support tailored to the individual needs of users, thereby improving customer satisfaction while keeping costs down.

[0549] The following describes the processing flow.

[0550] Step 1:

[0551] The server retrieves the user's contract information from the database. This contract information includes contract details, contract period, personal attributes, and past inquiry history. This information is retrieved from the database using SQL queries.

[0552] Step 2:

[0553] The server analyzes the collected contract information. It uses data analysis algorithms (e.g., clustering and categorization) to analyze user needs and inquiry trends. Based on the analysis results, it determines individual response strategies.

[0554] Step 3:

[0555] The server generates a personalized facial image for each user based on the analysis results. Image generation algorithms such as GANs (Generative Adversarial Networks) are used to generate these facial images. The generated facial images are customized based on the user's basic information (e.g., age, gender, region).

[0556] Step 4:

[0557] The server prepares individual customized response scenarios based on the analysis results. These scenarios include customizing a pre-prepared FAQ database and generating answer templates based on the user's past inquiries.

[0558] Step 5:

[0559] The server sends the generated facial image and customized scenario to the user's device. This transmission uses an HTTP request. The device prepares an interface to display the transmitted facial image and corresponding scenario.

[0560] Step 6:

[0561] A user contacts customer support. The device receives this inquiry and sends the inquiry details to the server.

[0562] Step 7:

[0563] The server analyzes the user's inquiry. Based on pre-prepared response templates, it generates the most appropriate response and sends it to the terminal.

[0564] Step 8:

[0565] The terminal displays the response received from the server to the user. At the same time, a generated image of the person in charge's face is also displayed.

[0566] Step 9:

[0567] After the support is complete, the device will display an interface for collecting user feedback. This includes a feedback form and a simple rating button.

[0568] Step 10:

[0569] Users provide feedback on the response they receive. This includes evaluations and opinions about the response.

[0570] Step 11:

[0571] The server collects user feedback and stores it in a database. The collected feedback is then analyzed and used as data to improve the quality of service.

[0572] (Example 1)

[0573] 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."

[0574] Mobile service providers are required to provide individual, dedicated customer support to a large number of users, but effectively and cost-effectively serving such a large user base is difficult. Traditional systems suffer from problems such as inconsistent quality of support, long response times, and excessive resource requirements.

[0575] 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.

[0576] In this invention, the server includes means for collecting user contract information, means for analyzing the collected contract information, means for generating a user-friendly facial image based on the analysis results, means for preparing individual customized response scenarios based on the analysis results, means for transmitting the generated facial image and customized response scenario to the user terminal, and means for receiving, analyzing, and generating an optimal response to inquiries via the user terminal. This makes it possible to provide efficient and cost-effective dedicated customer support to many users and improve customer satisfaction.

[0577] "Means for collecting user contract information" refers to methods or systems for obtaining data related to services contracted by users, and includes means of collecting information from a database that includes contract details, contract period, personal attributes, past inquiries, etc.

[0578] "Means for analyzing collected contract information" refers to methods and systems for analyzing collected user contract data and extracting specific patterns or trends, and includes data analysis algorithms such as clustering and categorization.

[0579] "Means for generating friendly facial images based on analysis results" refers to methods or systems for generating different friendly facial images for each user based on the results of data analysis, and these methods utilize generative adversarial networks (GANs).

[0580] "Means for preparing individual customized response scenarios based on analysis results" refers to methods or systems for preparing different customized response scenarios for each user using the results of data analysis, and includes means for generating response templates based on a pre-prepared FAQ database or past inquiry content.

[0581] "Means for sending generated facial images and customized response scenarios to the user's terminal" refers to a method or system for sending generated facial images and customized response scenarios to the user's terminal, and is a means that uses HTTP requests.

[0582] "Means for displaying face images and customized scenarios generated on a user terminal" refers to methods or systems for displaying face images and corresponding scenarios sent to a user terminal on a user interface.

[0583] "Means for receiving, analyzing, and generating optimal responses via user terminals" refers to methods or systems for receiving user inquiries via user terminals, analyzing their content, and generating optimal responses, and which utilize natural language processing (NLP) algorithms.

[0584] "Means for collecting user feedback and improving analysis results" refers to methods and systems for collecting and storing user feedback information and using it to improve the quality of analysis results and corresponding scenarios.

[0585] This invention relates to a system that provides mobile service providers with personalized, dedicated customer support for a large user base. This system consists of multiple means and processing steps to improve customer satisfaction while keeping costs down. Each means and its associated processing steps are as follows:

[0586] First, the server collects user contract information. This information includes contract details, contract period, personal attributes, and past inquiry history. This data is retrieved from the database using SQL queries. For example, the server connects to the user database and collects data by executing the SQL query "SELECT FROM user_contracts WHERE user_id = [User ID]".

[0587] Next, the server uses K-means clustering and categorization algorithms to analyze the collected contract information. For example, it performs clustering using K-means(n_clusters=5).fit(contract information) and extracts features to identify the needs and inquiry trends of the contract holders.

[0588] Based on the analysis results, the server generates face images using a GAN (Generative Adversarial Network). In this process, it selects a GAN model with parameters optimized for a specific user category and generates friendly face images using prompts. For example, if you select StyleGAN2 as the GAN model to use and input prompts such as "Attributes for the newly generated face image: friendly, smiling, adult male," the server will generate the image. The generated face image is then temporarily stored in the server's storage.

[0589] Furthermore, the server prepares individual customized response scenarios based on the analysis results. To do this, it extracts relevant questions using a pre-prepared FAQ database and past inquiries, and generates individual answer templates. For example, it executes an SQL query such as "SELECT faq_answer FROM FAQ WHERE question LIKE '%contract renewal%'" to customize the response scenario.

[0590] The generated facial image and customized response scenario are sent from the server to the user's terminal. The HTTP protocol is used for this transmission. For example, the data is sent in the format "POST / api / send_userdata HTTP / 1.1".

[0591] The user terminal stores the received facial image and corresponding scenario in memory and prepares to display them on the user interface. Specifically, the facial image and text are appropriately positioned.

[0592] When a user submits an inquiry, the terminal receives the inquiry details. The received content is sent to the server using an HTTP request such as "POST / api / send_query HTTP / 1.1". The server analyzes the received inquiry using a natural language processing (NLP) algorithm and generates the optimal response. One example of an NLP model used is GPT-3. Based on the inquiry, the server generates the optimal response to a question such as "Please tell me about contract renewal." The generated response is displayed on the user's terminal, along with a photo of the representative's face.

[0593] Once the interaction is complete, the user's terminal displays a feedback interface. This feedback includes a form and a simple rating button. When the user enters their feedback and presses the submit button, the server receives the feedback and saves it to the database. For example, an SQL query such as "INSERT INTO user_feedback (user_id, rating, comments) VALUES ([User ID], 5, 'I was satisfied with the interaction.')" is executed. The server analyzes the collected feedback and uses it to improve future interactions.

[0594] Example of a prompt

[0595] "Please generate a friendly, dedicated representative profile picture and support scenarios for new mobile service subscribers. Please also refer to their contract details and past inquiries."

[0596] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0597] Step 1:

[0598] The server collects user contract information from the database.

[0599] Input: User ID

[0600] Process: Execute an SQL query against the database and retrieve the corresponding row data. Example: "SELECT FROM user_contracts WHERE user_id = [User ID]"

[0601] Output: User contract information including contract details, contract period, personal attributes, and past inquiries.

[0602] Step 2:

[0603] The server analyzes the contract information it has collected.

[0604] Input: Collected contract information

[0605] Processing: The contract information is divided into clusters using the K-means clustering algorithm. Example: "K-means(n_clusters=5).fit(contract information)"

[0606] Output: Classification results and features for each user cluster

[0607] Step 3:

[0608] The server generates a friendly-looking facial image based on the analysis results.

[0609] Input: User classification results and features for each cluster

[0610] Processing: Generate face images using a GAN (Generative Adversarial Network) model. Example: "GAN model to use: StyleGAN2" "Attributes of the newly generated face image: Friendly, smiling, adult male"

[0611] Output: Generated facial image

[0612] Step 4:

[0613] The server prepares individual customized scenarios based on the analysis results.

[0614] Input: Cluster classification results, past inquiry details

[0615] Processing: Extract relevant questions from the FAQ database and generate individual answer templates. Example: "SELECT faq_answer FROM FAQ WHERE question LIKE '%contract renewal%'"

[0616] Output: Customized response scenarios and answer templates

[0617] Step 5:

[0618] The server sends the generated facial image and customized scenario to the user's terminal.

[0619] Input: Generated facial image and customizable scenario

[0620] Processing: Sends data to the user's terminal using the HTTP protocol. Example: "POST / api / send_userdata HTTP / 1.1"

[0621] Output: The transmitted data is stored on the user's terminal.

[0622] Step 6:

[0623] The user terminal displays the generated facial image and the customized scenario.

[0624] Input: Received facial image and corresponding scenario

[0625] Processing: Place and display face images and scenarios on the user interface.

[0626] Output: Face image and corresponding scenario displayed to the user

[0627] Step 7:

[0628] When a user submits an inquiry, the user's device receives the inquiry details and sends them to the server.

[0629] Input: User inquiry details

[0630] Processing: Receives the query content and sends it to the server as an HTTP request. Example: "POST / api / send_query HTTP / 1.1"

[0631] Output: Sent inquiry content

[0632] Step 8:

[0633] The server analyzes the inquiry and generates the most appropriate response.

[0634] Input: User inquiry details

[0635] Processing: Analyze the query using a natural language processing (NLP) algorithm and generate the optimal response. NLP model used: GPT-3

[0636] Output: Generated answer

[0637] Step 9:

[0638] The user's terminal displays the generated response to the user, along with the face image of the person in charge.

[0639] Input: Generated response and facial image

[0640] Processing: Display the answer on the user interface, along with the facial image.

[0641] Output: Responses and facial images displayed to the user

[0642] Step 10:

[0643] Once the issue is resolved, the user's device will display a feedback interface.

[0644] Input: Trigger for completion of response

[0645] Processing: Display a feedback form and a simple rating button on the interface.

[0646] Output: User-inputtable feedback interface

[0647] Step 11:

[0648] When a user enters feedback and presses the submit button, the server receives the feedback and saves it to the database.

[0649] Input: User feedback content

[0650] Process: Receive the feedback and save it to the database. Example: "INSERT INTO user_feedback (user_id, rating, comments) VALUES ([User ID], 5, 'I was satisfied with the service.')"

[0651] Output: Saved feedback data

[0652] Step 12:

[0653] The server will analyze the collected feedback and use it to improve future responses.

[0654] Input: Saved feedback data

[0655] Processing: Analyze feedback data and use it to improve the service.

[0656] Output: Insights for improved response scenarios and system updates

[0657] (Application Example 1)

[0658] 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 glasses 214 will be referred to as the "terminal."

[0659] In food service provision, a challenge is responding quickly and individually to customers' specific requests and inquiries. Traditional customer support systems often provide uniform responses, making it difficult to improve customer satisfaction and provide efficient support. In particular, in food service, responses must be based on customer preferences and past order history, and traditional systems often lacked flexibility and personalization.

[0660] 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.

[0661] In this invention, the server includes means for collecting user contract information, means for analyzing the collected contract information, means for preparing individual customized response scenarios for each user based on the analysis results, means for generating a user-friendly facial image for each user, means for transmitting the generated facial image and response scenario to the user terminal, means for displaying the generated facial image and response scenario on the user terminal, and means for functioning as a dedicated support person for food services via a smart device. This enables prompt and individual responses to customers' special requests and inquiries, improving customer satisfaction and providing efficient support.

[0662] "User contract information" refers to information such as the contract details, contract period, personal attributes, and past inquiries of food service users.

[0663] "Means of collection" refers to the technical means of obtaining user contract information from a database.

[0664] "Means of analysis" refers to technical means of analyzing collected information using data analysis algorithms such as clustering and categorization.

[0665] "Individualized customized response scenarios" refer to specific response plans or scenarios prepared for individual users based on the analysis results.

[0666] A "friendly face image" refers to a customized face image that users can find appealing.

[0667] "Generative means" refers to technical methods for generating friendly facial images using GANs (Generative Adversarial Networks).

[0668] "Means of transmission" refers to the technical means for transmitting the generated facial image and corresponding scenario to the user's terminal.

[0669] "Means of display" refers to the technical means for displaying face images and corresponding scenarios generated on the user's terminal.

[0670] A "smart device" refers to a portable information terminal with internet connectivity, such as a smartphone or tablet.

[0671] A "dedicated support representative" refers to a virtual support representative who is assigned to each individual user.

[0672] This invention relates to a dedicated support system for users in the food service industry. This system is configured to provide food service providers with personalized, dedicated customer support to a large user base and is implemented through the following elements.

[0673] The server first collects user contract information. This contract information includes the user's contract details, contract period, personal attributes, and past inquiry history. This data is retrieved from an SQL database using SQL queries.

[0674] Next, the server analyzes the collected contract information. Data analysis algorithms such as clustering and categorization are used for this analysis. Python's scikit-learn library is used for data analysis. Based on the analysis results, the needs and inquiry trends of the contract holders are identified.

[0675] Based on the analysis results, the server generates a user-friendly facial image for each user. Image generation algorithms such as GANs (Generative Adversarial Networks) are used to generate these facial images. The generated facial images are customized based on the user's basic information. Deep learning frameworks such as TensorFlow and PyTorch are used to implement the GANs.

[0676] The server prepares individual customized response scenarios based on the analysis results. This includes customizing the pre-prepared FAQ database and generating answer templates based on past inquiries. This improves the quality and efficiency of responses.

[0677] The prepared facial image and customized scenario are sent from the server to the user's terminal. HTTP requests are used for this transmission. The smart device (smartphone or tablet) prepares to display the transmitted facial image and corresponding scenario in the user interface.

[0678] When a user contacts customer support, the device receives the user's inquiry input and sends it to the server. The server analyzes the inquiry and generates the most appropriate response based on pre-prepared response templates. The device then displays the response to the user, along with a profile picture of the assigned support representative.

[0679] After the service is completed, the device displays an interface for collecting user feedback. This feedback includes a form and a simple rating button. The user enters their feedback, the server collects the feedback information, stores it in a database, and analyzes it to improve future services.

[0680] Specific example

[0681] For example, if user A has a habit of ordering a specific pizza every Friday, the server will use this information to create a dedicated support representative for user A and prepare support scenarios to address any special requests regarding the pizza. When user A requests support, they will be shown a message such as, "Hello, I am your dedicated support representative. Do you have any questions regarding your pizza order?"

[0682] Examples of input prompts for a generative AI model

[0683] "Based on User A's past order history, generate a facial image of a dedicated support representative and a customized scenario for ordering pizza."

[0684] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0685] Step 1:

[0686] The server collects user contract information.

[0687] Input: User ID

[0688] Operation: The server queries the SQL database to retrieve the user's contract details, contract period, personal attributes, and past query history.

[0689] Output: User contract information

[0690] Step 2:

[0691] The server analyzes the collected contract information.

[0692] Input: User contract information

[0693] Operation: The server analyzes contract information using data analysis algorithms such as clustering and categorization. Specifically, it uses the Python scikit-learn library to cluster the data.

[0694] Output: Analysis results (user needs and inquiry trends)

[0695] Step 3:

[0696] The server generates a user-friendly facial image based on the analysis results.

[0697] Input: Analysis results

[0698] Operation: The server uses a GAN (Generative Adversarial Network) to generate user-friendly facial images. TensorFlow and PyTorch are used as deep learning frameworks for this purpose.

[0699] Output: Friendly facial image

[0700] Step 4:

[0701] The server prepares individual customized response scenarios based on the analysis results.

[0702] Input: Analysis results

[0703] Operation: The server customizes a pre-prepared FAQ database and generates answer templates based on past inquiries.

[0704] Output: Customizable scenarios

[0705] Step 5:

[0706] The server sends the generated facial image and customized scenario to the user's terminal.

[0707] Input: Friendly facial image, customizable scenarios

[0708] Operation: The server uses an HTTP request to send the generated facial image and customizable scenario to the user's smart device.

[0709] Output: Face image and corresponding scenario sent to the user's terminal

[0710] Step 6:

[0711] The user's terminal displays the generated facial image and the customizable scenario.

[0712] Input: Face images sent from the server and customizable scenarios.

[0713] Operation: The user terminal displays the transmitted facial image and customizable scenario on the user interface.

[0714] Output: Displayed facial image and corresponding scenario

[0715] Step 7:

[0716] The user contacts customer support.

[0717] Input: User inquiry

[0718] Operation: The user contacts customer support via a smart device. The device sends this inquiry to the server.

[0719] Output: User queries sent to the server

[0720] Step 8:

[0721] The server analyzes the user's inquiry and generates the most appropriate response.

[0722] Input: User inquiry

[0723] Operation: The server analyzes the query text and generates the most suitable response based on pre-prepared response templates.

[0724] Output: Best Answer

[0725] Step 9:

[0726] The device displays the answer to the user, along with a generated image of the person in charge's face.

[0727] Input: Best response sent from the server, face image of the person in charge

[0728] Operation: The device displays the answer and the face image of the person in charge on the screen.

[0729] Output: Displayed response and the respondent's face image

[0730] Step 10:

[0731] The user enters feedback, and the server collects the feedback information and stores it in a database.

[0732] Input: User feedback

[0733] Operation: The terminal displays a feedback collection interface, and the user enters feedback. The entered feedback is sent to the server and stored in the database.

[0734] Output: Saved feedback information

[0735] Examples of input prompts for a generative AI model

[0736] "Based on User A's past order history, generate a facial image of a dedicated support representative and a customized scenario for ordering pizza."

[0737] 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.

[0738] This invention aims to further improve customer satisfaction by combining a system that provides mobile service providers with personalized, dedicated customer support for a large user base with an emotion engine that recognizes user emotions. This system consists of the means and processing steps described below.

[0739] 1. Data Collection

[0740] The server retrieves user contract information from the database. This contract information includes contract details, contract period, personal attributes, and past inquiry history. This data is retrieved from the database using SQL queries.

[0741] 2. Data Analysis

[0742] The server analyzes the collected contract information. It uses data analysis algorithms (e.g., clustering and categorization) to analyze user needs and inquiry trends. Based on the analysis results, it determines individual response strategies.

[0743] 3. Image generation for the person in charge

[0744] The server generates a personalized facial image for each user based on the analysis results. Image generation algorithms such as GANs (Generative Adversarial Networks) are used to generate these facial images. The generated facial images are customized based on the user's basic information (e.g., age, gender, region).

[0745] 4. Preparation for individualized support

[0746] The server prepares individual customized response scenarios based on the analysis results. These scenarios include customizing a pre-prepared FAQ database and generating answer templates based on the user's past inquiries.

[0747] 5. Interface preparation

[0748] The server sends the generated facial image and customized scenario to the user's device. An HTTP request is used for this transmission. The device then prepares an interface to display the transmitted facial image and corresponding scenario.

[0749] 6. Emotion recognition

[0750] The device uses an emotion engine to recognize emotions from the user's facial expressions and voice. The emotion engine identifies the emotions the user is feeling in real time through facial expression recognition and voice analysis.

[0751] 7. Start of response

[0752] A user contacts customer support. The device receives this inquiry and sends the inquiry details to the server.

[0753] 8. Emotion-based dynamic adjustment

[0754] The server analyzes the user's inquiry and sentiment data sent from the device. Based on the analysis results, it dynamically adjusts customized response scenarios and answer templates to generate the optimal response that corresponds to the user's current sentiment.

[0755] 9. Display the answer

[0756] The terminal displays the response received from the server to the user. At the same time, a generated image of the person in charge is also displayed, and the output of voice and text is adjusted according to the user's emotions.

[0757] 10. Feedback on the response

[0758] The device displays an interface for collecting user feedback, which includes a feedback form and a quick rating button.

[0759] 11. Gathering and analyzing feedback

[0760] Users provide feedback on the service they received, including evaluations and opinions about the service. The server collects this user feedback and stores it in a database. The collected feedback is then analyzed and used as data to improve the quality of the service.

[0761] Specific example

[0762] For new customers

[0763] 1. The server retrieves and analyzes information about newly contracted users.

[0764] 2. The server uses a GAN to generate a facial image of the user's dedicated representative.

[0765] 3. The server sends the generated facial image and customized scenario to the terminal, which then displays it.

[0766] 4. The device analyzes the user's facial expressions and voice using an emotion engine to recognize the user's emotions.

[0767] 5. When a user makes an inquiry, the device displays "Hello, I am your personal support representative. How can I help you today?"

[0768] Measures to prevent cancellations

[0769] 1. The server retrieves and analyzes data from users who show signs of canceling their subscription.

[0770] 2. The server determines a response scenario to prevent cancellation and generates a facial image using a GAN.

[0771] 3. The server sends the generated image and corresponding scenario to the terminal, and the terminal displays it.

[0772] 4. The device analyzes the user's facial expressions and voice using an emotion engine to recognize the user's emotions.

[0773] 5. When a user inquires about canceling their service, the device will display a message saying, "Hello, this is your dedicated representative. We understand you are considering canceling your service. May we hear about any points of dissatisfaction?" and will respond appropriately based on the user's feelings.

[0774] In this way, the present invention provides personalized customer support tailored to individual needs based on the user's emotions, further improving customer satisfaction.

[0775] The following describes the processing flow.

[0776] Step 1:

[0777] The server retrieves user contract information from the database. This contract information includes contract details, contract period, personal attributes, and past inquiry history. This data is retrieved from the database using SQL queries.

[0778] Step 2:

[0779] The server analyzes the collected contract information. It uses data analysis algorithms such as clustering and categorization to analyze user needs and inquiry trends. Based on the analysis results, it determines individual response strategies for each user.

[0780] Step 3:

[0781] The server generates a user-friendly facial image based on the analysis results. It uses image generation algorithms such as GAN (Generative Adversarial Network) to customize the facial image based on the user's basic information (age, gender, region, etc.).

[0782] Step 4:

[0783] The server prepares individual customized scenarios. It customizes a pre-configured FAQ database and generates answer templates based on the user's past inquiries.

[0784] Step 5:

[0785] The server sends the generated facial image and customized scenario to the user's terminal. The facial image and scenario data are sent to the terminal using an HTTP request.

[0786] Step 6:

[0787] The device displays the generated facial image and a customizable scenario. At this point, the user interface is prepared, and the scenario and facial image are visually displayed to the user.

[0788] Step 7:

[0789] The device analyzes the user's facial expressions and voice, and recognizes emotions using an emotion engine. The emotion engine utilizes facial recognition and voice analysis algorithms to identify the user's emotions (e.g., joy, anger, sadness, etc.) in real time.

[0790] Step 8:

[0791] A user contacts customer support. The device receives the user's inquiry and sends it to the server.

[0792] Step 9:

[0793] The server analyzes the user's inquiry and the sentiment data recognized by the sentiment engine. Based on the analysis results, it generates the optimal response corresponding to the user's emotions and dynamically adjusts the customized response scenario.

[0794] Step 10:

[0795] The server sends the generated response to the terminal. The response data is sent to the terminal as an HTTP response.

[0796] Step 11:

[0797] The terminal displays the response from the server to the user. Along with the generated image of the person in charge, it displays the response in voice or text format, adjusted to the user's emotions.

[0798] Step 12:

[0799] After the support is complete, the device will display an interface for collecting feedback from the user. It will provide a feedback form and a simple rating button to receive opinions and ratings from the user.

[0800] Step 13:

[0801] Users enter feedback. This feedback includes evaluations and opinions about the actions taken.

[0802] Step 14:

[0803] The server collects user feedback and stores it in a database. The collected feedback is analyzed and used as data to improve the quality of future responses.

[0804] (Example 2)

[0805] 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".

[0806] Traditional customer support systems can provide personalized support based on user contract information, but they lack the ability to recognize user emotions and dynamically adjust responses, limiting their potential for improving customer satisfaction. Furthermore, they lack mechanisms to analyze user feedback and improve responses, resulting in insufficient improvement in service quality.

[0807] 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.

[0808] In this invention, the server includes means for collecting user contract information, means for analyzing the collected contract information, means for preparing individual customized response scenarios for each user based on the analysis results, means for generating a user-friendly facial image for each user, means for transmitting the generated facial image and response scenario to the user terminal, means for collecting emotional data from the user, means for analyzing the user's emotional data and dynamically adjusting the customized response scenario, and means for displaying the generated facial image and dynamically adjusted response scenario on the user terminal. This enables individualized responses based on the user's emotions, thereby improving customer satisfaction.

[0809] "Contract information" refers to data such as the details of the contract a mobile service user has made, the contract period, personal attributes, and past inquiries.

[0810] A "database" is a system used to store and manage user contract information, inquiry details, and other data.

[0811] "Data analysis" is the process of analyzing user needs and inquiry trends based on collected contract information, and determining individual response strategies.

[0812] Clustering is a technique for grouping and classifying user data based on specific criteria.

[0813] A "face image" is an image of a human face that is generated to make the user feel a sense of familiarity.

[0814] GAN (Generative Adversarial Network) is a type of image generation algorithm that generates high-quality images by pitting two neural networks against each other.

[0815] A "customized scenario" is a scenario created for each user based on the analysis results, designed to address their specific needs.

[0816] An "emotion engine" is software that recognizes and analyzes emotions from a user's facial expressions and voice.

[0817] An "HTTP request" is a protocol used to send and receive data between a server and a terminal.

[0818] "Dynamic adjustment" is a process that optimizes response scenarios in real time based on user emotion data.

[0819] "Feedback" refers to the evaluations and opinions that users provide regarding the response.

[0820] Modes for carrying out the invention

[0821] This invention aims to further improve customer satisfaction by combining a system that provides mobile service providers with personalized, dedicated customer support for large user bases with an emotion engine that recognizes user emotions. This system links a server and a terminal and provides a function to dynamically adjust individual responses using user contract information and emotion data.

[0822] Data collection

[0823] The server retrieves user contract information from the database. This contract information includes contract details, contract period, personal attributes, and past inquiry history. This data is retrieved from the database using SQL queries. For example, an SQL query such as "SELECT FROM user_contracts WHERE user_id = '<user ID>'" is used.

[0824] Data Analysis

[0825] The server analyzes the collected contract information. It uses data analysis algorithms (e.g., clustering and categorization) to analyze user needs and inquiry trends. For clustering, for example, the KMeans algorithm is used.

[0826] Image generation for the person in charge

[0827] The server generates a personalized facial image for each user based on the analysis results. This facial image generation uses a Generative Adversarial Network (GAN). For example, a GAN model can be used to generate a customized facial image based on the user's age, gender, and location.

[0828] Preparation for individual support

[0829] The server prepares individual customized response scenarios based on the analysis results. These scenarios include customizing a pre-prepared FAQ database and generating answer templates based on the user's past inquiries. Appropriate content is extracted from the FAQ database and the answer templates are customized.

[0830] Interface preparation

[0831] The server sends the generated facial image and customizable scenario to the user's device. This communication uses HTTP requests. Based on the transmitted data, the device prepares a display interface and presents it to the user.

[0832] emotion recognition

[0833] The device uses an emotion engine to recognize emotions from the user's facial expressions and voice. This emotion engine identifies the emotions the user is feeling in real time through facial expression recognition and voice analysis. For example, the EmotionEngine is used to analyze the user's emotions.

[0834] Start of response

[0835] A user contacts customer support. The device receives this inquiry and sends the inquiry details to the server. For example, the user types "I can't connect to the internet" and sends it from the device to the server.

[0836] Emotion-based dynamic adjustment

[0837] The server analyzes the user's inquiry and sentiment data sent from the device. Based on the analysis results, it dynamically adjusts customized response scenarios and answer templates to generate the optimal response that corresponds to the user's current sentiment. It generates a dynamically adjusted response based on sentiment data and the inquiry content.

[0838] Display the answer

[0839] The terminal displays the response received from the server to the user. At the same time, a generated image of the support representative's face is also displayed, and the output of voice and text is adjusted according to the user's emotions. For example, it might display, "Hello, I'm your personal support representative. How can I help you today?"

[0840] Feedback on the response

[0841] The device displays an interface for collecting user feedback. This includes a feedback form and a quick rating button. It receives user feedback and sends it to the server.

[0842] Feedback collection and analysis

[0843] The server collects user feedback and stores it in a database. The collected feedback is then analyzed and used as data to improve the quality of service. For example, the feedback can be analyzed to improve the service.

[0844] Specific example

[0845] For new customers

[0846] 1. The server retrieves and analyzes information about newly contracted users.

[0847] 2. The server uses a GAN to generate a facial image of the user's dedicated representative.

[0848] 3. The server sends the generated facial image and customized scenario to the terminal, which then displays it.

[0849] 4. The device analyzes the user's facial expressions and voice using an emotion engine to recognize the user's emotions.

[0850] 5. When a user makes an inquiry, the device displays "Hello, I am your personal support representative. How can I help you today?"

[0851] Measures to prevent cancellations

[0852] 1. The server retrieves and analyzes data from users who show signs of canceling their subscription.

[0853] 2. The server determines a response scenario to prevent cancellation and generates a facial image using a GAN.

[0854] 3. The server sends the generated image and corresponding scenario to the terminal, and the terminal displays it.

[0855] 4. The device analyzes the user's facial expressions and voice using an emotion engine to recognize the user's emotions.

[0856] 5. When a user inquires about canceling their service, the device will display a message saying, "Hello, this is your dedicated representative. We understand you are considering canceling your service. May we hear about any points of dissatisfaction?" and will respond appropriately based on the user's feelings.

[0857] Examples of prompts include, "Create customer support scenarios tailored to the emotions of newly signed-up users," and "Generate response scenarios that include emotion recognition for users considering cancellation."

[0858] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0859] Step 1:

[0860] The server retrieves the user's contract information from the database.

[0861] Specifically, the server connects to the database and executes an SQL query to extract the user's contract information.

[0862] Input: User ID

[0863] Output: User contract information (contract details, contract period, personal attributes, past inquiries)

[0864] The server executes the SQL query "SELECT FROM user_contracts WHERE user_id = '<User ID>'" to retrieve relevant information from the database.

[0865] Step 2:

[0866] The server analyzes the collected contract information.

[0867] In terms of specific operations, the server performs analysis using data analysis algorithms (clustering and categorization). Python's KMeans algorithm, for example, can be used.

[0868] Input: User's contract information

[0869] Output: Analysis results (user needs and inquiry trends)

[0870] Based on the contract information collected by the server, users are classified using the KMeans algorithm, and analysis results are obtained.

[0871] Step 3:

[0872] The server generates a user-friendly facial image based on the analysis results.

[0873] Specifically, the server uses a GAN to generate a facial image based on the user's basic information.

[0874] Input: Analysis results, user's basic information (age, gender, region)

[0875] Output: Generated facial image

[0876] The server uses a GAN model to generate facial images based on, for example, the user's age and gender.

[0877] Step 4:

[0878] The server prepares individual customized response scenarios based on the analysis results.

[0879] Specifically, the server customizes a pre-prepared FAQ database and generates answer templates based on past inquiries.

[0880] Input: Analysis results, FAQ database, past inquiry content

[0881] Output: Customizable scenarios, response templates

[0882] The server creates customized response scenarios and answer templates based on the FAQ database and past inquiries.

[0883] Step 5:

[0884] The server sends the generated facial image and customized scenario to the user's device.

[0885] In terms of specific operations, the server sends data to the terminal using an HTTP request.

[0886] Input: Generated facial image, customizable scenario

[0887] Output: Transmission status

[0888] The server uses an HTTP request to send the generated facial image and corresponding scenario to the terminal.

[0889] Step 6:

[0890] The device uses an emotion engine to recognize emotions from the user's facial expressions and voice.

[0891] Specifically, the device performs facial recognition and voice analysis, and identifies emotions through an emotion engine.

[0892] Input: User facial expression data, voice data

[0893] Output: User sentiment data

[0894] The device uses EmotionEngine to analyze, for example, the user's facial expressions and voice in real time, and obtain emotional data.

[0895] Step 7:

[0896] The user contacts customer support.

[0897] In terms of specific actions, the user enters their question using a contact form on their device.

[0898] Input: User's inquiry

[0899] Output: Query data

[0900] The user enters their inquiry into the device, and that information is sent to the device.

[0901] Step 8:

[0902] The terminal receives this query and sends the query details to the server.

[0903] Specifically, the terminal sends an HTTP request to forward the inquiry details to the server.

[0904] Input: Inquiry data

[0905] Output: Transmission status

[0906] The device sends an HTTP request containing the inquiry details to the server.

[0907] Step 9:

[0908] The server analyzes the user's inquiry and sentiment data sent from the device.

[0909] Specifically, the server analyzes sentiment data and query content, and dynamically adjusts the customized response scenario.

[0910] Input: Inquiry details, sentiment data

[0911] Output: Dynamically adjusted response scenarios, response templates

[0912] The server integrates inquiry content and sentiment data to generate optimized response scenarios and answers.

[0913] Step 10:

[0914] The terminal displays the response received from the server to the user.

[0915] In terms of specific operation, the terminal displays the received response and the face image of the person in charge on the display interface.

[0916] Input: Response content from the server, generated facial image

[0917] Output: Content displayed to the user

[0918] Based on the information received by the terminal from the server, the response content and facial image are displayed on the screen.

[0919] Step 11:

[0920] The device displays an interface for collecting feedback from the user.

[0921] In terms of specific actions, the device will present the user with a feedback form and rating buttons.

[0922] Input: None

[0923] Output: Feedback input interface

[0924] The device displays a form or button for collecting feedback.

[0925] Step 12:

[0926] Users provide feedback on the response.

[0927] In terms of specific actions, users enter their ratings and opinions into a feedback form and submit it.

[0928] Input: User feedback content

[0929] Output: Feedback data

[0930] The user enters feedback and sends it to their device.

[0931] Step 13:

[0932] The server collects user feedback and stores it in a database.

[0933] Specifically, the server adds the received feedback data to the database.

[0934] Input: User feedback content

[0935] Output: Saved status

[0936] The server receives the feedback data and saves it to the database.

[0937] Step 14:

[0938] The server analyzes the collected feedback and uses it as data to improve the quality of service.

[0939] Specifically, the server analyzes feedback data to gain insights that can be used to improve the quality of the service.

[0940] Input: Feedback data

[0941] Output: Improvement suggestions, analysis results

[0942] The server analyzes the feedback data and creates specific suggestions for service improvement.

[0943] (Application Example 2)

[0944] 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."

[0945] Traditional content delivery services have not adequately addressed individual user needs and emotions, making it difficult to improve user satisfaction. In particular, the lack of real-time content recommendations based on emotional changes during viewing, and the absence of dynamic interface adjustments, sometimes led to user stress and decreased satisfaction.

[0946] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for collecting user contract information, means for analyzing the collected contract information, means for preparing individual customized response scenarios for each user based on the analysis results, means for generating a user-friendly face image for each user, means for transmitting the generated face image and response scenario to the user terminal, means for displaying the generated face image and response scenario on the user terminal, means for collecting and analyzing user emotion data, means for dynamically recommending content based on the user's emotions, means for dynamically adjusting the interface according to the user's emotions, and means for collecting and improving feedback from the user. This enables real-time personalized content recommendation and interface adjustment in accordance with the user's emotions and needs, thereby improving user satisfaction.

[0947] "User contract information" refers to information such as the user's contract details, contract period, personal attributes, and past inquiries.

[0948] "Analysis results" refer to conclusions and insights derived by the server based on the collected data.

[0949] A "customized response scenario" refers to a pre-prepared plan and flow of individual responses based on user needs and inquiry trends.

[0950] A "face image" is a friendly, customized image of a face created using a Generative Adversarial Network (GAN).

[0951] A "user terminal" refers to a device used by a user, such as a smartphone, tablet, or personal computer.

[0952] "Emotional data" refers to information about a user's emotions, obtained in real time from their facial expressions and voice.

[0953] "A means of dynamically recommending content" refers to a technology that recommends appropriate content in real time based on user sentiment data and viewing history.

[0954] "Methods for dynamically adjusting the interface" refer to techniques that change the design and layout of an application's user interface (UI) in response to the user's emotions.

[0955] "Methods for collecting feedback and making improvements" refers to technologies that collect evaluations and opinions from users and use that information to perform data analysis and implement improvement measures to enhance future services.

[0956] This invention is a system for delivering dynamic content based on user emotions. First, a server collects user contract information and analyzes it. The analysis uses data mining techniques such as clustering and categorization. Next, based on the analysis results, a customized response scenario is prepared for each user, and a friendly facial image is generated. A Generative Adversarial Network (GAN) is used to generate this facial image. The generated facial image and response scenario are sent to the user's terminal.

[0957] On the user's terminal, the generated facial image and corresponding scenario are displayed, and the user's emotional data is collected in real time. Facial expression recognition and voice analysis technologies using a camera and microphone are employed to collect emotional data. Based on the emotional analysis results, the server dynamically recommends specific content. For example, a user experiencing stress might be recommended relaxing music or a comedy movie.

[0958] Furthermore, the application's user interface (UI) is dynamically adjusted according to the user's emotions. This means, for example, changing the UI design to a simpler and calmer one if the user is feeling stressed.

[0959] In addition, user feedback is collected and incorporated into subsequent analyses and scenario generation. This ensures continuous improvement of the service.

[0960] As a concrete example, when a user opens the application while feeling stressed, the server recommends a relaxing music playlist. The UI also automatically changes to calming colors to help reduce user stress. Afterwards, the user enters feedback on the content they viewed into a form within the application, and this data is sent to the server and used to improve the recommendation algorithm for the next time.

[0961] The program to realize this invention will be implemented using a general-purpose programming language such as Python. The libraries used will be "some_emotion_recognition_library" for emotion recognition and "some_content_recommendation_library" for content recommendation. Libraries such as "requests" and "SQLAlchemy" will be used for HTTP requests and database access.

[0962] Examples of prompt statements to input into the generative AI model are as follows:

[0963] "Design an algorithm that recommends relaxing music and movies based on the user's viewing history data and real-time sentiment analysis data. This algorithm should also consider past rating data to recommend content that best suits the user's current mood in real time. Additionally, the UI theme should dynamically change according to the user's mood."

[0964] In this way, a system is realized that dynamically recommends content based on user emotions, adjusts the interface, and incorporates feedback.

[0965] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0966] Step 1:

[0967] The server collects user contract information from the database. Input is identification information such as the user ID, and output is user contract information (contract details, contract period, personal attributes, past inquiry history). Specifically, it executes SQL queries to extract the necessary data.

[0968] Step 2:

[0969] The server analyzes the collected contract information. The input is the contract information obtained in step 1, and the output is the analysis results regarding user needs and inquiry trends. Data analysis algorithms (e.g., clustering and categorization) are used to group the data and derive insights.

[0970] Step 3:

[0971] The server generates user-specific, customized response scenarios and friendly facial images based on the analysis results. The input is the analysis results from step 2, and the output is the customized scenario and facial image corresponding to the user. A Generative Adversarial Network (GAN) is used to generate facial images and customize pre-defined response scenarios.

[0972] Step 4:

[0973] The server sends the generated face image and corresponding scenario to the user terminal. The input is the face image and corresponding scenario generated in step 3, and the output is the face image and corresponding scenario sent to the user terminal. Data is sent using an HTTP request.

[0974] Step 5:

[0975] The user terminal displays the generated face image and a customizable scenario. The input is the face image and corresponding scenario sent from the server, and the output is a visual display that the user can confirm on their terminal. The interface displays the face image and text.

[0976] Step 6:

[0977] The user terminal collects and analyzes the user's emotional data (facial expressions and voice) in real time. The input is the user's facial and voice data, and the output is the analysis results regarding the user's emotions. Specifically, it uses a camera and microphone and an emotion recognition engine to identify emotions.

[0978] Step 7:

[0979] The server dynamically recommends the most suitable content to the user based on sentiment data. The input is the sentiment data from step 6 and the user's viewing history, and the output is the recommended content. A content recommendation algorithm is used to select the content best suited to the user's current sentiment.

[0980] Step 8:

[0981] The user's device dynamically adjusts the interface according to the user's emotions. The input is the content and emotion data recommended in step 7, and the output is the UI adjusted according to the emotion. For example, a user feeling stressed will be shown a UI with calming colors.

[0982] Step 9:

[0983] Users view content and then provide feedback. The input is the user's feedback, and the output is feedback data. By filling out the feedback form on their device, the data is sent to the server.

[0984] Step 10:

[0985] The server analyzes the collected feedback data to improve response scenarios and recommendation algorithms. The input is the feedback collected in step 9, and the output is the improved response scenarios and recommendation algorithms. Analysis based on the feedback data improves system performance.

[0986] The above steps enable a dynamic content delivery system based on user emotions.

[0987] 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.

[0988] 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.

[0989] 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.

[0990] [Third Embodiment]

[0991] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0992] 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.

[0993] 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).

[0994] 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.

[0995] 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.

[0996] 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).

[0997] 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.

[0998] 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.

[0999] 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.

[1000] 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.

[1001] 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.

[1002] 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".

[1003] This invention relates to a system that provides mobile service providers with personalized, dedicated customer support for a large user base. To reduce costs while improving customer satisfaction, this system comprises the means and processing steps described below.

[1004] 1. Data Collection

[1005] The server collects user contract information. This contract information includes contract details, contract period, personal attributes, and past inquiry history. This data is retrieved from the database using SQL queries.

[1006] 2. Data Analysis

[1007] The server analyzes the collected contract information. Data analysis algorithms such as clustering and categorization are used for the analysis. Based on the analysis results, the needs and inquiry trends of the contract holders are identified.

[1008] 3. Image generation for the person in charge

[1009] The server generates a personalized facial image for each user based on the analysis results. Image generation algorithms such as GANs (Generative Adversarial Networks) are used to generate the facial images. The generated facial images are customized based on the user's basic information.

[1010] 4. Preparation for individualized support

[1011] The server prepares individual customized response scenarios based on the analysis results. This includes customizing a pre-prepared FAQ database and generating answer templates based on past inquiries.

[1012] 5. Interface preparation

[1013] The server sends the generated facial image and customized scenario to the user's terminal. An HTTP request is used for this transmission. The terminal then prepares to display the transmitted facial image and scenario in its user interface.

[1014] 6. Start of response

[1015] When a user contacts customer support, the device receives the user's inquiry input and sends it to the server. The server analyzes the inquiry and generates the most appropriate response based on pre-prepared response templates. The device then displays the response to the user, along with a profile picture of the assigned support representative.

[1016] 7. Feedback on the response

[1017] After the service is completed, the device displays an interface for collecting user feedback. This feedback includes a form and a simple rating button. The user enters their feedback, the server collects the feedback information, stores it in a database, and analyzes it to improve future services.

[1018] Specific example

[1019] For new customers

[1020] 1. The server retrieves and analyzes information about newly contracted users.

[1021] 2. The server uses a GAN to generate a facial image of the user's dedicated representative.

[1022] 3. The server sends the generated facial image and customized scenario to the terminal, which then displays it.

[1023] 4. When a user makes an inquiry, the device displays "Hello, I am your personal support representative. How can I help you today?"

[1024] Measures to prevent cancellations

[1025] 1. The server retrieves and analyzes data from users who show signs of canceling their subscription.

[1026] 2. The server determines a response scenario to prevent cancellation and generates a facial image using a GAN.

[1027] 3. The server sends the generated image and corresponding scenario to the terminal, and the terminal displays it.

[1028] 4. When a user inquires about canceling their service, the device displays the message, "Hello, this is your dedicated representative. We understand you are considering canceling your service. Could you tell us what your concerns are?" and asks for the reason for cancellation.

[1029] In this way, the present invention can provide dedicated customer support tailored to the individual needs of users, thereby improving customer satisfaction while keeping costs down.

[1030] The following describes the processing flow.

[1031] Step 1:

[1032] The server retrieves the user's contract information from the database. This contract information includes contract details, contract period, personal attributes, and past inquiry history. This information is retrieved from the database using SQL queries.

[1033] Step 2:

[1034] The server analyzes the collected contract information. It uses data analysis algorithms (e.g., clustering and categorization) to analyze user needs and inquiry trends. Based on the analysis results, it determines individual response strategies.

[1035] Step 3:

[1036] The server generates a personalized facial image for each user based on the analysis results. Image generation algorithms such as GANs (Generative Adversarial Networks) are used to generate these facial images. The generated facial images are customized based on the user's basic information (e.g., age, gender, region).

[1037] Step 4:

[1038] The server prepares individual customized response scenarios based on the analysis results. These scenarios include customizing a pre-prepared FAQ database and generating answer templates based on the user's past inquiries.

[1039] Step 5:

[1040] The server sends the generated facial image and customized scenario to the user's device. This transmission uses an HTTP request. The device prepares an interface to display the transmitted facial image and corresponding scenario.

[1041] Step 6:

[1042] A user contacts customer support. The device receives this inquiry and sends the inquiry details to the server.

[1043] Step 7:

[1044] The server analyzes the user's inquiry. Based on pre-prepared response templates, it generates the most appropriate response and sends it to the terminal.

[1045] Step 8:

[1046] The terminal displays the response received from the server to the user. At the same time, a generated image of the person in charge's face is also displayed.

[1047] Step 9:

[1048] After the support is complete, the device will display an interface for collecting user feedback. This includes a feedback form and a simple rating button.

[1049] Step 10:

[1050] Users provide feedback on the response they receive. This includes evaluations and opinions about the response.

[1051] Step 11:

[1052] The server collects user feedback and stores it in a database. The collected feedback is then analyzed and used as data to improve the quality of service.

[1053] (Example 1)

[1054] 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."

[1055] Mobile service providers are required to provide individual, dedicated customer support to a large number of users, but effectively and cost-effectively serving such a large user base is difficult. Traditional systems suffer from problems such as inconsistent quality of support, long response times, and excessive resource requirements.

[1056] 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.

[1057] In this invention, the server includes means for collecting user contract information, means for analyzing the collected contract information, means for generating a user-friendly facial image based on the analysis results, means for preparing individual customized response scenarios based on the analysis results, means for transmitting the generated facial image and customized response scenario to the user terminal, and means for receiving, analyzing, and generating an optimal response to inquiries via the user terminal. This makes it possible to provide efficient and cost-effective dedicated customer support to many users and improve customer satisfaction.

[1058] "Means for collecting user contract information" refers to methods or systems for obtaining data related to services contracted by users, and includes means of collecting information from a database that includes contract details, contract period, personal attributes, past inquiries, etc.

[1059] "Means for analyzing collected contract information" refers to methods and systems for analyzing collected user contract data and extracting specific patterns or trends, and includes data analysis algorithms such as clustering and categorization.

[1060] "Means for generating friendly facial images based on analysis results" refers to methods or systems for generating different friendly facial images for each user based on the results of data analysis, and these methods utilize generative adversarial networks (GANs).

[1061] "Means for preparing individual customized response scenarios based on analysis results" refers to methods or systems for preparing different customized response scenarios for each user using the results of data analysis, and includes means for generating response templates based on a pre-prepared FAQ database or past inquiry content.

[1062] "Means for sending generated facial images and customized response scenarios to the user's terminal" refers to a method or system for sending generated facial images and customized response scenarios to the user's terminal, and is a means that uses HTTP requests.

[1063] "Means for displaying face images and customized scenarios generated on a user terminal" refers to methods or systems for displaying face images and corresponding scenarios sent to a user terminal on a user interface.

[1064] "Means for receiving, analyzing, and generating optimal responses via user terminals" refers to methods or systems for receiving user inquiries via user terminals, analyzing their content, and generating optimal responses, and which utilize natural language processing (NLP) algorithms.

[1065] "Means for collecting user feedback and improving analysis results" refers to methods and systems for collecting and storing user feedback information and using it to improve the quality of analysis results and corresponding scenarios.

[1066] This invention relates to a system that provides mobile service providers with personalized, dedicated customer support for a large user base. This system consists of multiple means and processing steps to improve customer satisfaction while keeping costs down. Each means and its associated processing steps are as follows:

[1067] First, the server collects user contract information. This information includes contract details, contract period, personal attributes, and past inquiry history. This data is retrieved from the database using SQL queries. For example, the server connects to the user database and collects data by executing the SQL query "SELECT FROM user_contracts WHERE user_id = [User ID]".

[1068] Next, the server uses K-means clustering and categorization algorithms to analyze the collected contract information. For example, it performs clustering using K-means(n_clusters=5).fit(contract information) and extracts features to identify the needs and inquiry trends of the contract holders.

[1069] Based on the analysis results, the server generates face images using a GAN (Generative Adversarial Network). In this process, it selects a GAN model with parameters optimized for a specific user category and generates friendly face images using prompts. For example, if you select StyleGAN2 as the GAN model to use and input prompts such as "Attributes for the newly generated face image: friendly, smiling, adult male," the server will generate the image. The generated face image is then temporarily stored in the server's storage.

[1070] Furthermore, the server prepares individual customized response scenarios based on the analysis results. To do this, it extracts relevant questions using a pre-prepared FAQ database and past inquiries, and generates individual answer templates. For example, it executes an SQL query such as "SELECT faq_answer FROM FAQ WHERE question LIKE '%contract renewal%'" to customize the response scenario.

[1071] The generated facial image and customized response scenario are sent from the server to the user's terminal. The HTTP protocol is used for this transmission. For example, the data is sent in the format "POST / api / send_userdata HTTP / 1.1".

[1072] The user terminal stores the received facial image and corresponding scenario in memory and prepares to display them on the user interface. Specifically, the facial image and text are appropriately positioned.

[1073] When a user submits an inquiry, the terminal receives the inquiry details. The received content is sent to the server using an HTTP request such as "POST / api / send_query HTTP / 1.1". The server analyzes the received inquiry using a natural language processing (NLP) algorithm and generates the optimal response. One example of an NLP model used is GPT-3. Based on the inquiry, the server generates the optimal response to a question such as "Please tell me about contract renewal." The generated response is displayed on the user's terminal, along with a photo of the representative's face.

[1074] Once the interaction is complete, the user's terminal displays a feedback interface. This feedback includes a form and a simple rating button. When the user enters their feedback and presses the submit button, the server receives the feedback and saves it to the database. For example, an SQL query such as "INSERT INTO user_feedback (user_id, rating, comments) VALUES ([User ID], 5, 'I was satisfied with the interaction.')" is executed. The server analyzes the collected feedback and uses it to improve future interactions.

[1075] Example of a prompt

[1076] "Please generate a friendly, dedicated representative profile picture and support scenarios for new mobile service subscribers. Please also refer to their contract details and past inquiries."

[1077] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1078] Step 1:

[1079] The server collects user contract information from the database.

[1080] Input: User ID

[1081] Process: Execute an SQL query against the database and retrieve the corresponding row data. Example: "SELECT FROM user_contracts WHERE user_id = [User ID]"

[1082] Output: User contract information including contract details, contract period, personal attributes, and past inquiries.

[1083] Step 2:

[1084] The server analyzes the contract information it has collected.

[1085] Input: Collected contract information

[1086] Processing: The contract information is divided into clusters using the K-means clustering algorithm. Example: "K-means(n_clusters=5).fit(contract information)"

[1087] Output: Classification results and features for each user cluster

[1088] Step 3:

[1089] The server generates a friendly-looking facial image based on the analysis results.

[1090] Input: User classification results and features for each cluster

[1091] Processing: Generate face images using a GAN (Generative Adversarial Network) model. Example: "GAN model to use: StyleGAN2" "Attributes of the newly generated face image: Friendly, smiling, adult male"

[1092] Output: Generated facial image

[1093] Step 4:

[1094] The server prepares individual customized scenarios based on the analysis results.

[1095] Input: Cluster classification results, past inquiry details

[1096] Processing: Extract relevant questions from the FAQ database and generate individual answer templates. Example: "SELECT faq_answer FROM FAQ WHERE question LIKE '%contract renewal%'"

[1097] Output: Customized response scenarios and answer templates

[1098] Step 5:

[1099] The server sends the generated facial image and customized scenario to the user's terminal.

[1100] Input: Generated facial image and customizable scenario

[1101] Processing: Sends data to the user's terminal using the HTTP protocol. Example: "POST / api / send_userdata HTTP / 1.1"

[1102] Output: The transmitted data is stored on the user's terminal.

[1103] Step 6:

[1104] The user terminal displays the generated facial image and the customized scenario.

[1105] Input: Received facial image and corresponding scenario

[1106] Processing: Place and display face images and scenarios on the user interface.

[1107] Output: Face image and corresponding scenario displayed to the user

[1108] Step 7:

[1109] When a user submits an inquiry, the user's device receives the inquiry details and sends them to the server.

[1110] Input: User inquiry details

[1111] Processing: Receives the query content and sends it to the server as an HTTP request. Example: "POST / api / send_query HTTP / 1.1"

[1112] Output: Sent inquiry content

[1113] Step 8:

[1114] The server analyzes the inquiry and generates the most appropriate response.

[1115] Input: User inquiry details

[1116] Processing: Analyze the query using a natural language processing (NLP) algorithm and generate the optimal response. NLP model used: GPT-3

[1117] Output: Generated answer

[1118] Step 9:

[1119] The user's terminal displays the generated response to the user, along with the face image of the person in charge.

[1120] Input: Generated response and facial image

[1121] Processing: Display the answer on the user interface, along with the facial image.

[1122] Output: Responses and facial images displayed to the user

[1123] Step 10:

[1124] Once the issue is resolved, the user's device will display a feedback interface.

[1125] Input: Trigger for completion of response

[1126] Processing: Display a feedback form and a simple rating button on the interface.

[1127] Output: User-inputtable feedback interface

[1128] Step 11:

[1129] When a user enters feedback and presses the submit button, the server receives the feedback and saves it to the database.

[1130] Input: User feedback content

[1131] Process: Receive the feedback and save it to the database. Example: "INSERT INTO user_feedback (user_id, rating, comments) VALUES ([User ID], 5, 'I was satisfied with the service.')"

[1132] Output: Saved feedback data

[1133] Step 12:

[1134] The server will analyze the collected feedback and use it to improve future responses.

[1135] Input: Saved feedback data

[1136] Processing: Analyze feedback data and use it to improve the service.

[1137] Output: Insights for improved response scenarios and system updates

[1138] (Application Example 1)

[1139] 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."

[1140] In food service provision, a challenge is responding quickly and individually to customers' specific requests and inquiries. Traditional customer support systems often provide uniform responses, making it difficult to improve customer satisfaction and provide efficient support. In particular, in food service, responses must be based on customer preferences and past order history, and traditional systems often lacked flexibility and personalization.

[1141] 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.

[1142] In this invention, the server includes means for collecting user contract information, means for analyzing the collected contract information, means for preparing individual customized response scenarios for each user based on the analysis results, means for generating a user-friendly facial image for each user, means for transmitting the generated facial image and response scenario to the user terminal, means for displaying the generated facial image and response scenario on the user terminal, and means for functioning as a dedicated support person for food services via a smart device. This enables prompt and individual responses to customers' special requests and inquiries, improving customer satisfaction and providing efficient support.

[1143] "User contract information" refers to information such as the contract details, contract period, personal attributes, and past inquiries of food service users.

[1144] "Means of collection" refers to the technical means of obtaining user contract information from a database.

[1145] "Means of analysis" refers to technical means of analyzing collected information using data analysis algorithms such as clustering and categorization.

[1146] "Individualized customized response scenarios" refer to specific response plans or scenarios prepared for individual users based on the analysis results.

[1147] A "friendly face image" refers to a customized face image that users can find appealing.

[1148] "Generative means" refers to technical methods for generating friendly facial images using GANs (Generative Adversarial Networks).

[1149] "Means of transmission" refers to the technical means for transmitting the generated facial image and corresponding scenario to the user's terminal.

[1150] "Means of display" refers to the technical means for displaying face images and corresponding scenarios generated on the user's terminal.

[1151] A "smart device" refers to a portable information terminal with internet connectivity, such as a smartphone or tablet.

[1152] A "dedicated support representative" refers to a virtual support representative who is assigned to each individual user.

[1153] This invention relates to a dedicated support system for users in the food service industry. This system is configured to provide food service providers with personalized, dedicated customer support to a large user base and is implemented through the following elements.

[1154] The server first collects user contract information. This contract information includes the user's contract details, contract period, personal attributes, and past inquiry history. This data is retrieved from an SQL database using SQL queries.

[1155] Next, the server analyzes the collected contract information. Data analysis algorithms such as clustering and categorization are used for this analysis. Python's scikit-learn library is used for data analysis. Based on the analysis results, the needs and inquiry trends of the contract holders are identified.

[1156] Based on the analysis results, the server generates a user-friendly facial image for each user. Image generation algorithms such as GANs (Generative Adversarial Networks) are used to generate these facial images. The generated facial images are customized based on the user's basic information. Deep learning frameworks such as TensorFlow and PyTorch are used to implement the GANs.

[1157] The server prepares individual customized response scenarios based on the analysis results. This includes customizing the pre-prepared FAQ database and generating answer templates based on past inquiries. This improves the quality and efficiency of responses.

[1158] The prepared facial image and customized scenario are sent from the server to the user's terminal. HTTP requests are used for this transmission. The smart device (smartphone or tablet) prepares to display the transmitted facial image and corresponding scenario in the user interface.

[1159] When a user contacts customer support, the device receives the user's inquiry input and sends it to the server. The server analyzes the inquiry and generates the most appropriate response based on pre-prepared response templates. The device then displays the response to the user, along with a profile picture of the assigned support representative.

[1160] After the service is completed, the device displays an interface for collecting user feedback. This feedback includes a form and a simple rating button. The user enters their feedback, the server collects the feedback information, stores it in a database, and analyzes it to improve future services.

[1161] Specific example

[1162] For example, if user A has a habit of ordering a specific pizza every Friday, the server will use this information to create a dedicated support representative for user A and prepare support scenarios to address any special requests regarding the pizza. When user A requests support, they will be shown a message such as, "Hello, I am your dedicated support representative. Do you have any questions regarding your pizza order?"

[1163] Examples of input prompts for a generative AI model

[1164] "Based on User A's past order history, generate a facial image of a dedicated support representative and a customized scenario for ordering pizza."

[1165] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1166] Step 1:

[1167] The server collects user contract information.

[1168] Input: User ID

[1169] Operation: The server queries the SQL database to retrieve the user's contract details, contract period, personal attributes, and past query history.

[1170] Output: User contract information

[1171] Step 2:

[1172] The server analyzes the collected contract information.

[1173] Input: User contract information

[1174] Operation: The server analyzes contract information using data analysis algorithms such as clustering and categorization. Specifically, it uses the Python scikit-learn library to cluster the data.

[1175] Output: Analysis results (user needs and inquiry trends)

[1176] Step 3:

[1177] The server generates a user-friendly facial image based on the analysis results.

[1178] Input: Analysis results

[1179] Operation: The server uses a GAN (Generative Adversarial Network) to generate user-friendly facial images. TensorFlow and PyTorch are used as deep learning frameworks for this purpose.

[1180] Output: Friendly facial image

[1181] Step 4:

[1182] The server prepares individual customized response scenarios based on the analysis results.

[1183] Input: Analysis results

[1184] Operation: The server customizes a pre-prepared FAQ database and generates answer templates based on past inquiries.

[1185] Output: Customizable scenarios

[1186] Step 5:

[1187] The server sends the generated facial image and customized scenario to the user's terminal.

[1188] Input: Friendly facial image, customizable scenarios

[1189] Operation: The server uses an HTTP request to send the generated facial image and customizable scenario to the user's smart device.

[1190] Output: Face image and corresponding scenario sent to the user's terminal

[1191] Step 6:

[1192] The user's terminal displays the generated facial image and the customizable scenario.

[1193] Input: Face images sent from the server and customizable scenarios.

[1194] Operation: The user terminal displays the transmitted facial image and customizable scenario on the user interface.

[1195] Output: Displayed facial image and corresponding scenario

[1196] Step 7:

[1197] The user contacts customer support.

[1198] Input: User inquiry

[1199] Operation: The user contacts customer support via a smart device. The device sends this inquiry to the server.

[1200] Output: User queries sent to the server

[1201] Step 8:

[1202] The server analyzes the user's inquiry and generates the most appropriate response.

[1203] Input: User inquiry

[1204] Operation: The server analyzes the query text and generates the most suitable response based on pre-prepared response templates.

[1205] Output: Best Answer

[1206] Step 9:

[1207] The device displays the answer to the user, along with a generated image of the person in charge's face.

[1208] Input: Best response sent from the server, face image of the person in charge

[1209] Operation: The device displays the answer and the face image of the person in charge on the screen.

[1210] Output: Displayed response and the respondent's face image

[1211] Step 10:

[1212] The user enters feedback, and the server collects the feedback information and stores it in a database.

[1213] Input: User feedback

[1214] Operation: The terminal displays a feedback collection interface, and the user enters feedback. The entered feedback is sent to the server and stored in the database.

[1215] Output: Saved feedback information

[1216] Examples of input prompts for a generative AI model

[1217] "Based on User A's past order history, generate a facial image of a dedicated support representative and a customized scenario for ordering pizza."

[1218] 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.

[1219] This invention aims to further improve customer satisfaction by combining a system that provides mobile service providers with personalized, dedicated customer support for a large user base with an emotion engine that recognizes user emotions. This system consists of the means and processing steps described below.

[1220] 1. Data Collection

[1221] The server retrieves user contract information from the database. This contract information includes contract details, contract period, personal attributes, and past inquiry history. This data is retrieved from the database using SQL queries.

[1222] 2. Data Analysis

[1223] The server analyzes the collected contract information. It uses data analysis algorithms (e.g., clustering and categorization) to analyze user needs and inquiry trends. Based on the analysis results, it determines individual response strategies.

[1224] 3. Image generation for the person in charge

[1225] The server generates a personalized facial image for each user based on the analysis results. Image generation algorithms such as GANs (Generative Adversarial Networks) are used to generate these facial images. The generated facial images are customized based on the user's basic information (e.g., age, gender, region).

[1226] 4. Preparation for individualized support

[1227] The server prepares individual customized response scenarios based on the analysis results. These scenarios include customizing a pre-prepared FAQ database and generating answer templates based on the user's past inquiries.

[1228] 5. Interface preparation

[1229] The server sends the generated facial image and customized scenario to the user's device. An HTTP request is used for this transmission. The device then prepares an interface to display the transmitted facial image and corresponding scenario.

[1230] 6. Emotion recognition

[1231] The device uses an emotion engine to recognize emotions from the user's facial expressions and voice. The emotion engine identifies the emotions the user is feeling in real time through facial expression recognition and voice analysis.

[1232] 7. Start of response

[1233] A user contacts customer support. The device receives this inquiry and sends the inquiry details to the server.

[1234] 8. Emotion-based dynamic adjustment

[1235] The server analyzes the user's inquiry and sentiment data sent from the device. Based on the analysis results, it dynamically adjusts customized response scenarios and answer templates to generate the optimal response that corresponds to the user's current sentiment.

[1236] 9. Display the answer

[1237] The terminal displays the response received from the server to the user. At the same time, a generated image of the person in charge is also displayed, and the output of voice and text is adjusted according to the user's emotions.

[1238] 10. Feedback on the response

[1239] The device displays an interface for collecting user feedback, which includes a feedback form and a quick rating button.

[1240] 11. Gathering and analyzing feedback

[1241] Users provide feedback on the service they received, including evaluations and opinions about the service. The server collects this user feedback and stores it in a database. The collected feedback is then analyzed and used as data to improve the quality of the service.

[1242] Specific example

[1243] For new customers

[1244] 1. The server retrieves and analyzes information about newly contracted users.

[1245] 2. The server uses a GAN to generate a facial image of the user's dedicated representative.

[1246] 3. The server sends the generated facial image and customized scenario to the terminal, which then displays it.

[1247] 4. The device analyzes the user's facial expressions and voice using an emotion engine to recognize the user's emotions.

[1248] 5. When a user makes an inquiry, the device displays "Hello, I am your personal support representative. How can I help you today?"

[1249] Measures to prevent cancellations

[1250] 1. The server retrieves and analyzes data from users who show signs of canceling their subscription.

[1251] 2. The server determines a response scenario to prevent cancellation and generates a facial image using a GAN.

[1252] 3. The server sends the generated image and corresponding scenario to the terminal, and the terminal displays it.

[1253] 4. The device analyzes the user's facial expressions and voice using an emotion engine to recognize the user's emotions.

[1254] 5. When a user inquires about canceling their service, the device will display a message saying, "Hello, this is your dedicated representative. We understand you are considering canceling your service. May we hear about any points of dissatisfaction?" and will respond appropriately based on the user's feelings.

[1255] In this way, the present invention provides personalized customer support tailored to individual needs based on the user's emotions, further improving customer satisfaction.

[1256] The following describes the processing flow.

[1257] Step 1:

[1258] The server retrieves user contract information from the database. This contract information includes contract details, contract period, personal attributes, and past inquiry history. This data is retrieved from the database using SQL queries.

[1259] Step 2:

[1260] The server analyzes the collected contract information. It uses data analysis algorithms such as clustering and categorization to analyze user needs and inquiry trends. Based on the analysis results, it determines individual response strategies for each user.

[1261] Step 3:

[1262] The server generates a user-friendly facial image based on the analysis results. It uses image generation algorithms such as GAN (Generative Adversarial Network) to customize the facial image based on the user's basic information (age, gender, region, etc.).

[1263] Step 4:

[1264] The server prepares individual customized scenarios. It customizes a pre-configured FAQ database and generates answer templates based on the user's past inquiries.

[1265] Step 5:

[1266] The server sends the generated facial image and customized scenario to the user's terminal. The facial image and scenario data are sent to the terminal using an HTTP request.

[1267] Step 6:

[1268] The device displays the generated facial image and a customizable scenario. At this point, the user interface is prepared, and the scenario and facial image are visually displayed to the user.

[1269] Step 7:

[1270] The device analyzes the user's facial expressions and voice, and recognizes emotions using an emotion engine. The emotion engine utilizes facial recognition and voice analysis algorithms to identify the user's emotions (e.g., joy, anger, sadness, etc.) in real time.

[1271] Step 8:

[1272] A user contacts customer support. The device receives the user's inquiry and sends it to the server.

[1273] Step 9:

[1274] The server analyzes the user's inquiry and the sentiment data recognized by the sentiment engine. Based on the analysis results, it generates the optimal response corresponding to the user's emotions and dynamically adjusts the customized response scenario.

[1275] Step 10:

[1276] The server sends the generated response to the terminal. The response data is sent to the terminal as an HTTP response.

[1277] Step 11:

[1278] The terminal displays the response from the server to the user. Along with the generated image of the person in charge, it displays the response in voice or text format, adjusted to the user's emotions.

[1279] Step 12:

[1280] After the support is complete, the device will display an interface for collecting feedback from the user. It will provide a feedback form and a simple rating button to receive opinions and ratings from the user.

[1281] Step 13:

[1282] Users enter feedback. This feedback includes evaluations and opinions about the actions taken.

[1283] Step 14:

[1284] The server collects user feedback and stores it in a database. The collected feedback is analyzed and used as data to improve the quality of future responses.

[1285] (Example 2)

[1286] 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."

[1287] Traditional customer support systems can provide personalized support based on user contract information, but they lack the ability to recognize user emotions and dynamically adjust responses, limiting their potential for improving customer satisfaction. Furthermore, they lack mechanisms to analyze user feedback and improve responses, resulting in insufficient improvement in service quality.

[1288] 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.

[1289] In this invention, the server includes means for collecting user contract information, means for analyzing the collected contract information, means for preparing individual customized response scenarios for each user based on the analysis results, means for generating a user-friendly facial image for each user, means for transmitting the generated facial image and response scenario to the user terminal, means for collecting emotional data from the user, means for analyzing the user's emotional data and dynamically adjusting the customized response scenario, and means for displaying the generated facial image and dynamically adjusted response scenario on the user terminal. This enables individualized responses based on the user's emotions, thereby improving customer satisfaction.

[1290] "Contract information" refers to data such as the details of the contract a mobile service user has made, the contract period, personal attributes, and past inquiries.

[1291] A "database" is a system used to store and manage user contract information, inquiry details, and other data.

[1292] "Data analysis" is the process of analyzing user needs and inquiry trends based on collected contract information, and determining individual response strategies.

[1293] Clustering is a technique for grouping and classifying user data based on specific criteria.

[1294] A "face image" is an image of a human face that is generated to make the user feel a sense of familiarity.

[1295] GAN (Generative Adversarial Network) is a type of image generation algorithm that generates high-quality images by pitting two neural networks against each other.

[1296] A "customized scenario" is a scenario created for each user based on the analysis results, designed to address their specific needs.

[1297] An "emotion engine" is software that recognizes and analyzes emotions from a user's facial expressions and voice.

[1298] An "HTTP request" is a protocol used to send and receive data between a server and a terminal.

[1299] "Dynamic adjustment" is a process that optimizes response scenarios in real time based on user emotion data.

[1300] "Feedback" refers to the evaluations and opinions that users provide regarding the response.

[1301] Modes for carrying out the invention

[1302] This invention aims to further improve customer satisfaction by combining a system that provides mobile service providers with personalized, dedicated customer support for large user bases with an emotion engine that recognizes user emotions. This system links a server and a terminal and provides a function to dynamically adjust individual responses using user contract information and emotion data.

[1303] Data collection

[1304] The server retrieves user contract information from the database. This contract information includes contract details, contract period, personal attributes, and past inquiry history. This data is retrieved from the database using SQL queries. For example, an SQL query such as "SELECT FROM user_contracts WHERE user_id = '<user ID>'" is used.

[1305] Data Analysis

[1306] The server analyzes the collected contract information. It uses data analysis algorithms (e.g., clustering and categorization) to analyze user needs and inquiry trends. For clustering, for example, the KMeans algorithm is used.

[1307] Image generation for the person in charge

[1308] The server generates a personalized facial image for each user based on the analysis results. This facial image generation uses a Generative Adversarial Network (GAN). For example, a GAN model can be used to generate a customized facial image based on the user's age, gender, and location.

[1309] Preparation for individual support

[1310] The server prepares individual customized response scenarios based on the analysis results. These scenarios include customizing a pre-prepared FAQ database and generating answer templates based on the user's past inquiries. Appropriate content is extracted from the FAQ database and the answer templates are customized.

[1311] Interface preparation

[1312] The server sends the generated facial image and customizable scenario to the user's device. This communication uses HTTP requests. Based on the transmitted data, the device prepares a display interface and presents it to the user.

[1313] emotion recognition

[1314] The device uses an emotion engine to recognize emotions from the user's facial expressions and voice. This emotion engine identifies the emotions the user is feeling in real time through facial expression recognition and voice analysis. For example, the EmotionEngine is used to analyze the user's emotions.

[1315] Start of response

[1316] A user contacts customer support. The device receives this inquiry and sends the inquiry details to the server. For example, the user types "I can't connect to the internet" and sends it from the device to the server.

[1317] Emotion-based dynamic adjustment

[1318] The server analyzes the user's inquiry and sentiment data sent from the device. Based on the analysis results, it dynamically adjusts customized response scenarios and answer templates to generate the optimal response that corresponds to the user's current sentiment. It generates a dynamically adjusted response based on sentiment data and the inquiry content.

[1319] Display the answer

[1320] The terminal displays the response received from the server to the user. At the same time, a generated image of the support representative's face is also displayed, and the output of voice and text is adjusted according to the user's emotions. For example, it might display, "Hello, I'm your personal support representative. How can I help you today?"

[1321] Feedback on the response

[1322] The device displays an interface for collecting user feedback. This includes a feedback form and a quick rating button. It receives user feedback and sends it to the server.

[1323] Feedback collection and analysis

[1324] The server collects user feedback and stores it in a database. The collected feedback is then analyzed and used as data to improve the quality of service. For example, the feedback can be analyzed to improve the service.

[1325] Specific example

[1326] For new customers

[1327] 1. The server retrieves and analyzes information about newly contracted users.

[1328] 2. The server uses a GAN to generate a facial image of the user's dedicated representative.

[1329] 3. The server sends the generated facial image and customized scenario to the terminal, which then displays it.

[1330] 4. The device analyzes the user's facial expressions and voice using an emotion engine to recognize the user's emotions.

[1331] 5. When a user makes an inquiry, the device displays "Hello, I am your personal support representative. How can I help you today?"

[1332] Measures to prevent cancellations

[1333] 1. The server retrieves and analyzes data from users who show signs of canceling their subscription.

[1334] 2. The server determines a response scenario to prevent cancellation and generates a facial image using a GAN.

[1335] 3. The server sends the generated image and corresponding scenario to the terminal, and the terminal displays it.

[1336] 4. The device analyzes the user's facial expressions and voice using an emotion engine to recognize the user's emotions.

[1337] 5. When a user inquires about canceling their service, the device will display a message saying, "Hello, this is your dedicated representative. We understand you are considering canceling your service. May we hear about any points of dissatisfaction?" and will respond appropriately based on the user's feelings.

[1338] Examples of prompts include, "Create customer support scenarios tailored to the emotions of newly signed-up users," and "Generate response scenarios that include emotion recognition for users considering cancellation."

[1339] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1340] Step 1:

[1341] The server retrieves the user's contract information from the database.

[1342] Specifically, the server connects to the database and executes an SQL query to extract the user's contract information.

[1343] Input: User ID

[1344] Output: User contract information (contract details, contract period, personal attributes, past inquiries)

[1345] The server executes the SQL query "SELECT FROM user_contracts WHERE user_id = '<User ID>'" to retrieve relevant information from the database.

[1346] Step 2:

[1347] The server analyzes the collected contract information.

[1348] In terms of specific operations, the server performs analysis using data analysis algorithms (clustering and categorization). Python's KMeans algorithm, for example, can be used.

[1349] Input: User's contract information

[1350] Output: Analysis results (user needs and inquiry trends)

[1351] Based on the contract information collected by the server, users are classified using the KMeans algorithm, and analysis results are obtained.

[1352] Step 3:

[1353] The server generates a user-friendly facial image based on the analysis results.

[1354] Specifically, the server uses a GAN to generate a facial image based on the user's basic information.

[1355] Input: Analysis results, user's basic information (age, gender, region)

[1356] Output: Generated facial image

[1357] The server uses a GAN model to generate facial images based on, for example, the user's age and gender.

[1358] Step 4:

[1359] The server prepares individual customized response scenarios based on the analysis results.

[1360] Specifically, the server customizes a pre-prepared FAQ database and generates answer templates based on past inquiries.

[1361] Input: Analysis results, FAQ database, past inquiry content

[1362] Output: Customizable scenarios, response templates

[1363] The server creates customized response scenarios and answer templates based on the FAQ database and past inquiries.

[1364] Step 5:

[1365] The server sends the generated facial image and customized scenario to the user's device.

[1366] In terms of specific operations, the server sends data to the terminal using an HTTP request.

[1367] Input: Generated facial image, customizable scenario

[1368] Output: Transmission status

[1369] The server uses an HTTP request to send the generated facial image and corresponding scenario to the terminal.

[1370] Step 6:

[1371] The device uses an emotion engine to recognize emotions from the user's facial expressions and voice.

[1372] Specifically, the device performs facial recognition and voice analysis, and identifies emotions through an emotion engine.

[1373] Input: User facial expression data, voice data

[1374] Output: User sentiment data

[1375] The device uses EmotionEngine to analyze, for example, the user's facial expressions and voice in real time, and obtain emotional data.

[1376] Step 7:

[1377] The user contacts customer support.

[1378] In terms of specific actions, the user enters their question using a contact form on their device.

[1379] Input: User's inquiry

[1380] Output: Query data

[1381] The user enters their inquiry into the device, and that information is sent to the device.

[1382] Step 8:

[1383] The terminal receives this query and sends the query details to the server.

[1384] Specifically, the terminal sends an HTTP request to forward the inquiry details to the server.

[1385] Input: Inquiry data

[1386] Output: Transmission status

[1387] The device sends an HTTP request containing the inquiry details to the server.

[1388] Step 9:

[1389] The server analyzes the user's inquiry and sentiment data sent from the device.

[1390] Specifically, the server analyzes sentiment data and query content, and dynamically adjusts the customized response scenario.

[1391] Input: Inquiry details, sentiment data

[1392] Output: Dynamically adjusted response scenarios, response templates

[1393] The server integrates inquiry content and sentiment data to generate optimized response scenarios and answers.

[1394] Step 10:

[1395] The terminal displays the response received from the server to the user.

[1396] In terms of specific operation, the terminal displays the received response and the face image of the person in charge on the display interface.

[1397] Input: Response content from the server, generated facial image

[1398] Output: Content displayed to the user

[1399] Based on the information received by the terminal from the server, the response content and facial image are displayed on the screen.

[1400] Step 11:

[1401] The device displays an interface for collecting feedback from the user.

[1402] In terms of specific actions, the device will present the user with a feedback form and rating buttons.

[1403] Input: None

[1404] Output: Feedback input interface

[1405] The device displays a form or button for collecting feedback.

[1406] Step 12:

[1407] Users provide feedback on the response.

[1408] In terms of specific actions, users enter their ratings and opinions into a feedback form and submit it.

[1409] Input: User feedback content

[1410] Output: Feedback data

[1411] The user enters feedback and sends it to their device.

[1412] Step 13:

[1413] The server collects user feedback and stores it in a database.

[1414] Specifically, the server adds the received feedback data to the database.

[1415] Input: User feedback content

[1416] Output: Saved status

[1417] The server receives the feedback data and saves it to the database.

[1418] Step 14:

[1419] The server analyzes the collected feedback and uses it as data to improve the quality of service.

[1420] Specifically, the server analyzes feedback data to gain insights that can be used to improve the quality of the service.

[1421] Input: Feedback data

[1422] Output: Improvement suggestions, analysis results

[1423] The server analyzes the feedback data and creates specific suggestions for service improvement.

[1424] (Application Example 2)

[1425] 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."

[1426] Traditional content delivery services have not adequately addressed individual user needs and emotions, making it difficult to improve user satisfaction. In particular, the lack of real-time content recommendations based on emotional changes during viewing, and the absence of dynamic interface adjustments, sometimes led to user stress and decreased satisfaction.

[1427] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for collecting user contract information, means for analyzing the collected contract information, means for preparing individual customized response scenarios for each user based on the analysis results, means for generating a user-friendly face image for each user, means for transmitting the generated face image and response scenario to the user terminal, means for displaying the generated face image and response scenario on the user terminal, means for collecting and analyzing user emotion data, means for dynamically recommending content based on the user's emotions, means for dynamically adjusting the interface according to the user's emotions, and means for collecting and improving feedback from the user. This enables real-time personalized content recommendation and interface adjustment in accordance with the user's emotions and needs, thereby improving user satisfaction.

[1428] "User contract information" refers to information such as the user's contract details, contract period, personal attributes, and past inquiries.

[1429] "Analysis results" refer to conclusions and insights derived by the server based on the collected data.

[1430] A "customized response scenario" refers to a pre-prepared plan and flow of individual responses based on user needs and inquiry trends.

[1431] A "face image" is a friendly, customized image of a face created using a Generative Adversarial Network (GAN).

[1432] A "user terminal" refers to a device used by a user, such as a smartphone, tablet, or personal computer.

[1433] "Emotional data" refers to information about a user's emotions, obtained in real time from their facial expressions and voice.

[1434] "A means of dynamically recommending content" refers to a technology that recommends appropriate content in real time based on user sentiment data and viewing history.

[1435] "Methods for dynamically adjusting the interface" refer to techniques that change the design and layout of an application's user interface (UI) in response to the user's emotions.

[1436] "Methods for collecting feedback and making improvements" refers to technologies that collect evaluations and opinions from users and use that information to perform data analysis and implement improvement measures to enhance future services.

[1437] This invention is a system for delivering dynamic content based on user emotions. First, a server collects user contract information and analyzes it. The analysis uses data mining techniques such as clustering and categorization. Next, based on the analysis results, a customized response scenario is prepared for each user, and a friendly facial image is generated. A Generative Adversarial Network (GAN) is used to generate this facial image. The generated facial image and response scenario are sent to the user's terminal.

[1438] On the user's terminal, the generated facial image and corresponding scenario are displayed, and the user's emotional data is collected in real time. Facial expression recognition and voice analysis technologies using a camera and microphone are employed to collect emotional data. Based on the emotional analysis results, the server dynamically recommends specific content. For example, a user experiencing stress might be recommended relaxing music or a comedy movie.

[1439] Furthermore, the application's user interface (UI) is dynamically adjusted according to the user's emotions. This means, for example, changing the UI design to a simpler and calmer one if the user is feeling stressed.

[1440] In addition, user feedback is collected and incorporated into subsequent analyses and scenario generation. This ensures continuous improvement of the service.

[1441] As a concrete example, when a user opens the application while feeling stressed, the server recommends a relaxing music playlist. The UI also automatically changes to calming colors to help reduce user stress. Afterwards, the user enters feedback on the content they viewed into a form within the application, and this data is sent to the server and used to improve the recommendation algorithm for the next time.

[1442] The program to realize this invention will be implemented using a general-purpose programming language such as Python. The libraries used will be "some_emotion_recognition_library" for emotion recognition and "some_content_recommendation_library" for content recommendation. Libraries such as "requests" and "SQLAlchemy" will be used for HTTP requests and database access.

[1443] Examples of prompt statements to input into the generative AI model are as follows:

[1444] "Design an algorithm that recommends relaxing music and movies based on the user's viewing history data and real-time sentiment analysis data. This algorithm should also consider past rating data to recommend content that best suits the user's current mood in real time. Additionally, the UI theme should dynamically change according to the user's mood."

[1445] In this way, a system is realized that dynamically recommends content based on user emotions, adjusts the interface, and incorporates feedback.

[1446] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1447] Step 1:

[1448] The server collects user contract information from the database. Input is identification information such as the user ID, and output is user contract information (contract details, contract period, personal attributes, past inquiry history). Specifically, it executes SQL queries to extract the necessary data.

[1449] Step 2:

[1450] The server analyzes the collected contract information. The input is the contract information obtained in step 1, and the output is the analysis results regarding user needs and inquiry trends. Data analysis algorithms (e.g., clustering and categorization) are used to group the data and derive insights.

[1451] Step 3:

[1452] The server generates user-specific, customized response scenarios and friendly facial images based on the analysis results. The input is the analysis results from step 2, and the output is the customized scenario and facial image corresponding to the user. A Generative Adversarial Network (GAN) is used to generate facial images and customize pre-defined response scenarios.

[1453] Step 4:

[1454] The server sends the generated face image and corresponding scenario to the user terminal. The input is the face image and corresponding scenario generated in step 3, and the output is the face image and corresponding scenario sent to the user terminal. Data is sent using an HTTP request.

[1455] Step 5:

[1456] The user terminal displays the generated face image and a customizable scenario. The input is the face image and corresponding scenario sent from the server, and the output is a visual display that the user can confirm on their terminal. The interface displays the face image and text.

[1457] Step 6:

[1458] The user terminal collects and analyzes the user's emotional data (facial expressions and voice) in real time. The input is the user's facial and voice data, and the output is the analysis results regarding the user's emotions. Specifically, it uses a camera and microphone and an emotion recognition engine to identify emotions.

[1459] Step 7:

[1460] The server dynamically recommends the most suitable content to the user based on sentiment data. The input is the sentiment data from step 6 and the user's viewing history, and the output is the recommended content. A content recommendation algorithm is used to select the content best suited to the user's current sentiment.

[1461] Step 8:

[1462] The user's device dynamically adjusts the interface according to the user's emotions. The input is the content and emotion data recommended in step 7, and the output is the UI adjusted according to the emotion. For example, a user feeling stressed will be shown a UI with calming colors.

[1463] Step 9:

[1464] Users view content and then provide feedback. The input is the user's feedback, and the output is feedback data. By filling out the feedback form on their device, the data is sent to the server.

[1465] Step 10:

[1466] The server analyzes the collected feedback data to improve response scenarios and recommendation algorithms. The input is the feedback collected in step 9, and the output is the improved response scenarios and recommendation algorithms. Analysis based on the feedback data improves system performance.

[1467] The above steps enable a dynamic content delivery system based on user emotions.

[1468] 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.

[1469] 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.

[1470] 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.

[1471] [Fourth Embodiment]

[1472] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1473] 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.

[1474] 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).

[1475] 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.

[1476] 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.

[1477] 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).

[1478] 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.

[1479] 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.

[1480] 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.

[1481] 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.

[1482] 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.

[1483] 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.

[1484] 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".

[1485] This invention relates to a system that provides mobile service providers with personalized, dedicated customer support for a large user base. To reduce costs while improving customer satisfaction, this system comprises the means and processing steps described below.

[1486] 1. Data Collection

[1487] The server collects user contract information. This contract information includes contract details, contract period, personal attributes, and past inquiry history. This data is retrieved from the database using SQL queries.

[1488] 2. Data Analysis

[1489] The server analyzes the collected contract information. Data analysis algorithms such as clustering and categorization are used for the analysis. Based on the analysis results, the needs and inquiry trends of the contract holders are identified.

[1490] 3. Image generation for the person in charge

[1491] The server generates a personalized facial image for each user based on the analysis results. Image generation algorithms such as GANs (Generative Adversarial Networks) are used to generate the facial images. The generated facial images are customized based on the user's basic information.

[1492] 4. Preparation for individualized support

[1493] The server prepares individual customized response scenarios based on the analysis results. This includes customizing a pre-prepared FAQ database and generating answer templates based on past inquiries.

[1494] 5. Interface preparation

[1495] The server sends the generated facial image and customized scenario to the user's terminal. An HTTP request is used for this transmission. The terminal then prepares to display the transmitted facial image and scenario in its user interface.

[1496] 6. Start of response

[1497] When a user contacts customer support, the device receives the user's inquiry input and sends it to the server. The server analyzes the inquiry and generates the most appropriate response based on pre-prepared response templates. The device then displays the response to the user, along with a profile picture of the assigned support representative.

[1498] 7. Feedback on the response

[1499] After the service is completed, the device displays an interface for collecting user feedback. This feedback includes a form and a simple rating button. The user enters their feedback, the server collects the feedback information, stores it in a database, and analyzes it to improve future services.

[1500] Specific example

[1501] For new customers

[1502] 1. The server retrieves and analyzes information about newly contracted users.

[1503] 2. The server uses a GAN to generate a facial image of the user's dedicated representative.

[1504] 3. The server sends the generated facial image and customized scenario to the terminal, which then displays it.

[1505] 4. When a user makes an inquiry, the device displays "Hello, I am your personal support representative. How can I help you today?"

[1506] Measures to prevent cancellations

[1507] 1. The server retrieves and analyzes data from users who show signs of canceling their subscription.

[1508] 2. The server determines a response scenario to prevent cancellation and generates a facial image using a GAN.

[1509] 3. The server sends the generated image and corresponding scenario to the terminal, and the terminal displays it.

[1510] 4. When a user inquires about canceling their service, the device displays the message, "Hello, this is your dedicated representative. We understand you are considering canceling your service. Could you tell us what your concerns are?" and asks for the reason for cancellation.

[1511] In this way, the present invention can provide dedicated customer support tailored to the individual needs of users, thereby improving customer satisfaction while keeping costs down.

[1512] The following describes the processing flow.

[1513] Step 1:

[1514] The server retrieves the user's contract information from the database. This contract information includes contract details, contract period, personal attributes, and past inquiry history. This information is retrieved from the database using SQL queries.

[1515] Step 2:

[1516] The server analyzes the collected contract information. It uses data analysis algorithms (e.g., clustering and categorization) to analyze user needs and inquiry trends. Based on the analysis results, it determines individual response strategies.

[1517] Step 3:

[1518] The server generates a personalized facial image for each user based on the analysis results. Image generation algorithms such as GANs (Generative Adversarial Networks) are used to generate these facial images. The generated facial images are customized based on the user's basic information (e.g., age, gender, region).

[1519] Step 4:

[1520] The server prepares individual customized response scenarios based on the analysis results. These scenarios include customizing a pre-prepared FAQ database and generating answer templates based on the user's past inquiries.

[1521] Step 5:

[1522] The server sends the generated facial image and customized scenario to the user's device. This transmission uses an HTTP request. The device prepares an interface to display the transmitted facial image and corresponding scenario.

[1523] Step 6:

[1524] A user contacts customer support. The device receives this inquiry and sends the inquiry details to the server.

[1525] Step 7:

[1526] The server analyzes the user's inquiry. Based on pre-prepared response templates, it generates the most appropriate response and sends it to the terminal.

[1527] Step 8:

[1528] The terminal displays the response received from the server to the user. At the same time, a generated image of the person in charge's face is also displayed.

[1529] Step 9:

[1530] After the support is complete, the device will display an interface for collecting user feedback. This includes a feedback form and a simple rating button.

[1531] Step 10:

[1532] Users provide feedback on the response they receive. This includes evaluations and opinions about the response.

[1533] Step 11:

[1534] The server collects user feedback and stores it in a database. The collected feedback is then analyzed and used as data to improve the quality of service.

[1535] (Example 1)

[1536] 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".

[1537] Mobile service providers are required to provide individual, dedicated customer support to a large number of users, but effectively and cost-effectively serving such a large user base is difficult. Traditional systems suffer from problems such as inconsistent quality of support, long response times, and excessive resource requirements.

[1538] 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.

[1539] In this invention, the server includes means for collecting user contract information, means for analyzing the collected contract information, means for generating a user-friendly facial image based on the analysis results, means for preparing individual customized response scenarios based on the analysis results, means for transmitting the generated facial image and customized response scenario to the user terminal, and means for receiving, analyzing, and generating an optimal response to inquiries via the user terminal. This makes it possible to provide efficient and cost-effective dedicated customer support to many users and improve customer satisfaction.

[1540] "Means for collecting user contract information" refers to methods or systems for obtaining data related to services contracted by users, and includes means of collecting information from a database that includes contract details, contract period, personal attributes, past inquiries, etc.

[1541] "Means for analyzing collected contract information" refers to methods and systems for analyzing collected user contract data and extracting specific patterns or trends, and includes data analysis algorithms such as clustering and categorization.

[1542] "Means for generating friendly facial images based on analysis results" refers to methods or systems for generating different friendly facial images for each user based on the results of data analysis, and these methods utilize generative adversarial networks (GANs).

[1543] "Means for preparing individual customized response scenarios based on analysis results" refers to methods or systems for preparing different customized response scenarios for each user using the results of data analysis, and includes means for generating response templates based on a pre-prepared FAQ database or past inquiry content.

[1544] "Means for sending generated facial images and customized response scenarios to the user's terminal" refers to a method or system for sending generated facial images and customized response scenarios to the user's terminal, and is a means that uses HTTP requests.

[1545] "Means for displaying face images and customized scenarios generated on a user terminal" refers to methods or systems for displaying face images and corresponding scenarios sent to a user terminal on a user interface.

[1546] "Means for receiving, analyzing, and generating optimal responses via user terminals" refers to methods or systems for receiving user inquiries via user terminals, analyzing their content, and generating optimal responses, and which utilize natural language processing (NLP) algorithms.

[1547] "Means for collecting user feedback and improving analysis results" refers to methods and systems for collecting and storing user feedback information and using it to improve the quality of analysis results and corresponding scenarios.

[1548] This invention relates to a system that provides mobile service providers with personalized, dedicated customer support for a large user base. This system consists of multiple means and processing steps to improve customer satisfaction while keeping costs down. Each means and its associated processing steps are as follows:

[1549] First, the server collects user contract information. This information includes contract details, contract period, personal attributes, and past inquiry history. This data is retrieved from the database using SQL queries. For example, the server connects to the user database and collects data by executing the SQL query "SELECT FROM user_contracts WHERE user_id = [User ID]".

[1550] Next, the server uses K-means clustering and categorization algorithms to analyze the collected contract information. For example, it performs clustering using K-means(n_clusters=5).fit(contract information) and extracts features to identify the needs and inquiry trends of the contract holders.

[1551] Based on the analysis results, the server generates face images using a GAN (Generative Adversarial Network). In this process, it selects a GAN model with parameters optimized for a specific user category and generates friendly face images using prompts. For example, if you select StyleGAN2 as the GAN model to use and input prompts such as "Attributes for the newly generated face image: friendly, smiling, adult male," the server will generate the image. The generated face image is then temporarily stored in the server's storage.

[1552] Furthermore, the server prepares individual customized response scenarios based on the analysis results. To do this, it extracts relevant questions using a pre-prepared FAQ database and past inquiries, and generates individual answer templates. For example, it executes an SQL query such as "SELECT faq_answer FROM FAQ WHERE question LIKE '%contract renewal%'" to customize the response scenario.

[1553] The generated facial image and customized response scenario are sent from the server to the user's terminal. The HTTP protocol is used for this transmission. For example, the data is sent in the format "POST / api / send_userdata HTTP / 1.1".

[1554] The user terminal stores the received facial image and corresponding scenario in memory and prepares to display them on the user interface. Specifically, the facial image and text are appropriately positioned.

[1555] When a user submits an inquiry, the terminal receives the inquiry details. The received content is sent to the server using an HTTP request such as "POST / api / send_query HTTP / 1.1". The server analyzes the received inquiry using a natural language processing (NLP) algorithm and generates the optimal response. One example of an NLP model used is GPT-3. Based on the inquiry, the server generates the optimal response to a question such as "Please tell me about contract renewal." The generated response is displayed on the user's terminal, along with a photo of the representative's face.

[1556] Once the interaction is complete, the user's terminal displays a feedback interface. This feedback includes a form and a simple rating button. When the user enters their feedback and presses the submit button, the server receives the feedback and saves it to the database. For example, an SQL query such as "INSERT INTO user_feedback (user_id, rating, comments) VALUES ([User ID], 5, 'I was satisfied with the interaction.')" is executed. The server analyzes the collected feedback and uses it to improve future interactions.

[1557] Example of a prompt

[1558] "Please generate a friendly, dedicated representative profile picture and support scenarios for new mobile service subscribers. Please also refer to their contract details and past inquiries."

[1559] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1560] Step 1:

[1561] The server collects user contract information from the database.

[1562] Input: User ID

[1563] Process: Execute an SQL query against the database and retrieve the corresponding row data. Example: "SELECT FROM user_contracts WHERE user_id = [User ID]"

[1564] Output: User contract information including contract details, contract period, personal attributes, and past inquiries.

[1565] Step 2:

[1566] The server analyzes the contract information it has collected.

[1567] Input: Collected contract information

[1568] Processing: The contract information is divided into clusters using the K-means clustering algorithm. Example: "K-means(n_clusters=5).fit(contract information)"

[1569] Output: Classification results and features for each user cluster

[1570] Step 3:

[1571] The server generates a friendly-looking facial image based on the analysis results.

[1572] Input: User classification results and features for each cluster

[1573] Processing: Generate face images using a GAN (Generative Adversarial Network) model. Example: "GAN model to use: StyleGAN2" "Attributes of the newly generated face image: Friendly, smiling, adult male"

[1574] Output: Generated facial image

[1575] Step 4:

[1576] The server prepares individual customized scenarios based on the analysis results.

[1577] Input: Cluster classification results, past inquiry details

[1578] Processing: Extract relevant questions from the FAQ database and generate individual answer templates. Example: "SELECT faq_answer FROM FAQ WHERE question LIKE '%contract renewal%'"

[1579] Output: Customized response scenarios and answer templates

[1580] Step 5:

[1581] The server sends the generated facial image and customized scenario to the user's terminal.

[1582] Input: Generated facial image and customizable scenario

[1583] Processing: Sends data to the user's terminal using the HTTP protocol. Example: "POST / api / send_userdata HTTP / 1.1"

[1584] Output: The transmitted data is stored on the user's terminal.

[1585] Step 6:

[1586] The user terminal displays the generated facial image and the customized scenario.

[1587] Input: Received facial image and corresponding scenario

[1588] Processing: Place and display face images and scenarios on the user interface.

[1589] Output: Face image and corresponding scenario displayed to the user

[1590] Step 7:

[1591] When a user submits an inquiry, the user's device receives the inquiry details and sends them to the server.

[1592] Input: User inquiry details

[1593] Processing: Receives the query content and sends it to the server as an HTTP request. Example: "POST / api / send_query HTTP / 1.1"

[1594] Output: Sent inquiry content

[1595] Step 8:

[1596] The server analyzes the inquiry and generates the most appropriate response.

[1597] Input: User inquiry details

[1598] Processing: Analyze the query using a natural language processing (NLP) algorithm and generate the optimal response. NLP model used: GPT-3

[1599] Output: Generated answer

[1600] Step 9:

[1601] The user's terminal displays the generated response to the user, along with the face image of the person in charge.

[1602] Input: Generated response and facial image

[1603] Processing: Display the answer on the user interface, along with the facial image.

[1604] Output: Responses and facial images displayed to the user

[1605] Step 10:

[1606] Once the issue is resolved, the user's device will display a feedback interface.

[1607] Input: Trigger for completion of response

[1608] Processing: Display a feedback form and a simple rating button on the interface.

[1609] Output: User-inputtable feedback interface

[1610] Step 11:

[1611] When a user enters feedback and presses the submit button, the server receives the feedback and saves it to the database.

[1612] Input: User feedback content

[1613] Process: Receive the feedback and save it to the database. Example: "INSERT INTO user_feedback (user_id, rating, comments) VALUES ([User ID], 5, 'I was satisfied with the service.')"

[1614] Output: Saved feedback data

[1615] Step 12:

[1616] The server will analyze the collected feedback and use it to improve future responses.

[1617] Input: Saved feedback data

[1618] Processing: Analyze feedback data and use it to improve the service.

[1619] Output: Insights for improved response scenarios and system updates

[1620] (Application Example 1)

[1621] 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".

[1622] In food service provision, a challenge is responding quickly and individually to customers' specific requests and inquiries. Traditional customer support systems often provide uniform responses, making it difficult to improve customer satisfaction and provide efficient support. In particular, in food service, responses must be based on customer preferences and past order history, and traditional systems often lacked flexibility and personalization.

[1623] 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.

[1624] In this invention, the server includes means for collecting user contract information, means for analyzing the collected contract information, means for preparing individual customized response scenarios for each user based on the analysis results, means for generating a user-friendly facial image for each user, means for transmitting the generated facial image and response scenario to the user terminal, means for displaying the generated facial image and response scenario on the user terminal, and means for functioning as a dedicated support person for food services via a smart device. This enables prompt and individual responses to customers' special requests and inquiries, improving customer satisfaction and providing efficient support.

[1625] "User contract information" refers to information such as the contract details, contract period, personal attributes, and past inquiries of food service users.

[1626] "Means of collection" refers to the technical means of obtaining user contract information from a database.

[1627] "Means of analysis" refers to technical means of analyzing collected information using data analysis algorithms such as clustering and categorization.

[1628] "Individualized customized response scenarios" refer to specific response plans or scenarios prepared for individual users based on the analysis results.

[1629] A "friendly face image" refers to a customized face image that users can find appealing.

[1630] "Generative means" refers to technical methods for generating friendly facial images using GANs (Generative Adversarial Networks).

[1631] "Means of transmission" refers to the technical means for transmitting the generated facial image and corresponding scenario to the user's terminal.

[1632] "Means of display" refers to the technical means for displaying face images and corresponding scenarios generated on the user's terminal.

[1633] A "smart device" refers to a portable information terminal with internet connectivity, such as a smartphone or tablet.

[1634] A "dedicated support representative" refers to a virtual support representative who is assigned to each individual user.

[1635] This invention relates to a dedicated support system for users in the food service industry. This system is configured to provide food service providers with personalized, dedicated customer support to a large user base and is implemented through the following elements.

[1636] The server first collects user contract information. This contract information includes the user's contract details, contract period, personal attributes, and past inquiry history. This data is retrieved from an SQL database using SQL queries.

[1637] Next, the server analyzes the collected contract information. Data analysis algorithms such as clustering and categorization are used for this analysis. Python's scikit-learn library is used for data analysis. Based on the analysis results, the needs and inquiry trends of the contract holders are identified.

[1638] Based on the analysis results, the server generates a user-friendly facial image for each user. Image generation algorithms such as GANs (Generative Adversarial Networks) are used to generate these facial images. The generated facial images are customized based on the user's basic information. Deep learning frameworks such as TensorFlow and PyTorch are used to implement the GANs.

[1639] The server prepares individual customized response scenarios based on the analysis results. This includes customizing the pre-prepared FAQ database and generating answer templates based on past inquiries. This improves the quality and efficiency of responses.

[1640] The prepared facial image and customized scenario are sent from the server to the user's terminal. HTTP requests are used for this transmission. The smart device (smartphone or tablet) prepares to display the transmitted facial image and corresponding scenario in the user interface.

[1641] When a user contacts customer support, the device receives the user's inquiry input and sends it to the server. The server analyzes the inquiry and generates the most appropriate response based on pre-prepared response templates. The device then displays the response to the user, along with a profile picture of the assigned support representative.

[1642] After the service is completed, the device displays an interface for collecting user feedback. This feedback includes a form and a simple rating button. The user enters their feedback, the server collects the feedback information, stores it in a database, and analyzes it to improve future services.

[1643] Specific example

[1644] For example, if user A has a habit of ordering a specific pizza every Friday, the server will use this information to create a dedicated support representative for user A and prepare support scenarios to address any special requests regarding the pizza. When user A requests support, they will be shown a message such as, "Hello, I am your dedicated support representative. Do you have any questions regarding your pizza order?"

[1645] Examples of input prompts for a generative AI model

[1646] "Based on User A's past order history, generate a facial image of a dedicated support representative and a customized scenario for ordering pizza."

[1647] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1648] Step 1:

[1649] The server collects user contract information.

[1650] Input: User ID

[1651] Operation: The server queries the SQL database to retrieve the user's contract details, contract period, personal attributes, and past query history.

[1652] Output: User contract information

[1653] Step 2:

[1654] The server analyzes the collected contract information.

[1655] Input: User contract information

[1656] Operation: The server analyzes contract information using data analysis algorithms such as clustering and categorization. Specifically, it uses the Python scikit-learn library to cluster the data.

[1657] Output: Analysis results (user needs and inquiry trends)

[1658] Step 3:

[1659] The server generates a user-friendly facial image based on the analysis results.

[1660] Input: Analysis results

[1661] Operation: The server uses a GAN (Generative Adversarial Network) to generate user-friendly facial images. TensorFlow and PyTorch are used as deep learning frameworks for this purpose.

[1662] Output: Friendly facial image

[1663] Step 4:

[1664] The server prepares individual customized response scenarios based on the analysis results.

[1665] Input: Analysis results

[1666] Operation: The server customizes a pre-prepared FAQ database and generates answer templates based on past inquiries.

[1667] Output: Customizable scenarios

[1668] Step 5:

[1669] The server sends the generated facial image and customized scenario to the user's terminal.

[1670] Input: Friendly facial image, customizable scenarios

[1671] Operation: The server uses an HTTP request to send the generated facial image and customizable scenario to the user's smart device.

[1672] Output: Face image and corresponding scenario sent to the user's terminal

[1673] Step 6:

[1674] The user's terminal displays the generated facial image and the customizable scenario.

[1675] Input: Face images sent from the server and customizable scenarios.

[1676] Operation: The user terminal displays the transmitted facial image and customizable scenario on the user interface.

[1677] Output: Displayed facial image and corresponding scenario

[1678] Step 7:

[1679] The user contacts customer support.

[1680] Input: User inquiry

[1681] Operation: The user contacts customer support via a smart device. The device sends this inquiry to the server.

[1682] Output: User queries sent to the server

[1683] Step 8:

[1684] The server analyzes the user's inquiry and generates the most appropriate response.

[1685] Input: User inquiry

[1686] Operation: The server analyzes the query text and generates the most suitable response based on pre-prepared response templates.

[1687] Output: Best Answer

[1688] Step 9:

[1689] The device displays the answer to the user, along with a generated image of the person in charge's face.

[1690] Input: Best response sent from the server, face image of the person in charge

[1691] Operation: The device displays the answer and the face image of the person in charge on the screen.

[1692] Output: Displayed response and the respondent's face image

[1693] Step 10:

[1694] The user enters feedback, and the server collects the feedback information and stores it in a database.

[1695] Input: User feedback

[1696] Operation: The terminal displays a feedback collection interface, and the user enters feedback. The entered feedback is sent to the server and stored in the database.

[1697] Output: Saved feedback information

[1698] Examples of input prompts for a generative AI model

[1699] "Based on User A's past order history, generate a facial image of a dedicated support representative and a customized scenario for ordering pizza."

[1700] 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.

[1701] This invention aims to further improve customer satisfaction by combining a system that provides mobile service providers with personalized, dedicated customer support for a large user base with an emotion engine that recognizes user emotions. This system consists of the means and processing steps described below.

[1702] 1. Data Collection

[1703] The server retrieves user contract information from the database. This contract information includes contract details, contract period, personal attributes, and past inquiry history. This data is retrieved from the database using SQL queries.

[1704] 2. Data Analysis

[1705] The server analyzes the collected contract information. It uses data analysis algorithms (e.g., clustering and categorization) to analyze user needs and inquiry trends. Based on the analysis results, it determines individual response strategies.

[1706] 3. Image generation for the person in charge

[1707] The server generates a personalized facial image for each user based on the analysis results. Image generation algorithms such as GANs (Generative Adversarial Networks) are used to generate these facial images. The generated facial images are customized based on the user's basic information (e.g., age, gender, region).

[1708] 4. Preparation for individualized support

[1709] The server prepares individual customized response scenarios based on the analysis results. These scenarios include customizing a pre-prepared FAQ database and generating answer templates based on the user's past inquiries.

[1710] 5. Interface preparation

[1711] The server sends the generated facial image and customized scenario to the user's device. An HTTP request is used for this transmission. The device then prepares an interface to display the transmitted facial image and corresponding scenario.

[1712] 6. Emotion recognition

[1713] The device uses an emotion engine to recognize emotions from the user's facial expressions and voice. The emotion engine identifies the emotions the user is feeling in real time through facial expression recognition and voice analysis.

[1714] 7. Start of response

[1715] A user contacts customer support. The device receives this inquiry and sends the inquiry details to the server.

[1716] 8. Emotion-based dynamic adjustment

[1717] The server analyzes the user's inquiry and sentiment data sent from the device. Based on the analysis results, it dynamically adjusts customized response scenarios and answer templates to generate the optimal response that corresponds to the user's current sentiment.

[1718] 9. Display the answer

[1719] The terminal displays the response received from the server to the user. At the same time, a generated image of the person in charge is also displayed, and the output of voice and text is adjusted according to the user's emotions.

[1720] 10. Feedback on the response

[1721] The device displays an interface for collecting user feedback, which includes a feedback form and a quick rating button.

[1722] 11. Gathering and analyzing feedback

[1723] Users provide feedback on the service they received, including evaluations and opinions about the service. The server collects this user feedback and stores it in a database. The collected feedback is then analyzed and used as data to improve the quality of the service.

[1724] Specific example

[1725] For new customers

[1726] 1. The server retrieves and analyzes information about newly contracted users.

[1727] 2. The server uses a GAN to generate a facial image of the user's dedicated representative.

[1728] 3. The server sends the generated facial image and customized scenario to the terminal, which then displays it.

[1729] 4. The device analyzes the user's facial expressions and voice using an emotion engine to recognize the user's emotions.

[1730] 5. When a user makes an inquiry, the device displays "Hello, I am your personal support representative. How can I help you today?"

[1731] Measures to prevent cancellations

[1732] 1. The server retrieves and analyzes data from users who show signs of canceling their subscription.

[1733] 2. The server determines a response scenario to prevent cancellation and generates a facial image using a GAN.

[1734] 3. The server sends the generated image and corresponding scenario to the terminal, and the terminal displays it.

[1735] 4. The device analyzes the user's facial expressions and voice using an emotion engine to recognize the user's emotions.

[1736] 5. When a user inquires about canceling their service, the device will display a message saying, "Hello, this is your dedicated representative. We understand you are considering canceling your service. May we hear about any points of dissatisfaction?" and will respond appropriately based on the user's feelings.

[1737] In this way, the present invention provides personalized customer support tailored to individual needs based on the user's emotions, further improving customer satisfaction.

[1738] The following describes the processing flow.

[1739] Step 1:

[1740] The server retrieves user contract information from the database. This contract information includes contract details, contract period, personal attributes, and past inquiry history. This data is retrieved from the database using SQL queries.

[1741] Step 2:

[1742] The server analyzes the collected contract information. It uses data analysis algorithms such as clustering and categorization to analyze user needs and inquiry trends. Based on the analysis results, it determines individual response strategies for each user.

[1743] Step 3:

[1744] The server generates a user-friendly facial image based on the analysis results. It uses image generation algorithms such as GAN (Generative Adversarial Network) to customize the facial image based on the user's basic information (age, gender, region, etc.).

[1745] Step 4:

[1746] The server prepares individual customized scenarios. It customizes a pre-configured FAQ database and generates answer templates based on the user's past inquiries.

[1747] Step 5:

[1748] The server sends the generated facial image and customized scenario to the user's terminal. The facial image and scenario data are sent to the terminal using an HTTP request.

[1749] Step 6:

[1750] The device displays the generated facial image and a customizable scenario. At this point, the user interface is prepared, and the scenario and facial image are visually displayed to the user.

[1751] Step 7:

[1752] The device analyzes the user's facial expressions and voice, and recognizes emotions using an emotion engine. The emotion engine utilizes facial recognition and voice analysis algorithms to identify the user's emotions (e.g., joy, anger, sadness, etc.) in real time.

[1753] Step 8:

[1754] A user contacts customer support. The device receives the user's inquiry and sends it to the server.

[1755] Step 9:

[1756] The server analyzes the user's inquiry and the sentiment data recognized by the sentiment engine. Based on the analysis results, it generates the optimal response corresponding to the user's emotions and dynamically adjusts the customized response scenario.

[1757] Step 10:

[1758] The server sends the generated response to the terminal. The response data is sent to the terminal as an HTTP response.

[1759] Step 11:

[1760] The terminal displays the response from the server to the user. Along with the generated image of the person in charge, it displays the response in voice or text format, adjusted to the user's emotions.

[1761] Step 12:

[1762] After the support is complete, the device will display an interface for collecting feedback from the user. It will provide a feedback form and a simple rating button to receive opinions and ratings from the user.

[1763] Step 13:

[1764] Users enter feedback. This feedback includes evaluations and opinions about the actions taken.

[1765] Step 14:

[1766] The server collects user feedback and stores it in a database. The collected feedback is analyzed and used as data to improve the quality of future responses.

[1767] (Example 2)

[1768] 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".

[1769] Traditional customer support systems can provide personalized support based on user contract information, but they lack the ability to recognize user emotions and dynamically adjust responses, limiting their potential for improving customer satisfaction. Furthermore, they lack mechanisms to analyze user feedback and improve responses, resulting in insufficient improvement in service quality.

[1770] 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.

[1771] In this invention, the server includes means for collecting user contract information, means for analyzing the collected contract information, means for preparing individual customized response scenarios for each user based on the analysis results, means for generating a user-friendly facial image for each user, means for transmitting the generated facial image and response scenario to the user terminal, means for collecting emotional data from the user, means for analyzing the user's emotional data and dynamically adjusting the customized response scenario, and means for displaying the generated facial image and dynamically adjusted response scenario on the user terminal. This enables individualized responses based on the user's emotions, thereby improving customer satisfaction.

[1772] "Contract information" refers to data such as the details of the contract a mobile service user has made, the contract period, personal attributes, and past inquiries.

[1773] A "database" is a system used to store and manage user contract information, inquiry details, and other data.

[1774] "Data analysis" is the process of analyzing user needs and inquiry trends based on collected contract information, and determining individual response strategies.

[1775] Clustering is a technique for grouping and classifying user data based on specific criteria.

[1776] A "face image" is an image of a human face that is generated to make the user feel a sense of familiarity.

[1777] GAN (Generative Adversarial Network) is a type of image generation algorithm that generates high-quality images by pitting two neural networks against each other.

[1778] A "customized scenario" is a scenario created for each user based on the analysis results, designed to address their specific needs.

[1779] An "emotion engine" is software that recognizes and analyzes emotions from a user's facial expressions and voice.

[1780] An "HTTP request" is a protocol used to send and receive data between a server and a terminal.

[1781] "Dynamic adjustment" is a process that optimizes response scenarios in real time based on user emotion data.

[1782] "Feedback" refers to the evaluations and opinions that users provide regarding the response.

[1783] Modes for carrying out the invention

[1784] This invention aims to further improve customer satisfaction by combining a system that provides mobile service providers with personalized, dedicated customer support for large user bases with an emotion engine that recognizes user emotions. This system links a server and a terminal and provides a function to dynamically adjust individual responses using user contract information and emotion data.

[1785] Data collection

[1786] The server retrieves user contract information from the database. This contract information includes contract details, contract period, personal attributes, and past inquiry history. This data is retrieved from the database using SQL queries. For example, an SQL query such as "SELECT FROM user_contracts WHERE user_id = '<user ID>'" is used.

[1787] Data Analysis

[1788] The server analyzes the collected contract information. It uses data analysis algorithms (e.g., clustering and categorization) to analyze user needs and inquiry trends. For clustering, for example, the KMeans algorithm is used.

[1789] Image generation for the person in charge

[1790] The server generates a personalized facial image for each user based on the analysis results. This facial image generation uses a Generative Adversarial Network (GAN). For example, a GAN model can be used to generate a customized facial image based on the user's age, gender, and location.

[1791] Preparation for individual support

[1792] The server prepares individual customized response scenarios based on the analysis results. These scenarios include customizing a pre-prepared FAQ database and generating answer templates based on the user's past inquiries. Appropriate content is extracted from the FAQ database and the answer templates are customized.

[1793] Interface preparation

[1794] The server sends the generated facial image and customizable scenario to the user's device. This communication uses HTTP requests. Based on the transmitted data, the device prepares a display interface and presents it to the user.

[1795] emotion recognition

[1796] The device uses an emotion engine to recognize emotions from the user's facial expressions and voice. This emotion engine identifies the emotions the user is feeling in real time through facial expression recognition and voice analysis. For example, the EmotionEngine is used to analyze the user's emotions.

[1797] Start of response

[1798] A user contacts customer support. The device receives this inquiry and sends the inquiry details to the server. For example, the user types "I can't connect to the internet" and sends it from the device to the server.

[1799] Emotion-based dynamic adjustment

[1800] The server analyzes the user's inquiry and sentiment data sent from the device. Based on the analysis results, it dynamically adjusts customized response scenarios and answer templates to generate the optimal response that corresponds to the user's current sentiment. It generates a dynamically adjusted response based on sentiment data and the inquiry content.

[1801] Display the answer

[1802] The terminal displays the response received from the server to the user. At the same time, a generated image of the support representative's face is also displayed, and the output of voice and text is adjusted according to the user's emotions. For example, it might display, "Hello, I'm your personal support representative. How can I help you today?"

[1803] Feedback on the response

[1804] The device displays an interface for collecting user feedback. This includes a feedback form and a quick rating button. It receives user feedback and sends it to the server.

[1805] Feedback collection and analysis

[1806] The server collects user feedback and stores it in a database. The collected feedback is then analyzed and used as data to improve the quality of service. For example, the feedback can be analyzed to improve the service.

[1807] Specific example

[1808] For new customers

[1809] 1. The server retrieves and analyzes information about newly contracted users.

[1810] 2. The server uses a GAN to generate a facial image of the user's dedicated representative.

[1811] 3. The server sends the generated facial image and customized scenario to the terminal, which then displays it.

[1812] 4. The device analyzes the user's facial expressions and voice using an emotion engine to recognize the user's emotions.

[1813] 5. When a user makes an inquiry, the device displays "Hello, I am your personal support representative. How can I help you today?"

[1814] Measures to prevent cancellations

[1815] 1. The server retrieves and analyzes data from users who show signs of canceling their subscription.

[1816] 2. The server determines a response scenario to prevent cancellation and generates a facial image using a GAN.

[1817] 3. The server sends the generated image and corresponding scenario to the terminal, and the terminal displays it.

[1818] 4. The device analyzes the user's facial expressions and voice using an emotion engine to recognize the user's emotions.

[1819] 5. When a user inquires about canceling their service, the device will display a message saying, "Hello, this is your dedicated representative. We understand you are considering canceling your service. May we hear about any points of dissatisfaction?" and will respond appropriately based on the user's feelings.

[1820] Examples of prompts include, "Create customer support scenarios tailored to the emotions of newly signed-up users," and "Generate response scenarios that include emotion recognition for users considering cancellation."

[1821] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1822] Step 1:

[1823] The server retrieves the user's contract information from the database.

[1824] Specifically, the server connects to the database and executes an SQL query to extract the user's contract information.

[1825] Input: User ID

[1826] Output: User contract information (contract details, contract period, personal attributes, past inquiries)

[1827] The server executes the SQL query "SELECT FROM user_contracts WHERE user_id = '<User ID>'" to retrieve relevant information from the database.

[1828] Step 2:

[1829] The server analyzes the collected contract information.

[1830] In terms of specific operations, the server performs analysis using data analysis algorithms (clustering and categorization). Python's KMeans algorithm, for example, can be used.

[1831] Input: User's contract information

[1832] Output: Analysis results (user needs and inquiry trends)

[1833] Based on the contract information collected by the server, users are classified using the KMeans algorithm, and analysis results are obtained.

[1834] Step 3:

[1835] The server generates a user-friendly facial image based on the analysis results.

[1836] Specifically, the server uses a GAN to generate a facial image based on the user's basic information.

[1837] Input: Analysis results, user's basic information (age, gender, region)

[1838] Output: Generated facial image

[1839] The server uses a GAN model to generate facial images based on, for example, the user's age and gender.

[1840] Step 4:

[1841] The server prepares individual customized response scenarios based on the analysis results.

[1842] Specifically, the server customizes a pre-prepared FAQ database and generates answer templates based on past inquiries.

[1843] Input: Analysis results, FAQ database, past inquiry content

[1844] Output: Customizable scenarios, response templates

[1845] The server creates customized response scenarios and answer templates based on the FAQ database and past inquiries.

[1846] Step 5:

[1847] The server sends the generated facial image and customized scenario to the user's device.

[1848] In terms of specific operations, the server sends data to the terminal using an HTTP request.

[1849] Input: Generated facial image, customizable scenario

[1850] Output: Transmission status

[1851] The server uses an HTTP request to send the generated facial image and corresponding scenario to the terminal.

[1852] Step 6:

[1853] The device uses an emotion engine to recognize emotions from the user's facial expressions and voice.

[1854] Specifically, the device performs facial recognition and voice analysis, and identifies emotions through an emotion engine.

[1855] Input: User facial expression data, voice data

[1856] Output: User sentiment data

[1857] The device uses EmotionEngine to analyze, for example, the user's facial expressions and voice in real time, and obtain emotional data.

[1858] Step 7:

[1859] The user contacts customer support.

[1860] In terms of specific actions, the user enters their question using a contact form on their device.

[1861] Input: User's inquiry

[1862] Output: Query data

[1863] The user enters their inquiry into the device, and that information is sent to the device.

[1864] Step 8:

[1865] The terminal receives this query and sends the query details to the server.

[1866] Specifically, the terminal sends an HTTP request to forward the inquiry details to the server.

[1867] Input: Inquiry data

[1868] Output: Transmission status

[1869] The device sends an HTTP request containing the inquiry details to the server.

[1870] Step 9:

[1871] The server analyzes the user's inquiry and sentiment data sent from the device.

[1872] Specifically, the server analyzes sentiment data and query content, and dynamically adjusts the customized response scenario.

[1873] Input: Inquiry details, sentiment data

[1874] Output: Dynamically adjusted response scenarios, response templates

[1875] The server integrates inquiry content and sentiment data to generate optimized response scenarios and answers.

[1876] Step 10:

[1877] The terminal displays the response received from the server to the user.

[1878] In terms of specific operation, the terminal displays the received response and the face image of the person in charge on the display interface.

[1879] Input: Response content from the server, generated facial image

[1880] Output: Content displayed to the user

[1881] Based on the information received by the terminal from the server, the response content and facial image are displayed on the screen.

[1882] Step 11:

[1883] The device displays an interface for collecting feedback from the user.

[1884] In terms of specific actions, the device will present the user with a feedback form and rating buttons.

[1885] Input: None

[1886] Output: Feedback input interface

[1887] The device displays a form or button for collecting feedback.

[1888] Step 12:

[1889] Users provide feedback on the response.

[1890] In terms of specific actions, users enter their ratings and opinions into a feedback form and submit it.

[1891] Input: User feedback content

[1892] Output: Feedback data

[1893] The user enters feedback and sends it to their device.

[1894] Step 13:

[1895] The server collects user feedback and stores it in a database.

[1896] Specifically, the server adds the received feedback data to the database.

[1897] Input: User feedback content

[1898] Output: Saved status

[1899] The server receives the feedback data and saves it to the database.

[1900] Step 14:

[1901] The server analyzes the collected feedback and uses it as data to improve the quality of service.

[1902] Specifically, the server analyzes feedback data to gain insights that can be used to improve the quality of the service.

[1903] Input: Feedback data

[1904] Output: Improvement suggestions, analysis results

[1905] The server analyzes the feedback data and creates specific suggestions for service improvement.

[1906] (Application Example 2)

[1907] 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".

[1908] Traditional content delivery services have not adequately addressed individual user needs and emotions, making it difficult to improve user satisfaction. In particular, the lack of real-time content recommendations based on emotional changes during viewing, and the absence of dynamic interface adjustments, sometimes led to user stress and decreased satisfaction.

[1909] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for collecting user contract information, means for analyzing the collected contract information, means for preparing individual customized response scenarios for each user based on the analysis results, means for generating a user-friendly face image for each user, means for transmitting the generated face image and response scenario to the user terminal, means for displaying the generated face image and response scenario on the user terminal, means for collecting and analyzing user emotion data, means for dynamically recommending content based on the user's emotions, means for dynamically adjusting the interface according to the user's emotions, and means for collecting and improving feedback from the user. This enables real-time personalized content recommendation and interface adjustment in accordance with the user's emotions and needs, thereby improving user satisfaction.

[1910] "User contract information" refers to information such as the user's contract details, contract period, personal attributes, and past inquiries.

[1911] "Analysis results" refer to conclusions and insights derived by the server based on the collected data.

[1912] A "customized response scenario" refers to a pre-prepared plan and flow of individual responses based on user needs and inquiry trends.

[1913] A "face image" is a friendly, customized image of a face created using a Generative Adversarial Network (GAN).

[1914] A "user terminal" refers to a device used by a user, such as a smartphone, tablet, or personal computer.

[1915] "Emotional data" refers to information about a user's emotions, obtained in real time from their facial expressions and voice.

[1916] "A means of dynamically recommending content" refers to a technology that recommends appropriate content in real time based on user sentiment data and viewing history.

[1917] "Methods for dynamically adjusting the interface" refer to techniques that change the design and layout of an application's user interface (UI) in response to the user's emotions.

[1918] "Methods for collecting feedback and making improvements" refers to technologies that collect evaluations and opinions from users and use that information to perform data analysis and implement improvement measures to enhance future services.

[1919] This invention is a system for delivering dynamic content based on user emotions. First, a server collects user contract information and analyzes it. The analysis uses data mining techniques such as clustering and categorization. Next, based on the analysis results, a customized response scenario is prepared for each user, and a friendly facial image is generated. A Generative Adversarial Network (GAN) is used to generate this facial image. The generated facial image and response scenario are sent to the user's terminal.

[1920] On the user's terminal, the generated facial image and corresponding scenario are displayed, and the user's emotional data is collected in real time. Facial expression recognition and voice analysis technologies using a camera and microphone are employed to collect emotional data. Based on the emotional analysis results, the server dynamically recommends specific content. For example, a user experiencing stress might be recommended relaxing music or a comedy movie.

[1921] Furthermore, the application's user interface (UI) is dynamically adjusted according to the user's emotions. This means, for example, changing the UI design to a simpler and calmer one if the user is feeling stressed.

[1922] In addition, user feedback is collected and incorporated into subsequent analyses and scenario generation. This ensures continuous improvement of the service.

[1923] As a concrete example, when a user opens the application while feeling stressed, the server recommends a relaxing music playlist. The UI also automatically changes to calming colors to help reduce user stress. Afterwards, the user enters feedback on the content they viewed into a form within the application, and this data is sent to the server and used to improve the recommendation algorithm for the next time.

[1924] The program to realize this invention will be implemented using a general-purpose programming language such as Python. The libraries used will be "some_emotion_recognition_library" for emotion recognition and "some_content_recommendation_library" for content recommendation. Libraries such as "requests" and "SQLAlchemy" will be used for HTTP requests and database access.

[1925] Examples of prompt statements to input into the generative AI model are as follows:

[1926] "Design an algorithm that recommends relaxing music and movies based on the user's viewing history data and real-time sentiment analysis data. This algorithm should also consider past rating data to recommend content that best suits the user's current mood in real time. Additionally, the UI theme should dynamically change according to the user's mood."

[1927] In this way, a system is realized that dynamically recommends content based on user emotions, adjusts the interface, and incorporates feedback.

[1928] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1929] Step 1:

[1930] The server collects user contract information from the database. Input is identification information such as the user ID, and output is user contract information (contract details, contract period, personal attributes, past inquiry history). Specifically, it executes SQL queries to extract the necessary data.

[1931] Step 2:

[1932] The server analyzes the collected contract information. The input is the contract information obtained in step 1, and the output is the analysis results regarding user needs and inquiry trends. Data analysis algorithms (e.g., clustering and categorization) are used to group the data and derive insights.

[1933] Step 3:

[1934] The server generates user-specific, customized response scenarios and friendly facial images based on the analysis results. The input is the analysis results from step 2, and the output is the customized scenario and facial image corresponding to the user. A Generative Adversarial Network (GAN) is used to generate facial images and customize pre-defined response scenarios.

[1935] Step 4:

[1936] The server sends the generated face image and corresponding scenario to the user terminal. The input is the face image and corresponding scenario generated in step 3, and the output is the face image and corresponding scenario sent to the user terminal. Data is sent using an HTTP request.

[1937] Step 5:

[1938] The user terminal displays the generated face image and a customizable scenario. The input is the face image and corresponding scenario sent from the server, and the output is a visual display that the user can confirm on their terminal. The interface displays the face image and text.

[1939] Step 6:

[1940] The user terminal collects and analyzes the user's emotional data (facial expressions and voice) in real time. The input is the user's facial and voice data, and the output is the analysis results regarding the user's emotions. Specifically, it uses a camera and microphone and an emotion recognition engine to identify emotions.

[1941] Step 7:

[1942] The server dynamically recommends the most suitable content to the user based on sentiment data. The input is the sentiment data from step 6 and the user's viewing history, and the output is the recommended content. A content recommendation algorithm is used to select the content best suited to the user's current sentiment.

[1943] Step 8:

[1944] The user's device dynamically adjusts the interface according to the user's emotions. The input is the content and emotion data recommended in step 7, and the output is the UI adjusted according to the emotion. For example, a user feeling stressed will be shown a UI with calming colors.

[1945] Step 9:

[1946] Users view content and then provide feedback. The input is the user's feedback, and the output is feedback data. By filling out the feedback form on their device, the data is sent to the server.

[1947] Step 10:

[1948] The server analyzes the collected feedback data to improve response scenarios and recommendation algorithms. The input is the feedback collected in step 9, and the output is the improved response scenarios and recommendation algorithms. Analysis based on the feedback data improves system performance.

[1949] The above steps enable a dynamic content delivery system based on user emotions.

[1950] 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.

[1951] 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.

[1952] 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 robot 414.

[1953] 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.

[1954] 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.

[1955] 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.

[1956] 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.

[1957] 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.

[1958] 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."

[1959] 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.

[1960] 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.

[1961] 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.

[1962] 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.

[1963] 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.

[1964] 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.

[1965] 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.

[1966] 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.

[1967] 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.

[1968] 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.

[1969] 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.

[1970] 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 as being incorporated by reference.

[1971] The following is further disclosed regarding the embodiments described above.

[1972] (Claim 1)

[1973] Means for collecting user contract information,

[1974] A means of analyzing the collected contract information,

[1975] A means of preparing individual customized scenarios for each user based on the analysis results,

[1976] A method for generating a user-friendly facial image for each user,

[1977] A means for transmitting the generated facial image and corresponding scenario to the user terminal,

[1978] A means for displaying a face image and corresponding scenario generated on the user terminal,

[1979] A system that includes this.

[1980] (Claim 2)

[1981] The system according to claim 1, further comprising means for analyzing the content of a user's inquiry and generating an optimal response.

[1982] (Claim 3)

[1983] The system according to claim 1, further comprising means for collecting user feedback and improving the analysis results.

[1984] "Example 1"

[1985] (Claim 1)

[1986] Means for collecting user contract information,

[1987] A means of analyzing the collected contract information,

[1988] A means of preparing individual customized scenarios for each user based on the analysis results,

[1989] A means for generating friendly facial images based on analysis results,

[1990] A means for transmitting the generated facial image and customizable scenario to the user terminal,

[1991] A means for displaying a face image generated on the user terminal and a customizable scenario,

[1992] A means of receiving inquiries via a user terminal, analyzing them, and generating the optimal response,

[1993] A system that includes this.

[1994] (Claim 2)

[1995] The system according to claim 1, further comprising means for collecting user feedback and improving the analysis results.

[1996] (Claim 3)

[1997] The system according to claim 1, comprising means for generating friendly facial images using a generative adversarial network.

[1998] "Application Example 1"

[1999] (Claim 1)

[2000] Means for collecting user contract information,

[2001] A means of analyzing the collected contract information,

[2002] A means of preparing individual customized scenarios for each user based on the analysis results,

[2003] A method for generating a user-friendly facial image for each user,

[2004] A means for transmitting the generated facial image and corresponding scenario to the user terminal,

[2005] A means for displaying a face image and corresponding scenario generated on the user terminal,

[2006] A means of acting as a dedicated support person for food services via s...

Claims

1. Means for collecting user contract information, A means of analyzing the collected contract information, A means of preparing individual customized scenarios for each user based on the analysis results, A method for generating a user-friendly facial image for each user, A means for transmitting the generated facial image and corresponding scenario to the user terminal, A means for displaying a face image and corresponding scenario generated on the user terminal, A system that includes this.

2. The system according to claim 1, further comprising means for analyzing the content of a user's inquiry and generating an optimal response.

3. The system according to claim 1, further comprising means for collecting user feedback and improving the analysis results.

Images