system

Abstract

We provide the system. [Solution] A means for receiving text data entered by the user, A means for analyzing the aforementioned text data and calculating the frequency of specific keywords or phrases, A means for identifying the user's challenges based on the aforementioned analysis results and assigning appropriate roles to a generative artificial intelligence, A means for generating response messages according to roles using the aforementioned generative artificial intelligence, Means for providing the generated response message to the user, 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, the method including the 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] In modern times, users face various problems, and means for quickly and effectively solving these problems are required. However, it is not easy for users to be aware of the problems they have and find solutions thereto, especially when specialized knowledge is required, which is difficult for users. The present invention aims to provide a system that allows users to identify their own problems and easily obtain solutions thereto.

Means for Solving the Problems

[0005] The present invention solves the above problem with a system that includes means for receiving text data entered by a user, means for analyzing the text data and calculating the frequency of specific keywords and phrases, means for identifying the user's problem based on the analysis results and assigning an appropriate role to a generative artificial intelligence, means for generating a response message according to the role using the generative artificial intelligence, and means for providing the generated response message to the user. Furthermore, by including specific advice regarding the user's problem in the generated response message, the user can recognize their own problem and obtain a solution. In addition, by calculating the frequency of keywords related to a specific category through analysis using a natural language processing algorithm and assigning roles corresponding to professionals such as doctors, counselors, and tax accountants, it is possible to provide the user with expert and highly accurate advice.

[0006] A "user" refers to a person who uses this system and is the entity that inputs text data into the system.

[0007] "Text data" refers to information in written form that a user inputs into a system, and is primarily expressed in natural language.

[0008] "Means of receiving" refers to components that have the function of receiving text data entered by the user and importing it for processing.

[0009] "Means of analysis" refers to a function that processes received text data, extracts specific keywords or phrases, and calculates their frequency.

[0010] "Generative artificial intelligence" refers to a type of artificial intelligence used by systems to generate response messages with specific roles assigned to them.

[0011] A "response message" refers to a message generated by a generative artificial intelligence system for the user, and includes advice and information regarding the user's challenges.

[0012] "Natural language processing" refers to the technology used for computers to understand, analyze, and generate human language, and is used for analyzing text data.

[0013] A "keyword" refers to a specific important word or phrase within text data, serving as a crucial clue for identifying the problem.

[0014] "Assigning a role" means setting a role for a generative artificial intelligence that mimics a person with specific expertise (e.g., a doctor, counselor, or tax accountant) based on the analysis results.

[0015] A "prompt" refers to input information that includes instructions or conditions for a generative artificial intelligence system to generate an appropriate response message.

[0016] A "problem" refers to a problem or difficulty that a user is facing, and it is something that the system identifies and provides advice on. [Brief explanation of the drawing]

[0017] [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] This 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] This 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

[0018] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0019] First, the terms used in the following description will be described.

[0020] In the following embodiments, a processor with a reference numeral (hereinafter simply referred to as "processor") may be one arithmetic unit or a combination of a plurality of arithmetic units. Also, the processor may be one type of arithmetic unit or a combination of a plurality of types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0021] 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.

[0022] 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.

[0023] 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).

[0024] 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."

[0025] [First Embodiment]

[0026] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0027] 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.

[0028] 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).

[0029] 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.

[0030] 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.

[0031] 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.

[0032] 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.

[0033] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0034] 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.

[0035] 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.

[0036] 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.

[0037] 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".

[0038] This invention is an interactive system for users to receive advice on specific problems. The system analyzes text data entered by the user and uses generative artificial intelligence to provide appropriate response messages based on the data's content, thereby effectively solving the user's problems.

[0039] System Configuration

[0040] 1. Launch the user interface

[0041] The device launches the application. The application displays a user interface that includes a chat box and a send button for the user to enter text data.

[0042] 2. Initial contact and message reception

[0043] The server sends a question to the user as its initial message. For example, a message like, "Hello! What are you interested in today?"

[0044] The user enters text data as a response and presses the send button. The entered message is sent from the terminal to the server.

[0045] 3. Analysis of text data

[0046] The server analyzes the received text data using natural language processing (NLP) algorithms. The analysis extracts specific keywords and phrases, and calculates their frequency.

[0047] For example, if the input is "I haven't been able to sleep lately," the keyword "can't sleep" will be analyzed.

[0048] 4. Determining the role of generative artificial intelligence

[0049] The server identifies the user's problem based on the analysis results. It then assigns appropriate roles to generative artificial intelligence based on frequently occurring keywords.

[0050] For example, if the keyword "can't sleep" appears frequently, the role of "doctor" might be assigned to that keyword.

[0051] 5. Generating a response message

[0052] The generative artificial intelligence generates response messages using appropriate prompts according to its assigned role. For example, a message like, "That sounds difficult. Let's try a few things to get a good night's sleep. Going to bed at the same time every night and having time to relax before bed can be effective."

[0053] 6. Providing response messages to users

[0054] The server sends a generated response message to the terminal. The message is displayed in the user interface and can be viewed by the user.

[0055] Specific example

[0056] Example 1: Health-related cases

[0057] The server sends the first message to the user: "Hello! What are you interested in today?"

[0058] User response: "I've been having trouble sleeping lately."

[0059] The server analyzes the text data. It extracts the keyword "can't sleep" and checks its frequent occurrences.

[0060] The server assigns the role of "doctor" to the generative artificial intelligence and generates prompts.

[0061] Generative AI: "That sounds difficult. Let's try a few methods to get a good night's sleep. For example, going to bed at the same time every night and having time to relax before bed can be effective."

[0062] The server sends a response message to the user, which is displayed on the terminal.

[0063] Example 2: Cases related to finance

[0064] The server sends the user an initial message: "What are you interested in today?"

[0065] User response: "I'm worried because I haven't been managing my household finances well lately."

[0066] The server analyzes the text data and extracts the keywords "household finances" and "anxiety."

[0067] The server assigns the role of "tax accountant" to the generative artificial intelligence and generates a prompt.

[0068] Generative AI: "Household financial management is very important, isn't it? The first step is to meticulously record your income and expenses and reduce unnecessary spending. If you need more specific advice, may I ask for more details?"

[0069] The server sends a response message to the user, which is displayed on the terminal.

[0070] This system allows users to instantly recognize their challenges and receive expert advice. This is expected to significantly improve users' quality of life.

[0071] The following describes the processing flow.

[0072] Step 1:

[0073] The device launches an application via the user. The application displays a user interface, including a chat box where the user can enter text data and a send button.

[0074] Step 2:

[0075] The server sends the user an initial message asking, "Hello! What are you interested in today?"

[0076] Step 3:

[0077] The user enters text data into the chat box and presses the send button. For example, they might type, "I've been having trouble sleeping lately."

[0078] Step 4:

[0079] The terminal sends the message entered by the user to the server.

[0080] Step 5:

[0081] The server saves the text data it receives.

[0082] Step 6:

[0083] The server uses natural language processing (NLP) algorithms to analyze the received text data. Specifically, it extracts keywords and phrases from the text data and calculates their frequency of occurrence. For example, it analyzes the frequency of the keyword "can't sleep."

[0084] Step 7:

[0085] The server identifies the user's problem based on the analysis results. The analysis confirms that keywords related to "difficulty sleeping" appear frequently, and identifies "possible sleep disorder" as the user's problem.

[0086] Step 8:

[0087] The server assigns an appropriate role to the generative artificial intelligence. Based on the identified task, it assigns, for example, the role of "doctor."

[0088] Step 9:

[0089] The server generates and sends an appropriate prompt to the generative artificial intelligence. For example, the prompt might say, "The user is having trouble sleeping. Provide advice on improving sleep."

[0090] Step 10:

[0091] The generative artificial intelligence generates a response message based on the role it has been assigned. For example, it might say, "That sounds difficult. Let's try a few things to get a good night's sleep. Going to bed at the same time every night and having time to relax before bed can be effective."

[0092] Step 11:

[0093] The server sends the generated response message to the terminal.

[0094] Step 12:

[0095] The device displays the response message in the user's chat box.

[0096] Step 13:

[0097] The user receives the generated response message and can enter further questions or feedback as needed.

[0098] Step 14:

[0099] The terminal sends an additional message to the server, and the process from step 5 is repeated.

[0100] (Example 1)

[0101] 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."

[0102] In conventional conversational systems, it can be difficult for users to obtain appropriate advice for the specific problems they face. This is because the input text data is not effectively analyzed, and expert responses that meet the user's needs are not generated. Furthermore, the user interface may not be intuitive, and the flow of conversation may be interrupted, which is also a problem. Therefore, there is a need for a conversational system that can efficiently analyze the text data entered by the user and quickly provide expert and specific advice.

[0103] 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.

[0104] In this invention, the server includes means for receiving text data entered by a user, means for analyzing the text data and calculating the frequency of specific keywords or phrases, means for identifying the user's problem based on the analysis results and assigning an appropriate role to a generative artificial intelligence, means for generating a response message according to the role using the generative artificial intelligence, means for providing the generated response message to the user, means for activating a user interface, means for sending an initial message to the user, and means for displaying the response message on the user interface. This enables the user to immediately recognize their problem and quickly obtain the specific and expert advice they need.

[0105] A "user" is someone who uses this system to seek advice.

[0106] "Text data" refers to the words and sentences entered by the user, which are the objects that are analyzed by the system.

[0107] "Analysis" refers to the process of decomposing and extracting features from received text data using natural language processing algorithms.

[0108] "Keywords" are important words or phrases contained within the text data entered by the user.

[0109] "Frequency" refers to the number of times a particular keyword or phrase appears within text data.

[0110] A "problem" refers to an issue or difficulty identified based on text data entered by the user into the system.

[0111] "Generative artificial intelligence" refers to algorithms and machine learning models that have the ability to generate appropriate response messages based on input data and a specified role.

[0112] "Role" refers to the professional position or character that a generative artificial intelligence assumes when generating a response message.

[0113] A "response message" is a text containing solutions or advice generated by a generative artificial intelligence system in response to a user's problem.

[0114] A "user interface" is software that provides screen displays and input methods for users to interact with a system.

[0115] An "initial message" is a message that a server sends to a user via the user interface for the first time, such as a question or confirmation.

[0116] "Means of receiving" refers to the mechanisms and methods for importing text data entered by a user into the system.

[0117] "Means of providing" refers to the mechanisms and methods for displaying the generated response message to the user.

[0118] "Means of display" refers to mechanisms or methods for visually showing response messages on a user interface.

[0119] This invention is an interactive system for users to receive advice on specific problems. The system analyzes text data entered by the user and uses generative artificial intelligence to provide appropriate response messages based on the data's content, thereby effectively solving the user's problems.

[0120] System Configuration

[0121] Launching the user interface

[0122] The device runs a specific application (for example, "adviceApp.exe"). The application displays a user interface that includes a chat box and a send button for the user to enter text data.

[0123] Initial contact and message reception

[0124] The server immediately sends an initial message to the user after the application starts. For example, it might send a message like, "Hello! What are you interested in today?"

[0125] When a user enters text data into a text box and presses the submit button, that data is sent from the device to the server. For example, a user might enter, "I've been having trouble sleeping lately."

[0126] The terminal acquires the input data, converts it to JSON format, and sends it to the server.

[0127] Text data analysis

[0128] The server analyzes the received text data using natural language processing (NLP) algorithms. Python libraries such as "NLTK" and "spaCy" are used for the analysis.

[0129] Specifically, the server segments the text data, extracts keywords, and calculates their frequency. For example, it might extract the keyword "can't sleep" from the text "I've been having trouble sleeping lately."

[0130] Determining the Role of Generative Artificial Intelligence

[0131] The server determines the appropriate role for a generative artificial intelligence based on the frequency of keywords extracted through NLP analysis. For example, if the keyword "can't sleep" appears frequently, it will assign the role of "doctor."

[0132] The system compares a list of frequently occurring keywords with a database and then instructs the generative artificial intelligence to assign a role based on the results.

[0133] Generating a response message

[0134] The generative artificial intelligence generates response messages to the user according to its assigned role. For example, it uses a model such as OpenAI's GPT-3®.

[0135] The server generates a prompt such as, "That sounds difficult. Let's try a few things to help you get a good night's sleep. Going to bed at the same time every night and having time to relax before bed can be effective."

[0136] Providing response messages to users

[0137] The server sends the generated response message to the terminal. This message is often sent in JSON format.

[0138] The device displays received messages in the chat box. For example, a message like, "That sounds tough. Let's try a few things to help you get a good night's sleep," might be displayed.

[0139] Specific example

[0140] Example 1: Health-related cases

[0141] The server sends the first message to the user: "Hello! What are you interested in today?"

[0142] User response: "I've been having trouble sleeping lately."

[0143] The server analyzes the text data and extracts the keyword "can't sleep".

[0144] The server assigns the role of "doctor" to the generative artificial intelligence and generates prompts.

[0145] Generative AI: "That sounds difficult. Let's try a few methods to get a good night's sleep. For example, going to bed at the same time every night and having time to relax before bed can be effective."

[0146] The server sends a response message to the user, which is then displayed on the terminal.

[0147] Example 2: Cases related to finance

[0148] The server sends the user an initial message: "What are you interested in today?"

[0149] User response: "I'm worried because I haven't been managing my household finances well lately."

[0150] The server analyzes the text data and extracts the keywords "household finances" and "anxiety."

[0151] The server assigns the role of "tax accountant" to the generative artificial intelligence and generates prompts.

[0152] Generative AI: "Household financial management is very important, isn't it? The first step is to meticulously record your income and expenses and reduce unnecessary spending. If you need more specific advice, may I ask for more details?"

[0153] The server sends a response message to the user, which is then displayed on the terminal.

[0154] This invention allows users to instantly recognize their challenges and improve their quality of life by obtaining expert and specific advice.

[0155] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0156] Step 1:

[0157] The device runs a specific application (e.g., "adviceApp.exe"). The user interface launches, displaying a chat box for the user to enter text data and a send button.

[0158] Input: User-initiated application execution

[0159] Output: Display of a user interface including a chat box and a send button.

[0160] Step 2:

[0161] The server immediately sends an initial message to the user after the application starts. For example, it might display a message like, "Hello! What are you interested in today?"

[0162] Input: Application startup signal

[0163] Output: Generation and sending of the initial message

[0164] Step 3:

[0165] The user enters text data into a text box and presses the submit button. This data is sent from the terminal to the server. For example, the user might enter "I've been having trouble sleeping lately."

[0166] Input: User text input and submission operation

[0167] Output: Convert text data to JSON format and send to the server.

[0168] Step 4:

[0169] The server analyzes the received text data using natural language processing (NLP) algorithms. For example, it uses Python libraries such as "NLTK" and "spaCy" to tokenize the text data, extract keywords, and calculate their frequencies.

[0170] Input: Text data in JSON format

[0171] Output: Extracted keywords and their frequencies

[0172] Step 5:

[0173] The server determines the appropriate role for the generative artificial intelligence based on the extracted results. For example, if the keyword "can't sleep" appears frequently, it will assign the role of "doctor."

[0174] Input: Extracted keywords and their frequency

[0175] Output: Role instructions for generative artificial intelligence

[0176] Step 6:

[0177] The generative artificial intelligence generates response messages to the user according to its assigned role. For example, using OpenAI's GPT-3 model, it might generate a prompt such as, "That sounds tough. Let's try a few things to get a good night's sleep."

[0178] Input: Role instructions and generative artificial intelligence

[0179] Output: Response message

[0180] Step 7:

[0181] The server sends a generated response message to the terminal. This message is sent in JSON format.

[0182] Input: Generated response message

[0183] Output: Send a message in JSON format to the terminal.

[0184] Step 8:

[0185] The device parses the JSON-formatted message it receives and displays it in the chat box. For example, a message like, "That sounds tough. Let's try a few things to get a good night's sleep," might be displayed.

[0186] Input: Response message in JSON format

[0187] Output: Message displayed in the chat box

[0188] (Application Example 1)

[0189] 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."

[0190] Traditional content delivery services often fail to adequately recommend content based on what users have watched previously. This results in users spending a long time finding content that interests them, leading to decreased satisfaction. Furthermore, there was a lack of systems that provided real-time, specific advice to help users address their particular challenges. This inconsistent user experience negatively impacted customer satisfaction and retention rates.

[0191] 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.

[0192] In this invention, the server includes means for receiving text data entered by a user, means for analyzing the text data and calculating the frequency of specific keywords or phrases, means for identifying the user's problem based on the analysis results and assigning an appropriate role to a generative artificial intelligence, means for generating a response message according to the role using the generative artificial intelligence, means for recommending content that the user should watch next based on the content they have watched, and means for providing the generated response message to the user. By combining the user's viewing history with real-time text data analysis, it becomes possible to provide the user with more accurate content recommendations and specific advice.

[0193] "Means for receiving text data entered by the user" refers to a function for sending and receiving text and messages entered by the user using a terminal to a server.

[0194] "Means for analyzing the text data and calculating the frequency of specific keywords or phrases" refers to a function that analyzes the input text data using a natural language processing algorithm, extracts important words and phrases, and calculates their occurrence count and frequency.

[0195] "A means of identifying the user's challenges based on the analysis results and assigning an appropriate role to the generative artificial intelligence" refers to a function that identifies the problems the user faces from the analyzed keywords and phrases and assigns a role suitable for solving those problems to the generative artificial intelligence.

[0196] "Means for generating response messages according to roles using the aforementioned generative artificial intelligence" refers to a function in which the generative artificial intelligence creates an appropriate response message based on the assigned role.

[0197] "A means of recommending content to watch next based on the content a user has watched" refers to a function that analyzes highly relevant content based on a user's past viewing history and presents recommended content to watch next.

[0198] "Means for providing the generated response message to the user" refers to a function that sends the generated response message to the user's terminal and displays it.

[0199] This invention provides an interactive system for users to receive advice on specific problems, and in particular, for making recommendations based on viewing history in content distribution services. This system consists of a server, terminals, and a generative artificial intelligence (AI) model.

[0200] System Configuration

[0201] 1. Server

[0202] The server receives text data entered by the user and analyzes it. Natural language processing (NLP) algorithms are used for the analysis, and as a result, specific keywords and phrases are extracted.

[0203] The server identifies the user's problem based on the analysis results and assigns an appropriate role to the generative artificial intelligence. The response message generated by this role will be based on the user's viewing history and input text.

[0204] The server also has a function to recommend content that users should watch next, based on the content they have watched. Specifically, it analyzes and provides new content related to movies and TV shows that the user has recently watched.

[0205] 2. Terminal

[0206] The terminal provides a user interface, such as a chat box and a send button, for the user to input text data. The data entered by the user is sent to the server.

[0207] Furthermore, the terminal displays response messages sent from the server, providing an environment where users can receive real-time advice and view recommended content.

[0208] 3. Generative Artificial Intelligence

[0209] The generative artificial intelligence generates response messages based on appropriate prompts, following the roles assigned by the server. This uses OpenAI's GPT-3 model to provide specific advice and content recommendations for solving the user's problems.

[0210] The response message is generated based on a pre-configured prompt, taking into account the text data entered by the user and their viewing history.

[0211] Specific example

[0212] For example, suppose a user recently watched the movie "Inception." In this case, the server would take the following steps:

[0213] The server sends the user an initial message: "Hi! Would you like to talk about the movies you've watched recently?"

[0214] The user enters "Inception".

[0215] The server analyzes the text data and extracts the keyword "Inception".

[0216] The server assigns the role of "film critic" to the generative artificial intelligence and generates prompts.

[0217] Generative artificial intelligence generates a response message: "Since you've watched 'Inception,' I recommend 'Interstellar' or 'The Matrix' next. These films also have compelling storylines and profound philosophical themes, so please enjoy them."

[0218] The server sends the generated response message to the user, and it is displayed on the terminal.

[0219] Example of a prompt

[0220] Movies recently watched by users: Inception

[0221] Please recommend some movies.

[0222] This system combines users' viewing history with real-time text data analysis to provide users with more accurate content recommendations and specific advice.

[0223] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0224] Step 1:

[0225] The device launches the application. At this stage, the user interface is displayed, providing a chat box for the user to enter text data and a send button. Input: Launch of the user interface. Output: Display of the chat box and send button.

[0226] Step 2:

[0227] The server sends an initial message to the user. For example, the message might say, "Hello! Shall we talk about the movies you've watched recently?" Input: Initial state of the application. Output: Display of the initial message.

[0228] Step 3:

[0229] The user enters text data about the title of a movie they watched or other related information, and then presses the submit button. This data is sent from the terminal to the server. Input: User's text input (e.g., Inception). Output: Text data sent to the server.

[0230] Step 4:

[0231] The server analyzes the received text data using a natural language processing (NLP) algorithm. This analysis extracts specific keywords and phrases, and calculates their frequency. Input: Title of a movie watched (e.g., Inception). Output: Extracted keywords and their frequencies.

[0232] Step 5:

[0233] The server identifies the user's problem based on the analysis results and assigns an appropriate role to the generative artificial intelligence. For example, if the keyword "Inception" is extracted, the role of "film critic" will be assigned to the generative artificial intelligence. Input: Keyword analysis results. Output: Role assignment to the generative artificial intelligence.

[0234] Step 6:

[0235] A generative artificial intelligence generates a response message using appropriate prompts based on the assigned role. This response message includes specific advice and recommended content for the user. Input: Role assignment and prompt. Output: Response message (e.g., "Since you watched 'Inception,' we recommend 'Interstellar' or 'The Matrix' next").

[0236] Step 7:

[0237] The server sends a generated response message to the terminal and displays it on the user interface. The user views this message and is helped to decide what content to view next. Input: Response message. Output: Message displayed on the user interface.

[0238] This series of processing steps allows users to receive appropriate content recommendations and specific advice in real time, based on their viewing history. Furthermore, the use of generative artificial intelligence provides a deeper understanding of and solutions to the user's challenges.

[0239] 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.

[0240] This invention is an interactive system for users to receive advice on specific issues and emotions. The system analyzes text data entered by the user and the user's emotional information, and uses generative artificial intelligence to provide appropriate response messages based on the content, thereby effectively resolving the user's issues and emotional problems.

[0241] System Configuration

[0242] 1. Launch the user interface

[0243] The device launches the application. The application displays a user interface, including a chat box and a send button for the user to enter text data.

[0244] 2. Initial contact and message reception

[0245] The server sends a question to the user as its initial message. For example, a message like, "Hello! What are you interested in today?"

[0246] The user enters text data as a response and presses the send button. For example, they might enter, "I've been having trouble sleeping lately."

[0247] 3. Analysis of text data

[0248] The server analyzes the received text data using natural language processing (NLP) algorithms. The analysis extracts specific keywords and phrases, and calculates their frequency. For example, it analyzes the frequency of the keyword "can't sleep."

[0249] 4. Emotion recognition by an emotion engine

[0250] During the process of analyzing text data, the server uses an emotion engine to recognize the user's emotions. It extracts emotions such as "anxiety" and "stress" from the user's input.

[0251] 5. Determining the role of generative artificial intelligence

[0252] The server identifies the user's issues and emotions based on the analysis results and emotion recognition results. From the analysis results and emotion information, "possible sleep disorder" is identified as an issue and "anxiety" as an emotion.

[0253] The server assigns an appropriate role to the generative artificial intelligence. For example, it might assign the role of "doctor."

[0254] 6. Generating a response message

[0255] The generative artificial intelligence generates response messages using appropriate prompts based on its assigned role. For example, "That sounds difficult. Let's try a few things to get a good night's sleep. Going to bed at the same time every night and having time to relax before bed can be effective."

[0256] Taking the results of emotion recognition into consideration, for example, for "anxiety," add a message to care for those emotions, such as, "Also, try breathing exercises and light exercise to help you relax. You are not alone, we support you."

[0257] 7. Providing response messages to users

[0258] The server sends a generated response message to the terminal. The message is displayed in the user interface and can be viewed by the user.

[0259] Specific example

[0260] Example 1: Health-related cases (using the emotion engine)

[0261] The server sends the first message to the user: "Hello! What are you interested in today?"

[0262] User response: "I've been having trouble sleeping lately, and it's making me a little anxious."

[0263] The server analyzes the text data. It extracts the keyword "can't sleep" and recognizes the emotion "anxiety."

[0264] The server assigns the role of "doctor" to the generative artificial intelligence and generates prompts.

[0265] Generative AI: "That sounds difficult. Let's try a few things to help you get a good night's sleep. For example, going to bed at the same time every night and having time to relax before bed can be effective. You can also try breathing exercises or light exercise to help you relax. You are not alone, we are here to support you."

[0266] The server sends a response message to the user, which is displayed on the terminal.

[0267] Example 2: Financial-related cases (using the emotion engine)

[0268] The server sends the user an initial message: "What are you interested in today?"

[0269] User response: "I'm worried because I haven't been managing my household finances well lately."

[0270] The server analyzes the text data. It extracts the keywords "household finances" and "anxiety" and recognizes the emotion of "anxiety."

[0271] The server assigns the role of "tax accountant" to the generative artificial intelligence and generates a prompt.

[0272] Generative AI: "Household financial management is very important, isn't it? The first step is to meticulously record your income and expenses and reduce unnecessary spending. If you need more specific advice, may I ask for more details? Also, please feel free to contact me anytime to reduce any anxiety you may feel during the process."

[0273] The server sends a response message to the user, which is displayed on the terminal.

[0274] In this way, users can instantly receive advice tailored to specific challenges and emotions. By adding an emotion engine, it becomes possible to provide advice that takes users' emotions into consideration, leading to higher satisfaction.

[0275] The following describes the processing flow.

[0276] Step 1:

[0277] The device launches an application via the user. The application displays a user interface, including a chat box where the user can enter text data and a send button.

[0278] Step 2:

[0279] The server sends the user an initial message asking, "Hello! What are you interested in today?"

[0280] Step 3:

[0281] The user inputs text data into the chat box and presses the send button. For example, input "Recently, I often have trouble sleeping and feel a bit anxious."

[0282] Step 4:

[0283] The terminal sends the message input by the user to the server.

[0284] Step 5:

[0285] The server saves the received text data.

[0286] Step 6:

[0287] The server uses a natural language processing (NLP) algorithm to analyze the received text data. Specifically, it extracts keywords and phrases in the text data and calculates their frequencies. For example, it analyzes the frequency of the keyword "insomnia".

[0288] Step 7:

[0289] The server uses an emotion engine to recognize the user's emotion from the text data. Specifically, it extracts emotions such as "anxiety" and "stress".

[0290] Step 8:

[0291] Based on the analysis results and emotion recognition results, the server identifies the user's issues and emotions. For example, from the analysis results, "the possibility of sleep disorder" is identified as an issue, and "anxiety" is identified as an emotion.

[0292] Step 9:

[0293] The server assigns an appropriate role to the generative artificial intelligence. For example, it assigns the role of "doctor" based on the issues and emotions.

[0294] Step 10:

[0295] The server generates and sends an appropriate prompt to the generative artificial intelligence. For example, it might generate a prompt that says, "The user is having trouble sleeping. Provide advice on improving sleep."

[0296] Step 11:

[0297] The generative artificial intelligence generates a response message based on the role it has been assigned. For example, it might say, "That sounds difficult. Let's try a few things to get a good night's sleep. Going to bed at the same time every night and having time to relax before bed can be effective. You can also try breathing exercises and light exercise to help you relax. You are not alone, we support you."

[0298] Step 12:

[0299] The server sends the generated response message to the terminal.

[0300] Step 13:

[0301] The device displays the response message in the user's chat box.

[0302] Step 14:

[0303] The user receives the generated response message and can enter further questions or feedback as needed.

[0304] Step 15:

[0305] The terminal sends an additional message to the server, and the process from step 5 is repeated.

[0306] This allows users to receive advice on specific issues and emotions, and the system can provide responses that are appropriate to the user's situation.

[0307] (Example 2)

[0308] Next, Example 2 will be described. In the following description, the data processing device 12 is referred to as a "server", and the smart device 14 is referred to as a "terminal".

[0309] In recent years, there has been a demand for a dialogue system that can provide appropriate advice for specific problems and feelings of users. However, in the conventional dialogue system, there is a problem that it is difficult to recognize emotions in addition to analyzing the user's text data and provide appropriate advice. In particular, the advice provided without considering the user's emotion information is insufficient for the user and the satisfaction is low.

[0310] The specific processing by the specific processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving text data input by the user, means for analyzing the text data and calculating the frequency of specific keywords and phrases, means for classifying emotions for recognizing the user's emotions from the analyzed text data, means for specifying the user's problems based on the analysis result and the emotion recognition result and assigning an appropriate role to the generative artificial intelligence, means for generating a response message according to the role using the generative artificial intelligence, and means for providing the generated response message to the user. Thereby, it becomes possible to provide appropriate advice considering the emotional aspect for the specific problems of the user.

[0311] The "means for receiving text data input by the user" refers to a function for the server to receive the text data input by the user using the terminal.

[0312] The "means for calculating the frequency of specific keywords and phrases" refers to a function for analyzing the received text data and measuring the number of occurrences of preset keywords and phrases.

[0313] "Emotion classification means" refers to a function that uses algorithms or models to identify and classify a user's emotions during the process of analyzing text data.

[0314] "Means for assigning appropriate roles to generative artificial intelligence" refers to a function for setting appropriate roles (e.g., doctor, tax accountant, etc.) to generative artificial intelligence based on analysis results and emotion recognition results.

[0315] "Means for generating response messages" refers to a function that uses generative artificial intelligence to create response messages according to the assigned role.

[0316] "Means of providing a response message to the user" refers to a function that sends the generated response message to the user's terminal and displays it on the user interface.

[0317] This invention is an interactive system for users to receive advice on specific issues and emotions. The system analyzes text data entered by the user and the user's emotional information, and uses generative artificial intelligence to provide appropriate response messages based on that analysis.

[0318] System Configuration

[0319] Launching the user interface

[0320] The device launches the application. The application displays a user interface, including a chat box and a send button for the user to enter text data.

[0321] Sending and receiving the first message

[0322] The server sends the first message. The message says, "Hello! What are you interested in today?"

[0323] The user responds to this message by entering text data such as, "I've been having trouble sleeping lately," and then presses the send button.

[0324] Receiving and analyzing text data

[0325] The server receives text data sent by the user. The received data is analyzed using natural language processing (NLP) algorithms. Here, NLP libraries such as Python's NLTK and spaCy are used to extract the keyword "can't sleep".

[0326] Emotion recognition by an emotion engine

[0327] During the process of analyzing text data, the server uses an emotion engine to recognize the user's emotions. Emotion classification models (such as BERT or RoBERTa) are used to extract emotions like "anxiety" and "stress." Specifically, the text is input to the emotion classification model, and emotion labels are returned as output.

[0328] Identifying the issues and emotions, and determining the role of generative artificial intelligence.

[0329] The server identifies the user's issues and emotions based on the analysis results and emotion recognition results. For example, it might identify "potential sleep disorder" and "anxiety."

[0330] The server assigns the role of "doctor" to the generative artificial intelligence. Specifically, it selects the appropriate role from an internal database and provides the role information to the generative AI model as part of a prompt.

[0331] Generating a response message

[0332] The generative artificial intelligence creates a prompt message according to its assigned role. Here, for example, GPT-3 or a similar generative AI model is used. The prompt message would be something like, "The user is feeling anxious because they can't sleep. Please provide advice from a doctor's perspective."

[0333] Generative artificial intelligence generates a response message based on the input prompt. Based on that prompt, it creates content such as, "That sounds difficult. Let's try a few methods to get a good night's sleep. For example, going to bed at the same time every night or taking some time to relax before bed can be effective."

[0334] The generative artificial intelligence also takes emotion recognition results into consideration and adds the message, "Also, try breathing exercises and light exercise to help you relax. You are not alone; we support you."

[0335] Providing a response message to the user

[0336] The server sends the generated response message to the terminal. Specifically, the message data is sent to the user interface via the HTTP response.

[0337] The device displays received messages in the user interface. Users can see the response messages in the chat box on the screen.

[0338] Specific example

[0339] Example 1: Health-related cases

[0340] The server sends the first message to the user: "Hello! What are you interested in today?"

[0341] User response: "I've been having trouble sleeping lately, and it's making me a little anxious."

[0342] The server analyzes the text data, extracts the keyword "can't sleep," and recognizes the emotion of "anxiety."

[0343] The server assigns the role of "doctor" to the generative artificial intelligence and generates prompts.

[0344] The generative artificial intelligence created the following message: "That must be difficult. Let's try a few things to get a good night's sleep. For example, going to bed at the same time every night and having time to relax before bed can be effective. You can also try breathing exercises or light exercise to help you relax. You are not alone, we are here for you."

[0345] The server sends a response message to the user, which is then displayed on the terminal.

[0346] Example 2: Cases related to finance

[0347] The server sends the user an initial message: "What are you interested in today?"

[0348] User response: "I'm worried because I haven't been managing my household finances well lately."

[0349] The server analyzes the text data, extracts "household finances" and "anxiety," and recognizes the emotion of "anxiety."

[0350] The server assigns the role of "tax accountant" to the generative artificial intelligence and generates prompts.

[0351] The generative AI created the following message: "Household financial management is very important, isn't it? The first step is to meticulously record your income and expenses and reduce unnecessary spending. If you need more specific advice, may I ask for more details? Also, please feel free to contact us anytime to alleviate any anxiety you may feel during this process."

[0352] The server sends a response message to the user, which is then displayed on the terminal.

[0353] Thus, the present invention makes it possible to provide appropriate advice that takes into account the emotional aspects of the user's specific problems.

[0354] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0355] Step 1:

[0356] The device launches the application.

[0357] Input: The terminal the user is operating from.

[0358] Action: The application launches and the user interface is displayed.

[0359] Output: Screen including chat box and send button

[0360] Step 2:

[0361] The server sends the initial message.

[0362] Input: Initial message template on the server

[0363] Action: The server generates the message "Hello! What are you interested in today?" and sends it to the terminal.

[0364] Output: Initial message displayed to the user

[0365] Step 3:

[0366] The user responds to the initial message.

[0367] Input: User's text data (e.g., "I've been having trouble sleeping lately")

[0368] Action: The user enters text and presses the submit button.

[0369] Output: Text data sent to the server

[0370] Step 4:

[0371] The server receives text data sent by the user.

[0372] Input: Text data entered and submitted by the user.

[0373] Operation: The server receives text data and saves it to the database.

[0374] Output: Saved text data

[0375] Step 5:

[0376] The server analyzes the text data using natural language processing (NLP) algorithms.

[0377] Input: Saved text data

[0378] Specific operation: Use Python's NLTK or spaCy to extract specific keywords or phrases (e.g., "can't sleep").

[0379] Output: Keyword frequency data from the analysis results

[0380] Step 6:

[0381] The server uses an emotion engine to recognize the user's emotions.

[0382] Input: User's text data

[0383] Specific operation: Use emotion classification models such as BERT or RoBERTa to extract emotion labels such as "anxiety" and "stress."

[0384] Output: Sentiment labels as recognition results

[0385] Step 7:

[0386] The server identifies the user's issues and emotions based on the analysis results and emotion recognition results.

[0387] Input: Analysis results (keyword frequency data) and sentiment labels

[0388] Specific actions: The issue identified is "possible sleep disorder," and the emotion identified is "anxiety."

[0389] Output: Identification of issues and emotions

[0390] Step 8:

[0391] The server assigns appropriate roles to the generative artificial intelligence.

[0392] Input: Identification results of issues and emotions

[0393] Specific operation: Select an appropriate role (e.g., "doctor") from the internal database and provide role information to the generative artificial intelligence model.

[0394] Output: Role information assigned to the generative artificial intelligence.

[0395] Step 9:

[0396] The generative artificial intelligence generates the response message.

[0397] Input: User's identified issues, emotions, and assigned roles

[0398] Specific operation: Uses a generative AI model such as GPT-3 to generate a response message based on the prompt.

[0399] Example prompt: "The user is experiencing anxiety due to insomnia. Please provide advice from a doctor's perspective."

[0400] Output: Generated response message ("That sounds difficult. Let's try a few things to help you get a good night's sleep...")

[0401] Step 10:

[0402] The server sends the generated response message to the terminal.

[0403] Input: Generated response message

[0404] Specific operation: Message data is sent to the user interface via an HTTP response.

[0405] Output: Response message displayed on the user terminal

[0406] Step 11:

[0407] The response message received by the terminal is displayed on the user interface.

[0408] Input: Response message sent from the server

[0409] Operation: The message is displayed in the chat box and can be viewed by the user.

[0410] Output: Response message displayed to the user

[0411] This series of processes allows users to receive appropriate advice regarding specific issues and emotions.

[0412] (Application Example 2)

[0413] 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".

[0414] Conventional interactive systems often fail to adequately address user needs, as response messages generated using algorithms that identify problems in response to user input frequently do not adequately address user needs. Furthermore, responses that take into account user emotions are not provided, making it difficult to completely resolve user dissatisfaction and anxiety. As a result, improving the user experience has been challenging. This invention aims to solve this problem and provide an interactive system that can address both user problems and emotions.

[0415] 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 receiving text data input by the user, means for analyzing the text data and calculating the frequency of specific keywords and phrases, means for identifying the user's issues and emotions based on the analysis results and assigning an appropriate role to a generative artificial intelligence, means for generating a response message according to the role using the generative artificial intelligence, means for providing the generated response message to the user, and means for adding supplementary information and reassuring messages according to the user's emotions using an emotion recognition engine. This makes it possible to provide specific and useful advice regarding the user's issues and to provide appropriate care for the user's emotions.

[0416] A "user" is a person who uses an interactive system to seek advice on specific issues or emotions.

[0417] "Text data" refers to the written information entered by the user, including the user's concerns and emotions.

[0418] "Analysis" is the process of processing text data to calculate the frequency of specific keywords or phrases.

[0419] An "emotion recognition engine" is an algorithm that extracts a user's emotions from text data.

[0420] "Generative artificial intelligence" is artificial intelligence that generates appropriate response messages in response to a given prompt.

[0421] A "response message" is a message generated by a generative artificial intelligence system that addresses the user's problems and provides emotional support.

[0422] A "server" is a core computer device in a system that receives user input data, analyzes it, assigns roles, and generates and provides response messages.

[0423] A "role" refers to the specific response characteristics and areas of expertise assigned to a generative artificial intelligence.

[0424] "Supplemental information" refers to additional explanations or advice added to a response message that are tailored to the user's emotional state.

[0425] A "reassuring message" is a message provided to care for the user's emotions and offer a sense of mental reassurance.

[0426] A "natural language processing algorithm" is a computational method used to analyze text data to extract keywords and recognize emotions.

[0427] This invention is an interactive system for users to receive advice on specific issues and emotions. The system analyzes text data entered by the user and the user's emotional information, and uses generative artificial intelligence to provide appropriate response messages based on that analysis. The embodiments for carrying out this invention are described in detail below.

[0428] System Configuration

[0429] hardware

[0430] User terminal: Smartphones, tablets, personal computers, etc. These are devices used by users to input text data and view received response messages.

[0431] Server: A computer device that performs backend processing. It receives user input data, analyzes it, assigns roles, and generates and sends response messages.

[0432] software

[0433] Programming language: Python

[0434] Natural language processing algorithms: Use TensorFlow (registered trademark) or PyTorch to parse user input data.

[0435] Emotion recognition engine: Extracts emotions from text data using BERT models or similar emotion recognition models.

[0436] Generative AI Model: Using GPT-3 or similar generative models, it generates response messages tailored to the user's needs.

[0437] Web framework: Use Flask or Django to manage communication between the server and the user's device.

[0438] User Interface: We provide an intuitive interface for users using React or Vue.js.

[0439] Program processing

[0440] When the server receives text data from a user, it first analyzes the input data using a natural language processing algorithm. This analysis calculates the frequency of specific keywords and phrases, identifying the user's issues and emotions. At this stage, an emotion recognition engine is used to extract emotions from the user's text data.

[0441] Next, based on the analysis results and emotion recognition results, an appropriate role is assigned to the generative artificial intelligence. The generative artificial intelligence, once assigned a role, generates a response message based on the prompt text. At this time, the emotion recognition results are also taken into consideration, and supplementary information and reassuring messages corresponding to the user's emotions are added.

[0442] Finally, the generated response message is sent from the server to the user's terminal and displayed on the user interface. This allows the user to receive appropriate, emotionally sensitive responses along with specific advice.

[0443] Specific example

[0444] For example, consider a scenario where a user enters "I want a new smartphone, but I'm unsure which one to get" in a virtual store.

[0445] The server first receives this text data and begins analysis. The analysis extracts the keywords "smartphone" and "hesitant," and the emotion recognition engine recognizes the emotion "anxiety."

[0446] Next, the prompt given to the generative AI model will be as follows:

[0447] "I want a new smartphone, but I don't know which one to get. What do you recommend?"

[0448] Generative artificial intelligence generates the following response message based on this prompt:

[0449] "For the latest models, we'll introduce some specific new models and highly-rated models. Also, certain mid-range models are highly regarded."

[0450] The following supplementary information is added to the recognized emotion of "anxiety":

[0451] "Making your first expensive purchase can be nerve-wracking, right? Please feel free to ask any questions you may have. We'll work together to address any concerns you may have."

[0452] In this way, users can gain a sense of security along with specific and helpful advice. This approach significantly improves the user experience.

[0453] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0454] Step 1:

[0455] The server receives text data entered by the user. Text data entered by the user through the device is sent to the server. For example, if a user enters "I want a new smartphone, but I don't know which one to get," that text data is sent to the server.

[0456] Step 2:

[0457] The server analyzes the received text data using natural language processing algorithms. Specifically, the server extracts specific keywords and phrases from the text data and calculates their frequency. For example, the keywords "smartphone" and "hesitant" are extracted, and their frequencies are calculated.

[0458] Step 3:

[0459] The server applies an emotion recognition engine to the analyzed text data. This engine extracts the user's emotions from the text data. For example, the emotion "anxious" might be recognized from the word "hesitant." The input is the analyzed text data, and the output is the extracted emotion information.

[0460] Step 4:

[0461] The server identifies the user's problem based on the analysis results and emotion recognition results, and assigns an appropriate role to the generative artificial intelligence. For example, if the user's problem is identified as "choosing a smartphone" and the user's emotion is identified as "anxiety," the generative artificial intelligence is assigned the role of "product advisor."

[0462] Step 5:

[0463] The server provides a prompt to a generative artificial intelligence (AI) system to generate a response message. The prompt is generated based on the user's input data and emotions. For example, if the prompt "I want a new smartphone, but I don't know which one to get. What do you recommend?" is input to the generative AI system, the response message "For the latest models, we can recommend several specific new models and highly-rated models. Also, certain mid-range models are highly rated." is output.

[0464] Step 6:

[0465] The server adds supplementary information and reassuring messages aimed at emotional support to the generated response message. For example, it might add information such as, "Making your first expensive purchase can be nerve-wracking. Please feel free to ask any questions you may have. We'll work together to address your concerns."

[0466] Step 7:

[0467] The server sends the generated response message to the user's terminal and displays it on the user interface. This allows the user to receive an emotionally sensitive message along with appropriate advice. For example, the user's terminal might display a message such as, "If you're looking for the latest model, we can recommend some of the newest and most highly-rated models. We also have some highly-rated mid-range models. Making your first expensive purchase can be nerve-wracking, so please feel free to ask any questions you may have. We'll help you resolve your concerns."

[0468] 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.

[0469] 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.

[0470] 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.

[0471] [Second Embodiment]

[0472] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0473] 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.

[0474] 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).

[0475] 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.

[0476] 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.

[0477] 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).

[0478] 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.

[0479] 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.

[0480] 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.

[0481] 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.

[0482] 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.

[0483] 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".

[0484] This invention is an interactive system for users to receive advice on specific problems. The system analyzes text data entered by the user and uses generative artificial intelligence to provide appropriate response messages based on the data's content, thereby effectively solving the user's problems.

[0485] System Configuration

[0486] 1. Launch the user interface

[0487] The device launches the application. The application displays a user interface that includes a chat box and a send button for the user to enter text data.

[0488] 2. Initial contact and message reception

[0489] The server sends a question to the user as its initial message. For example, a message like, "Hello! What are you interested in today?"

[0490] The user enters text data as a response and presses the send button. The entered message is sent from the terminal to the server.

[0491] 3. Analysis of text data

[0492] The server analyzes the received text data using natural language processing (NLP) algorithms. The analysis extracts specific keywords and phrases, and calculates their frequency.

[0493] For example, if the input is "I haven't been able to sleep lately," the keyword "can't sleep" will be analyzed.

[0494] 4. Determining the role of generative artificial intelligence

[0495] The server identifies the user's problem based on the analysis results. It then assigns appropriate roles to generative artificial intelligence based on frequently occurring keywords.

[0496] For example, if the keyword "can't sleep" appears frequently, the role of "doctor" might be assigned to that keyword.

[0497] 5. Generating a response message

[0498] The generative artificial intelligence generates response messages using appropriate prompts according to its assigned role. For example, a message like, "That sounds difficult. Let's try a few things to get a good night's sleep. Going to bed at the same time every night and having time to relax before bed can be effective."

[0499] 6. Providing response messages to users

[0500] The server sends a generated response message to the terminal. The message is displayed in the user interface and can be viewed by the user.

[0501] Specific example

[0502] Example 1: Health-related cases

[0503] The server sends the first message to the user: "Hello! What are you interested in today?"

[0504] User response: "I've been having trouble sleeping lately."

[0505] The server analyzes the text data. It extracts the keyword "can't sleep" and checks its frequent occurrences.

[0506] The server assigns the role of "doctor" to the generative artificial intelligence and generates prompts.

[0507] Generative AI: "That sounds difficult. Let's try a few methods to get a good night's sleep. For example, going to bed at the same time every night and having time to relax before bed can be effective."

[0508] The server sends a response message to the user, which is displayed on the terminal.

[0509] Example 2: Cases related to finance

[0510] The server sends the user an initial message: "What are you interested in today?"

[0511] User response: "I'm worried because I haven't been managing my household finances well lately."

[0512] The server analyzes the text data and extracts the keywords "household finances" and "anxiety."

[0513] The server assigns the role of "tax accountant" to the generative artificial intelligence and generates a prompt.

[0514] Generative AI: "Household financial management is very important, isn't it? The first step is to meticulously record your income and expenses and reduce unnecessary spending. If you need more specific advice, may I ask for more details?"

[0515] The server sends a response message to the user, which is displayed on the terminal.

[0516] This system allows users to instantly recognize their challenges and receive expert advice. This is expected to significantly improve users' quality of life.

[0517] The following describes the processing flow.

[0518] Step 1:

[0519] The device launches an application via the user. The application displays a user interface, including a chat box where the user can enter text data and a send button.

[0520] Step 2:

[0521] The server sends the user an initial message asking, "Hello! What are you interested in today?"

[0522] Step 3:

[0523] The user enters text data into the chat box and presses the send button. For example, they might type, "I've been having trouble sleeping lately."

[0524] Step 4:

[0525] The terminal sends the message entered by the user to the server.

[0526] Step 5:

[0527] The server saves the text data it receives.

[0528] Step 6:

[0529] The server uses natural language processing (NLP) algorithms to analyze the received text data. Specifically, it extracts keywords and phrases from the text data and calculates their frequency of occurrence. For example, it analyzes the frequency of the keyword "can't sleep."

[0530] Step 7:

[0531] The server identifies the user's problem based on the analysis results. The analysis confirms that keywords related to "difficulty sleeping" appear frequently, and identifies "possible sleep disorder" as the user's problem.

[0532] Step 8:

[0533] The server assigns an appropriate role to the generative artificial intelligence. Based on the identified task, it assigns, for example, the role of "doctor."

[0534] Step 9:

[0535] The server generates and sends an appropriate prompt to the generative artificial intelligence. For example, the prompt might say, "The user is having trouble sleeping. Provide advice on improving sleep."

[0536] Step 10:

[0537] The generative artificial intelligence generates a response message based on the role it has been assigned. For example, it might say, "That sounds difficult. Let's try a few things to get a good night's sleep. Going to bed at the same time every night and having time to relax before bed can be effective."

[0538] Step 11:

[0539] The server sends the generated response message to the terminal.

[0540] Step 12:

[0541] The device displays the response message in the user's chat box.

[0542] Step 13:

[0543] The user receives the generated response message and can enter further questions or feedback as needed.

[0544] Step 14:

[0545] The terminal sends an additional message to the server, and the process from step 5 is repeated.

[0546] (Example 1)

[0547] 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."

[0548] In conventional conversational systems, it can be difficult for users to obtain appropriate advice for the specific problems they face. This is because the input text data is not effectively analyzed, and expert responses that meet the user's needs are not generated. Furthermore, the user interface may not be intuitive, and the flow of conversation may be interrupted, which is also a problem. Therefore, there is a need for a conversational system that can efficiently analyze the text data entered by the user and quickly provide expert and specific advice.

[0549] 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.

[0550] In this invention, the server includes means for receiving text data entered by a user, means for analyzing the text data and calculating the frequency of specific keywords or phrases, means for identifying the user's problem based on the analysis results and assigning an appropriate role to a generative artificial intelligence, means for generating a response message according to the role using the generative artificial intelligence, means for providing the generated response message to the user, means for activating a user interface, means for sending an initial message to the user, and means for displaying the response message on the user interface. This enables the user to immediately recognize their problem and quickly obtain the specific and expert advice they need.

[0551] A "user" is someone who uses this system to seek advice.

[0552] "Text data" refers to the words and sentences entered by the user, which are the objects that are analyzed by the system.

[0553] "Analysis" refers to the process of decomposing and extracting features from received text data using natural language processing algorithms.

[0554] "Keywords" are important words or phrases contained within the text data entered by the user.

[0555] "Frequency" refers to the number of times a particular keyword or phrase appears within text data.

[0556] A "problem" refers to an issue or difficulty identified based on text data entered by the user into the system.

[0557] "Generative artificial intelligence" refers to algorithms and machine learning models that have the ability to generate appropriate response messages based on input data and a specified role.

[0558] "Role" refers to the professional position or character that a generative artificial intelligence assumes when generating a response message.

[0559] A "response message" is a text containing solutions or advice generated by a generative artificial intelligence system in response to a user's problem.

[0560] A "user interface" is software that provides screen displays and input methods for users to interact with a system.

[0561] An "initial message" is a message that a server sends to a user via the user interface for the first time, such as a question or confirmation.

[0562] "Means of receiving" refers to the mechanisms and methods for importing text data entered by a user into the system.

[0563] "Means of providing" refers to the mechanisms and methods for displaying the generated response message to the user.

[0564] "Means of display" refers to mechanisms or methods for visually showing response messages on a user interface.

[0565] This invention is an interactive system for users to receive advice on specific problems. The system analyzes text data entered by the user and uses generative artificial intelligence to provide appropriate response messages based on the data's content, thereby effectively solving the user's problems.

[0566] System Configuration

[0567] Launching the user interface

[0568] The device runs a specific application (for example, "adviceApp.exe"). The application displays a user interface that includes a chat box and a send button for the user to enter text data.

[0569] Initial contact and message reception

[0570] The server immediately sends an initial message to the user after the application starts. For example, it might send a message like, "Hello! What are you interested in today?"

[0571] When a user enters text data into a text box and presses the submit button, that data is sent from the device to the server. For example, a user might enter, "I've been having trouble sleeping lately."

[0572] The terminal acquires the input data, converts it to JSON format, and sends it to the server.

[0573] Text data analysis

[0574] The server analyzes the received text data using natural language processing (NLP) algorithms. Python libraries such as "NLTK" and "spaCy" are used for the analysis.

[0575] Specifically, the server segments the text data, extracts keywords, and calculates their frequency. For example, it might extract the keyword "can't sleep" from the text "I've been having trouble sleeping lately."

[0576] Determining the Role of Generative Artificial Intelligence

[0577] The server determines the appropriate role for a generative artificial intelligence based on the frequency of keywords extracted through NLP analysis. For example, if the keyword "can't sleep" appears frequently, it will assign the role of "doctor."

[0578] The system compares a list of frequently occurring keywords with a database and then instructs the generative artificial intelligence to assign a role based on the results.

[0579] Generating a response message

[0580] The generative artificial intelligence generates response messages to the user according to its assigned role. For example, it might use a model such as OpenAI's GPT-3.

[0581] The server generates a prompt such as, "That sounds difficult. Let's try a few things to help you get a good night's sleep. Going to bed at the same time every night and having time to relax before bed can be effective."

[0582] Providing response messages to users

[0583] The server sends the generated response message to the terminal. This message is often sent in JSON format.

[0584] The device displays received messages in the chat box. For example, a message like, "That sounds tough. Let's try a few things to help you get a good night's sleep," might be displayed.

[0585] Specific example

[0586] Example 1: Health-related cases

[0587] The server sends the first message to the user: "Hello! What are you interested in today?"

[0588] User response: "I've been having trouble sleeping lately."

[0589] The server analyzes the text data and extracts the keyword "can't sleep".

[0590] The server assigns the role of "doctor" to the generative artificial intelligence and generates prompts.

[0591] Generative AI: "That sounds difficult. Let's try a few methods to get a good night's sleep. For example, going to bed at the same time every night and having time to relax before bed can be effective."

[0592] The server sends a response message to the user, which is then displayed on the terminal.

[0593] Example 2: Cases related to finance

[0594] The server sends the user an initial message: "What are you interested in today?"

[0595] User response: "I'm worried because I haven't been managing my household finances well lately."

[0596] The server analyzes the text data and extracts the keywords "household finances" and "anxiety."

[0597] The server assigns the role of "tax accountant" to the generative artificial intelligence and generates prompts.

[0598] Generative AI: "Household financial management is very important, isn't it? The first step is to meticulously record your income and expenses and reduce unnecessary spending. If you need more specific advice, may I ask for more details?"

[0599] The server sends a response message to the user, which is then displayed on the terminal.

[0600] This invention allows users to instantly recognize their challenges and improve their quality of life by obtaining expert and specific advice.

[0601] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0602] Step 1:

[0603] The device runs a specific application (e.g., "adviceApp.exe"). The user interface launches, displaying a chat box for the user to enter text data and a send button.

[0604] Input: User-initiated application execution

[0605] Output: Display of a user interface including a chat box and a send button.

[0606] Step 2:

[0607] The server immediately sends an initial message to the user after the application starts. For example, it might display a message like, "Hello! What are you interested in today?"

[0608] Input: Application startup signal

[0609] Output: Generation and sending of the initial message

[0610] Step 3:

[0611] The user enters text data into a text box and presses the submit button. This data is sent from the terminal to the server. For example, the user might enter "I've been having trouble sleeping lately."

[0612] Input: User text input and submission operation

[0613] Output: Convert text data to JSON format and send to the server.

[0614] Step 4:

[0615] The server analyzes the received text data using natural language processing (NLP) algorithms. For example, it uses Python libraries such as "NLTK" and "spaCy" to tokenize the text data, extract keywords, and calculate their frequencies.

[0616] Input: Text data in JSON format

[0617] Output: Extracted keywords and their frequencies

[0618] Step 5:

[0619] The server determines the appropriate role for the generative artificial intelligence based on the extracted results. For example, if the keyword "can't sleep" appears frequently, it will assign the role of "doctor."

[0620] Input: Extracted keywords and their frequency

[0621] Output: Role instructions for generative artificial intelligence

[0622] Step 6:

[0623] The generative artificial intelligence generates response messages to the user according to its assigned role. For example, using OpenAI's GPT-3 model, it might generate a prompt such as, "That sounds tough. Let's try a few things to get a good night's sleep."

[0624] Input: Role instructions and generative artificial intelligence

[0625] Output: Response message

[0626] Step 7:

[0627] The server sends a generated response message to the terminal. This message is sent in JSON format.

[0628] Input: Generated response message

[0629] Output: Send a message in JSON format to the terminal.

[0630] Step 8:

[0631] The device parses the JSON-formatted message it receives and displays it in the chat box. For example, a message like, "That sounds tough. Let's try a few things to get a good night's sleep," might be displayed.

[0632] Input: Response message in JSON format

[0633] Output: Message displayed in the chat box

[0634] (Application Example 1)

[0635] 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."

[0636] Traditional content delivery services often fail to adequately recommend content based on what users have watched previously. This results in users spending a long time finding content that interests them, leading to decreased satisfaction. Furthermore, there was a lack of systems that provided real-time, specific advice to help users address their particular challenges. This inconsistent user experience negatively impacted customer satisfaction and retention rates.

[0637] 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.

[0638] In this invention, the server includes means for receiving text data entered by a user, means for analyzing the text data and calculating the frequency of specific keywords or phrases, means for identifying the user's problem based on the analysis results and assigning an appropriate role to a generative artificial intelligence, means for generating a response message according to the role using the generative artificial intelligence, means for recommending content that the user should watch next based on the content they have watched, and means for providing the generated response message to the user. By combining the user's viewing history with real-time text data analysis, it becomes possible to provide the user with more accurate content recommendations and specific advice.

[0639] "Means for receiving text data entered by the user" refers to a function for sending and receiving text and messages entered by the user using a terminal to a server.

[0640] "Means for analyzing the text data and calculating the frequency of specific keywords or phrases" refers to a function that analyzes the input text data using a natural language processing algorithm, extracts important words and phrases, and calculates their occurrence count and frequency.

[0641] "A means of identifying the user's challenges based on the analysis results and assigning an appropriate role to the generative artificial intelligence" refers to a function that identifies the problems the user faces from the analyzed keywords and phrases and assigns a role suitable for solving those problems to the generative artificial intelligence.

[0642] "Means for generating response messages according to roles using the aforementioned generative artificial intelligence" refers to a function in which the generative artificial intelligence creates an appropriate response message based on the assigned role.

[0643] "A means of recommending content to watch next based on the content a user has watched" refers to a function that analyzes highly relevant content based on a user's past viewing history and presents recommended content to watch next.

[0644] "Means for providing the generated response message to the user" refers to a function that sends the generated response message to the user's terminal and displays it.

[0645] This invention provides an interactive system for users to receive advice on specific problems, and in particular, for making recommendations based on viewing history in content distribution services. This system consists of a server, terminals, and a generative artificial intelligence (AI) model.

[0646] System Configuration

[0647] 1. Server

[0648] The server receives text data entered by the user and analyzes it. Natural language processing (NLP) algorithms are used for the analysis, and as a result, specific keywords and phrases are extracted.

[0649] The server identifies the user's problem based on the analysis results and assigns an appropriate role to the generative artificial intelligence. The response message generated by this role will be based on the user's viewing history and input text.

[0650] The server also has a function to recommend content that users should watch next, based on the content they have watched. Specifically, it analyzes and provides new content related to movies and TV shows that the user has recently watched.

[0651] 2. Terminal

[0652] The terminal provides a user interface, such as a chat box and a send button, for the user to input text data. The data entered by the user is sent to the server.

[0653] Furthermore, the terminal displays response messages sent from the server, providing an environment where users can receive real-time advice and view recommended content.

[0654] 3. Generative Artificial Intelligence

[0655] The generative artificial intelligence generates response messages based on appropriate prompts, following the roles assigned by the server. This uses OpenAI's GPT-3 model to provide specific advice and content recommendations for solving the user's problems.

[0656] The response message is generated based on a pre-configured prompt, taking into account the text data entered by the user and their viewing history.

[0657] Specific example

[0658] For example, suppose a user recently watched the movie "Inception." In this case, the server would take the following steps:

[0659] The server sends the user an initial message: "Hi! Would you like to talk about the movies you've watched recently?"

[0660] The user enters "Inception".

[0661] The server analyzes the text data and extracts the keyword "Inception".

[0662] The server assigns the role of "film critic" to the generative artificial intelligence and generates prompts.

[0663] Generative artificial intelligence generates a response message: "Since you've watched 'Inception,' I recommend 'Interstellar' or 'The Matrix' next. These films also have compelling storylines and profound philosophical themes, so please enjoy them."

[0664] The server sends the generated response message to the user, and it is displayed on the terminal.

[0665] Example of a prompt

[0666] Movies recently watched by users: Inception

[0667] Please recommend some movies.

[0668] This system combines users' viewing history with real-time text data analysis to provide users with more accurate content recommendations and specific advice.

[0669] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0670] Step 1:

[0671] The device launches the application. At this stage, the user interface is displayed, providing a chat box for the user to enter text data and a send button. Input: Launch of the user interface. Output: Display of the chat box and send button.

[0672] Step 2:

[0673] The server sends an initial message to the user. For example, the message might say, "Hello! Shall we talk about the movies you've watched recently?" Input: Initial state of the application. Output: Display of the initial message.

[0674] Step 3:

[0675] The user enters text data about the title of a movie they watched or other related information, and then presses the submit button. This data is sent from the terminal to the server. Input: User's text input (e.g., Inception). Output: Text data sent to the server.

[0676] Step 4:

[0677] The server analyzes the received text data using a natural language processing (NLP) algorithm. This analysis extracts specific keywords and phrases, and calculates their frequency. Input: Title of a movie watched (e.g., Inception). Output: Extracted keywords and their frequencies.

[0678] Step 5:

[0679] The server identifies the user's problem based on the analysis results and assigns an appropriate role to the generative artificial intelligence. For example, if the keyword "Inception" is extracted, the role of "film critic" will be assigned to the generative artificial intelligence. Input: Keyword analysis results. Output: Role assignment to the generative artificial intelligence.

[0680] Step 6:

[0681] A generative artificial intelligence generates a response message using appropriate prompts based on the assigned role. This response message includes specific advice and recommended content for the user. Input: Role assignment and prompt. Output: Response message (e.g., "Since you watched 'Inception,' we recommend 'Interstellar' or 'The Matrix' next").

[0682] Step 7:

[0683] The server sends a generated response message to the terminal and displays it on the user interface. The user views this message and is helped to decide what content to view next. Input: Response message. Output: Message displayed on the user interface.

[0684] This series of processing steps allows users to receive appropriate content recommendations and specific advice in real time, based on their viewing history. Furthermore, the use of generative artificial intelligence provides a deeper understanding of and solutions to the user's challenges.

[0685] 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.

[0686] This invention is an interactive system for users to receive advice on specific issues and emotions. The system analyzes text data entered by the user and the user's emotional information, and uses generative artificial intelligence to provide appropriate response messages based on the content, thereby effectively resolving the user's issues and emotional problems.

[0687] System Configuration

[0688] 1. Launch the user interface

[0689] The device launches the application. The application displays a user interface, including a chat box and a send button for the user to enter text data.

[0690] 2. Initial contact and message reception

[0691] The server sends a question to the user as its initial message. For example, a message like, "Hello! What are you interested in today?"

[0692] The user enters text data as a response and presses the send button. For example, they might enter, "I've been having trouble sleeping lately."

[0693] 3. Analysis of text data

[0694] The server analyzes the received text data using natural language processing (NLP) algorithms. The analysis extracts specific keywords and phrases, and calculates their frequency. For example, it analyzes the frequency of the keyword "can't sleep."

[0695] 4. Emotion recognition by an emotion engine

[0696] During the process of analyzing text data, the server uses an emotion engine to recognize the user's emotions. It extracts emotions such as "anxiety" and "stress" from the user's input.

[0697] 5. Determining the role of generative artificial intelligence

[0698] The server identifies the user's issues and emotions based on the analysis results and emotion recognition results. From the analysis results and emotion information, "possible sleep disorder" is identified as an issue and "anxiety" as an emotion.

[0699] The server assigns an appropriate role to the generative artificial intelligence. For example, it might assign the role of "doctor."

[0700] 6. Generating a response message

[0701] The generative artificial intelligence generates response messages using appropriate prompts based on its assigned role. For example, "That sounds difficult. Let's try a few things to get a good night's sleep. Going to bed at the same time every night and having time to relax before bed can be effective."

[0702] Taking the results of emotion recognition into consideration, for example, for "anxiety," add a message to care for those emotions, such as, "Also, try breathing exercises and light exercise to help you relax. You are not alone, we support you."

[0703] 7. Providing response messages to users

[0704] The server sends a generated response message to the terminal. The message is displayed in the user interface and can be viewed by the user.

[0705] Specific example

[0706] Example 1: Health-related cases (using the emotion engine)

[0707] The server sends the first message to the user: "Hello! What are you interested in today?"

[0708] User response: "I've been having trouble sleeping lately, and it's making me a little anxious."

[0709] The server analyzes the text data. It extracts the keyword "can't sleep" and recognizes the emotion "anxiety."

[0710] The server assigns the role of "doctor" to the generative artificial intelligence and generates prompts.

[0711] Generative AI: "That sounds difficult. Let's try a few things to help you get a good night's sleep. For example, going to bed at the same time every night and having time to relax before bed can be effective. You can also try breathing exercises or light exercise to help you relax. You are not alone, we are here to support you."

[0712] The server sends a response message to the user, which is displayed on the terminal.

[0713] Example 2: Financial-related cases (using the emotion engine)

[0714] The server sends the user an initial message: "What are you interested in today?"

[0715] User response: "I'm worried because I haven't been managing my household finances well lately."

[0716] The server analyzes the text data. It extracts the keywords "household finances" and "anxiety" and recognizes the emotion of "anxiety."

[0717] The server assigns the role of "tax accountant" to the generative artificial intelligence and generates a prompt.

[0718] Generative AI: "Household financial management is very important, isn't it? The first step is to meticulously record your income and expenses and reduce unnecessary spending. If you need more specific advice, may I ask for more details? Also, please feel free to contact me anytime to reduce any anxiety you may feel during the process."

[0719] The server sends a response message to the user, which is displayed on the terminal.

[0720] In this way, users can instantly receive advice tailored to specific challenges and emotions. By adding an emotion engine, it becomes possible to provide advice that takes users' emotions into consideration, leading to higher satisfaction.

[0721] The following describes the processing flow.

[0722] Step 1:

[0723] The device launches an application via the user. The application displays a user interface, including a chat box where the user can enter text data and a send button.

[0724] Step 2:

[0725] The server sends the user an initial message asking, "Hello! What are you interested in today?"

[0726] Step 3:

[0727] The user enters text data into the chat box and presses the send button. For example, they might type, "I've been having trouble sleeping lately, and it's making me a little anxious."

[0728] Step 4:

[0729] The terminal sends the message entered by the user to the server.

[0730] Step 5:

[0731] The server saves the text data it receives.

[0732] Step 6:

[0733] The server uses natural language processing (NLP) algorithms to analyze the received text data. Specifically, it extracts keywords and phrases from the text data and calculates their frequency of occurrence. For example, it analyzes the frequency of the keyword "can't sleep."

[0734] Step 7:

[0735] The server uses an emotion engine to recognize the user's emotions from text data. Specifically, it extracts emotions such as "anxiety" and "stress."

[0736] Step 8:

[0737] The server identifies the user's issues and emotions based on the analysis results and emotion recognition results. For example, the analysis results might identify "possible sleep disorder" as an issue, and "anxiety" as the emotion.

[0738] Step 9:

[0739] The server assigns appropriate roles to the generative artificial intelligence. For example, it might assign the role of "doctor" based on the task and emotions.

[0740] Step 10:

[0741] The server generates and sends an appropriate prompt to the generative artificial intelligence. For example, it might generate a prompt that says, "The user is having trouble sleeping. Provide advice on improving sleep."

[0742] Step 11:

[0743] The generative artificial intelligence generates a response message based on the role it has been assigned. For example, it might say, "That sounds difficult. Let's try a few things to get a good night's sleep. Going to bed at the same time every night and having time to relax before bed can be effective. You can also try breathing exercises and light exercise to help you relax. You are not alone, we support you."

[0744] Step 12:

[0745] The server sends the generated response message to the terminal.

[0746] Step 13:

[0747] The device displays the response message in the user's chat box.

[0748] Step 14:

[0749] The user receives the generated response message and can enter further questions or feedback as needed.

[0750] Step 15:

[0751] The terminal sends an additional message to the server, and the process from step 5 is repeated.

[0752] This allows users to receive advice on specific issues and emotions, and the system can provide responses that are appropriate to the user's situation.

[0753] (Example 2)

[0754] 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".

[0755] In recent years, there has been a growing demand for dialogue systems that provide appropriate advice tailored to the specific challenges and emotions users face. However, conventional dialogue systems have struggled to recognize emotions in addition to analyzing user text data, making it difficult to provide appropriate advice. In particular, advice provided without considering the user's emotional information is often insufficient and results in low user satisfaction.

[0756] The identification processing performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving text data input by the user, means for analyzing the text data and calculating the frequency of specific keywords and phrases, means for classifying emotions to recognize the user's emotions from the analyzed text data, means for identifying the user's problem based on the analysis results and emotion recognition results and assigning an appropriate role to the generative artificial intelligence, means for generating a response message according to the role using the generative artificial intelligence, and means for providing the generated response message to the user. This makes it possible to provide appropriate advice that also takes emotions into consideration for the user's specific problem.

[0757] "Means for receiving text data entered by the user" refers to the function that allows the server to receive text data entered by the user using a terminal.

[0758] "Means for calculating the frequency of specific keywords or phrases" refers to a function that analyzes received text data and measures the number of occurrences of pre-set keywords or phrases.

[0759] "Emotion classification means" refers to a function that uses algorithms or models to identify and classify a user's emotions during the process of analyzing text data.

[0760] "Means for assigning appropriate roles to generative artificial intelligence" refers to a function for setting appropriate roles (e.g., doctor, tax accountant, etc.) to generative artificial intelligence based on analysis results and emotion recognition results.

[0761] "Means for generating response messages" refers to a function that uses generative artificial intelligence to create response messages according to the assigned role.

[0762] "Means of providing a response message to the user" refers to a function that sends the generated response message to the user's terminal and displays it on the user interface.

[0763] This invention is an interactive system for users to receive advice on specific issues and emotions. The system analyzes text data entered by the user and the user's emotional information, and uses generative artificial intelligence to provide appropriate response messages based on that analysis.

[0764] System Configuration

[0765] Launching the user interface

[0766] The device launches the application. The application displays a user interface, including a chat box and a send button for the user to enter text data.

[0767] Sending and receiving the first message

[0768] The server sends the first message. The message says, "Hello! What are you interested in today?"

[0769] The user responds to this message by entering text data such as, "I've been having trouble sleeping lately," and then presses the send button.

[0770] Receiving and analyzing text data

[0771] The server receives text data sent by the user. The received data is analyzed using natural language processing (NLP) algorithms. Here, NLP libraries such as Python's NLTK and spaCy are used to extract the keyword "can't sleep".

[0772] Emotion recognition by an emotion engine

[0773] During the process of analyzing text data, the server uses an emotion engine to recognize the user's emotions. Emotion classification models (such as BERT or RoBERTa) are used to extract emotions like "anxiety" and "stress." Specifically, the text is input to the emotion classification model, and emotion labels are returned as output.

[0774] Identifying the issues and emotions, and determining the role of generative artificial intelligence.

[0775] The server identifies the user's issues and emotions based on the analysis results and emotion recognition results. For example, it might identify "potential sleep disorder" and "anxiety."

[0776] The server assigns the role of "doctor" to the generative artificial intelligence. Specifically, it selects the appropriate role from an internal database and provides the role information to the generative AI model as part of a prompt.

[0777] Generating a response message

[0778] The generative artificial intelligence creates a prompt message according to its assigned role. Here, for example, GPT-3 or a similar generative AI model is used. The prompt message would be something like, "The user is feeling anxious because they can't sleep. Please provide advice from a doctor's perspective."

[0779] Generative artificial intelligence generates a response message based on the input prompt. Based on that prompt, it creates content such as, "That sounds difficult. Let's try a few methods to get a good night's sleep. For example, going to bed at the same time every night or taking some time to relax before bed can be effective."

[0780] The generative artificial intelligence also takes emotion recognition results into consideration and adds the message, "Also, try breathing exercises and light exercise to help you relax. You are not alone; we support you."

[0781] Providing a response message to the user

[0782] The server sends the generated response message to the terminal. Specifically, the message data is sent to the user interface via the HTTP response.

[0783] The device displays received messages in the user interface. Users can see the response messages in the chat box on the screen.

[0784] Specific example

[0785] Example 1: Health-related cases

[0786] The server sends the first message to the user: "Hello! What are you interested in today?"

[0787] User response: "I've been having trouble sleeping lately, and it's making me a little anxious."

[0788] The server analyzes the text data, extracts the keyword "can't sleep," and recognizes the emotion of "anxiety."

[0789] The server assigns the role of "doctor" to the generative artificial intelligence and generates prompts.

[0790] The generative artificial intelligence created the following message: "That must be difficult. Let's try a few things to get a good night's sleep. For example, going to bed at the same time every night and having time to relax before bed can be effective. You can also try breathing exercises or light exercise to help you relax. You are not alone, we are here for you."

[0791] The server sends a response message to the user, which is then displayed on the terminal.

[0792] Example 2: Cases related to finance

[0793] The server sends the user an initial message: "What are you interested in today?"

[0794] User response: "I'm worried because I haven't been managing my household finances well lately."

[0795] The server analyzes the text data, extracts "household finances" and "anxiety," and recognizes the emotion of "anxiety."

[0796] The server assigns the role of "tax accountant" to the generative artificial intelligence and generates prompts.

[0797] The generative AI created the following message: "Household financial management is very important, isn't it? The first step is to meticulously record your income and expenses and reduce unnecessary spending. If you need more specific advice, may I ask for more details? Also, please feel free to contact us anytime to alleviate any anxiety you may feel during this process."

[0798] The server sends a response message to the user, which is then displayed on the terminal.

[0799] Thus, the present invention makes it possible to provide appropriate advice that takes into account the emotional aspects of the user's specific problems.

[0800] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0801] Step 1:

[0802] The device launches the application.

[0803] Input: The terminal the user is operating from.

[0804] Action: The application launches and the user interface is displayed.

[0805] Output: Screen including chat box and send button

[0806] Step 2:

[0807] The server sends the initial message.

[0808] Input: Initial message template on the server

[0809] Action: The server generates the message "Hello! What are you interested in today?" and sends it to the terminal.

[0810] Output: Initial message displayed to the user

[0811] Step 3:

[0812] The user responds to the initial message.

[0813] Input: User's text data (e.g., "I've been having trouble sleeping lately")

[0814] Action: The user enters text and presses the submit button.

[0815] Output: Text data sent to the server

[0816] Step 4:

[0817] The server receives text data sent by the user.

[0818] Input: Text data entered and submitted by the user.

[0819] Operation: The server receives text data and saves it to the database.

[0820] Output: Saved text data

[0821] Step 5:

[0822] The server analyzes the text data using natural language processing (NLP) algorithms.

[0823] Input: Saved text data

[0824] Specific operation: Use Python's NLTK or spaCy to extract specific keywords or phrases (e.g., "can't sleep").

[0825] Output: Keyword frequency data from the analysis results

[0826] Step 6:

[0827] The server uses an emotion engine to recognize the user's emotions.

[0828] Input: User's text data

[0829] Specific operation: Use emotion classification models such as BERT or RoBERTa to extract emotion labels such as "anxiety" and "stress."

[0830] Output: Sentiment labels as recognition results

[0831] Step 7:

[0832] The server identifies the user's issues and emotions based on the analysis results and emotion recognition results.

[0833] Input: Analysis results (keyword frequency data) and sentiment labels

[0834] Specific actions: The issue identified is "possible sleep disorder," and the emotion identified is "anxiety."

[0835] Output: Identification of issues and emotions

[0836] Step 8:

[0837] The server assigns appropriate roles to the generative artificial intelligence.

[0838] Input: Identification results of issues and emotions

[0839] Specific operation: Select an appropriate role (e.g., "doctor") from the internal database and provide role information to the generative artificial intelligence model.

[0840] Output: Role information assigned to the generative artificial intelligence.

[0841] Step 9:

[0842] The generative artificial intelligence generates the response message.

[0843] Input: User's identified issues, emotions, and assigned roles

[0844] Specific operation: Uses a generative AI model such as GPT-3 to generate a response message based on the prompt.

[0845] Example prompt: "The user is experiencing anxiety due to insomnia. Please provide advice from a doctor's perspective."

[0846] Output: Generated response message ("That sounds difficult. Let's try a few things to help you get a good night's sleep...")

[0847] Step 10:

[0848] The server sends the generated response message to the terminal.

[0849] Input: Generated response message

[0850] Specific operation: Message data is sent to the user interface via an HTTP response.

[0851] Output: Response message displayed on the user terminal

[0852] Step 11:

[0853] The response message received by the terminal is displayed on the user interface.

[0854] Input: Response message sent from the server

[0855] Operation: The message is displayed in the chat box and can be viewed by the user.

[0856] Output: Response message displayed to the user

[0857] This series of processes allows users to receive appropriate advice regarding specific issues and emotions.

[0858] (Application Example 2)

[0859] 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."

[0860] Conventional interactive systems often fail to adequately address user needs, as response messages generated using algorithms that identify problems in response to user input frequently do not adequately address user needs. Furthermore, responses that take into account user emotions are not provided, making it difficult to completely resolve user dissatisfaction and anxiety. As a result, improving the user experience has been challenging. This invention aims to solve this problem and provide an interactive system that can address both user problems and emotions.

[0861] 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 receiving text data input by the user, means for analyzing the text data and calculating the frequency of specific keywords and phrases, means for identifying the user's issues and emotions based on the analysis results and assigning an appropriate role to a generative artificial intelligence, means for generating a response message according to the role using the generative artificial intelligence, means for providing the generated response message to the user, and means for adding supplementary information and reassuring messages according to the user's emotions using an emotion recognition engine. This makes it possible to provide specific and useful advice regarding the user's issues and to provide appropriate care for the user's emotions.

[0862] A "user" is a person who uses an interactive system to seek advice on specific issues or emotions.

[0863] "Text data" refers to the written information entered by the user, including the user's concerns and emotions.

[0864] "Analysis" is the process of processing text data to calculate the frequency of specific keywords or phrases.

[0865] An "emotion recognition engine" is an algorithm that extracts a user's emotions from text data.

[0866] "Generative artificial intelligence" is artificial intelligence that generates appropriate response messages in response to a given prompt.

[0867] A "response message" is a message generated by a generative artificial intelligence system that addresses the user's problems and provides emotional support.

[0868] A "server" is a core computer device in a system that receives user input data, analyzes it, assigns roles, and generates and provides response messages.

[0869] A "role" refers to the specific response characteristics and areas of expertise assigned to a generative artificial intelligence.

[0870] "Supplemental information" refers to additional explanations or advice added to a response message that are tailored to the user's emotional state.

[0871] A "reassuring message" is a message provided to care for the user's emotions and offer a sense of mental reassurance.

[0872] A "natural language processing algorithm" is a computational method used to analyze text data to extract keywords and recognize emotions.

[0873] This invention is an interactive system for users to receive advice on specific issues and emotions. The system analyzes text data entered by the user and the user's emotional information, and uses generative artificial intelligence to provide appropriate response messages based on that analysis. The embodiments for carrying out this invention are described in detail below.

[0874] System Configuration

[0875] hardware

[0876] User terminal: Smartphones, tablets, personal computers, etc. These are devices used by users to input text data and view received response messages.

[0877] Server: A computer device that performs backend processing. It receives user input data, analyzes it, assigns roles, and generates and sends response messages.

[0878] software

[0879] Programming language: Python

[0880] Natural language processing algorithms: Use TensorFlow or PyTorch to parse user input data.

[0881] Emotion recognition engine: Extracts emotions from text data using BERT models or similar emotion recognition models.

[0882] Generative AI Model: Using GPT-3 or similar generative models, it generates response messages tailored to the user's needs.

[0883] Web framework: Use Flask or Django to manage communication between the server and the user's device.

[0884] User Interface: We provide an intuitive interface for users using React or Vue.js.

[0885] Program processing

[0886] When the server receives text data from a user, it first analyzes the input data using a natural language processing algorithm. This analysis calculates the frequency of specific keywords and phrases, identifying the user's issues and emotions. At this stage, an emotion recognition engine is used to extract emotions from the user's text data.

[0887] Next, based on the analysis results and emotion recognition results, an appropriate role is assigned to the generative artificial intelligence. The generative artificial intelligence, once assigned a role, generates a response message based on the prompt text. At this time, the emotion recognition results are also taken into consideration, and supplementary information and reassuring messages corresponding to the user's emotions are added.

[0888] Finally, the generated response message is sent from the server to the user's terminal and displayed on the user interface. This allows the user to receive appropriate, emotionally sensitive responses along with specific advice.

[0889] Specific example

[0890] For example, consider a scenario where a user enters "I want a new smartphone, but I'm unsure which one to get" in a virtual store.

[0891] The server first receives this text data and begins analysis. The analysis extracts the keywords "smartphone" and "hesitant," and the emotion recognition engine recognizes the emotion "anxiety."

[0892] Next, the prompt given to the generative AI model will be as follows:

[0893] "I want a new smartphone, but I don't know which one to get. What do you recommend?"

[0894] Generative artificial intelligence generates the following response message based on this prompt:

[0895] "For the latest models, we'll introduce some specific new models and highly-rated models. Also, certain mid-range models are highly regarded."

[0896] The following supplementary information is added to the recognized emotion of "anxiety":

[0897] "Making your first expensive purchase can be nerve-wracking, right? Please feel free to ask any questions you may have. We'll work together to address any concerns you may have."

[0898] In this way, users can gain a sense of security along with specific and helpful advice. This approach significantly improves the user experience.

[0899] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0900] Step 1:

[0901] The server receives text data entered by the user. Text data entered by the user through the device is sent to the server. For example, if a user enters "I want a new smartphone, but I don't know which one to get," that text data is sent to the server.

[0902] Step 2:

[0903] The server analyzes the received text data using natural language processing algorithms. Specifically, the server extracts specific keywords and phrases from the text data and calculates their frequency. For example, the keywords "smartphone" and "hesitant" are extracted, and their frequencies are calculated.

[0904] Step 3:

[0905] The server applies an emotion recognition engine to the analyzed text data. This engine extracts the user's emotions from the text data. For example, the emotion "anxious" might be recognized from the word "hesitant." The input is the analyzed text data, and the output is the extracted emotion information.

[0906] Step 4:

[0907] The server identifies the user's problem based on the analysis results and emotion recognition results, and assigns an appropriate role to the generative artificial intelligence. For example, if the user's problem is identified as "choosing a smartphone" and the user's emotion is identified as "anxiety," the generative artificial intelligence is assigned the role of "product advisor."

[0908] Step 5:

[0909] The server provides a prompt to a generative artificial intelligence (AI) system to generate a response message. The prompt is generated based on the user's input data and emotions. For example, if the prompt "I want a new smartphone, but I don't know which one to get. What do you recommend?" is input to the generative AI system, the response message "For the latest models, we can recommend several specific new models and highly-rated models. Also, certain mid-range models are highly rated." is output.

[0910] Step 6:

[0911] The server adds supplementary information and reassuring messages aimed at emotional support to the generated response message. For example, it might add information such as, "Making your first expensive purchase can be nerve-wracking. Please feel free to ask any questions you may have. We'll work together to address your concerns."

[0912] Step 7:

[0913] The server sends the generated response message to the user's terminal and displays it on the user interface. This allows the user to receive an emotionally sensitive message along with appropriate advice. For example, the user's terminal might display a message such as, "If you're looking for the latest model, we can recommend some of the newest and most highly-rated models. We also have some highly-rated mid-range models. Making your first expensive purchase can be nerve-wracking, so please feel free to ask any questions you may have. We'll help you resolve your concerns."

[0914] 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.

[0915] 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.

[0916] 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.

[0917] [Third Embodiment]

[0918] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0919] 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.

[0920] 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).

[0921] 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.

[0922] 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.

[0923] 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).

[0924] 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.

[0925] 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.

[0926] 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.

[0927] 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.

[0928] 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.

[0929] 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".

[0930] This invention is an interactive system for users to receive advice on specific problems. The system analyzes text data entered by the user and uses generative artificial intelligence to provide appropriate response messages based on the data's content, thereby effectively solving the user's problems.

[0931] System Configuration

[0932] 1. Launch the user interface

[0933] The device launches the application. The application displays a user interface that includes a chat box and a send button for the user to enter text data.

[0934] 2. Initial contact and message reception

[0935] The server sends a question to the user as its initial message. For example, a message like, "Hello! What are you interested in today?"

[0936] The user enters text data as a response and presses the send button. The entered message is sent from the terminal to the server.

[0937] 3. Analysis of text data

[0938] The server analyzes the received text data using natural language processing (NLP) algorithms. The analysis extracts specific keywords and phrases, and calculates their frequency.

[0939] For example, if the input is "I haven't been able to sleep lately," the keyword "can't sleep" will be analyzed.

[0940] 4. Determining the role of generative artificial intelligence

[0941] The server identifies the user's problem based on the analysis results. It then assigns appropriate roles to generative artificial intelligence based on frequently occurring keywords.

[0942] For example, if the keyword "can't sleep" appears frequently, the role of "doctor" might be assigned to that keyword.

[0943] 5. Generating a response message

[0944] The generative artificial intelligence generates response messages using appropriate prompts according to its assigned role. For example, a message like, "That sounds difficult. Let's try a few things to get a good night's sleep. Going to bed at the same time every night and having time to relax before bed can be effective."

[0945] 6. Providing response messages to users

[0946] The server sends a generated response message to the terminal. The message is displayed in the user interface and can be viewed by the user.

[0947] Specific example

[0948] Example 1: Health-related cases

[0949] The server sends the first message to the user: "Hello! What are you interested in today?"

[0950] User response: "I've been having trouble sleeping lately."

[0951] The server analyzes the text data. It extracts the keyword "can't sleep" and checks its frequent occurrences.

[0952] The server assigns the role of "doctor" to the generative artificial intelligence and generates prompts.

[0953] Generative AI: "That sounds difficult. Let's try a few methods to get a good night's sleep. For example, going to bed at the same time every night and having time to relax before bed can be effective."

[0954] The server sends a response message to the user, which is displayed on the terminal.

[0955] Example 2: Cases related to finance

[0956] The server sends the user an initial message: "What are you interested in today?"

[0957] User response: "I'm worried because I haven't been managing my household finances well lately."

[0958] The server analyzes the text data and extracts the keywords "household finances" and "anxiety."

[0959] The server assigns the role of "tax accountant" to the generative artificial intelligence and generates a prompt.

[0960] Generative AI: "Household financial management is very important, isn't it? The first step is to meticulously record your income and expenses and reduce unnecessary spending. If you need more specific advice, may I ask for more details?"

[0961] The server sends a response message to the user, which is displayed on the terminal.

[0962] This system allows users to instantly recognize their challenges and receive expert advice. This is expected to significantly improve users' quality of life.

[0963] The following describes the processing flow.

[0964] Step 1:

[0965] The device launches an application via the user. The application displays a user interface, including a chat box where the user can enter text data and a send button.

[0966] Step 2:

[0967] The server sends the user an initial message asking, "Hello! What are you interested in today?"

[0968] Step 3:

[0969] The user enters text data into the chat box and presses the send button. For example, they might type, "I've been having trouble sleeping lately."

[0970] Step 4:

[0971] The terminal sends the message entered by the user to the server.

[0972] Step 5:

[0973] The server saves the text data it receives.

[0974] Step 6:

[0975] The server uses natural language processing (NLP) algorithms to analyze the received text data. Specifically, it extracts keywords and phrases from the text data and calculates their frequency of occurrence. For example, it analyzes the frequency of the keyword "can't sleep."

[0976] Step 7:

[0977] The server identifies the user's problem based on the analysis results. The analysis confirms that keywords related to "difficulty sleeping" appear frequently, and identifies "possible sleep disorder" as the user's problem.

[0978] Step 8:

[0979] The server assigns an appropriate role to the generative artificial intelligence. Based on the identified task, it assigns, for example, the role of "doctor."

[0980] Step 9:

[0981] The server generates and sends an appropriate prompt to the generative artificial intelligence. For example, the prompt might say, "The user is having trouble sleeping. Provide advice on improving sleep."

[0982] Step 10:

[0983] The generative artificial intelligence generates a response message based on the role it has been assigned. For example, it might say, "That sounds difficult. Let's try a few things to get a good night's sleep. Going to bed at the same time every night and having time to relax before bed can be effective."

[0984] Step 11:

[0985] The server sends the generated response message to the terminal.

[0986] Step 12:

[0987] The device displays the response message in the user's chat box.

[0988] Step 13:

[0989] The user receives the generated response message and can enter further questions or feedback as needed.

[0990] Step 14:

[0991] The terminal sends an additional message to the server, and the process from step 5 is repeated.

[0992] (Example 1)

[0993] 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."

[0994] In conventional conversational systems, it can be difficult for users to obtain appropriate advice for the specific problems they face. This is because the input text data is not effectively analyzed, and expert responses that meet the user's needs are not generated. Furthermore, the user interface may not be intuitive, and the flow of conversation may be interrupted, which is also a problem. Therefore, there is a need for a conversational system that can efficiently analyze the text data entered by the user and quickly provide expert and specific advice.

[0995] 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.

[0996] In this invention, the server includes means for receiving text data entered by a user, means for analyzing the text data and calculating the frequency of specific keywords or phrases, means for identifying the user's problem based on the analysis results and assigning an appropriate role to a generative artificial intelligence, means for generating a response message according to the role using the generative artificial intelligence, means for providing the generated response message to the user, means for activating a user interface, means for sending an initial message to the user, and means for displaying the response message on the user interface. This enables the user to immediately recognize their problem and quickly obtain the specific and expert advice they need.

[0997] A "user" is someone who uses this system to seek advice.

[0998] "Text data" refers to the words and sentences entered by the user, which are the objects that are analyzed by the system.

[0999] "Analysis" refers to the process of decomposing and extracting features from received text data using natural language processing algorithms.

[1000] "Keywords" are important words or phrases contained within the text data entered by the user.

[1001] "Frequency" refers to the number of times a particular keyword or phrase appears within text data.

[1002] A "problem" refers to an issue or difficulty identified based on text data entered by the user into the system.

[1003] "Generative artificial intelligence" refers to algorithms and machine learning models that have the ability to generate appropriate response messages based on input data and a specified role.

[1004] "Role" refers to the professional position or character that a generative artificial intelligence assumes when generating a response message.

[1005] A "response message" is a text containing solutions or advice generated by a generative artificial intelligence system in response to a user's problem.

[1006] A "user interface" is software that provides screen displays and input methods for users to interact with a system.

[1007] An "initial message" is a message that a server sends to a user via the user interface for the first time, such as a question or confirmation.

[1008] "Means of receiving" refers to the mechanisms and methods for importing text data entered by a user into the system.

[1009] "Means of providing" refers to the mechanisms and methods for displaying the generated response message to the user.

[1010] "Means of display" refers to mechanisms or methods for visually showing response messages on a user interface.

[1011] This invention is an interactive system for users to receive advice on specific problems. The system analyzes text data entered by the user and uses generative artificial intelligence to provide appropriate response messages based on the data's content, thereby effectively solving the user's problems.

[1012] System Configuration

[1013] Launching the user interface

[1014] The device runs a specific application (for example, "adviceApp.exe"). The application displays a user interface that includes a chat box and a send button for the user to enter text data.

[1015] Initial contact and message reception

[1016] The server immediately sends an initial message to the user after the application starts. For example, it might send a message like, "Hello! What are you interested in today?"

[1017] When a user enters text data into a text box and presses the submit button, that data is sent from the device to the server. For example, a user might enter, "I've been having trouble sleeping lately."

[1018] The terminal acquires the input data, converts it to JSON format, and sends it to the server.

[1019] Text data analysis

[1020] The server analyzes the received text data using natural language processing (NLP) algorithms. Python libraries such as "NLTK" and "spaCy" are used for the analysis.

[1021] Specifically, the server segments the text data, extracts keywords, and calculates their frequency. For example, it might extract the keyword "can't sleep" from the text "I've been having trouble sleeping lately."

[1022] Determining the Role of Generative Artificial Intelligence

[1023] The server determines the appropriate role for a generative artificial intelligence based on the frequency of keywords extracted through NLP analysis. For example, if the keyword "can't sleep" appears frequently, it will assign the role of "doctor."

[1024] The system compares a list of frequently occurring keywords with a database and then instructs the generative artificial intelligence to assign a role based on the results.

[1025] Generating a response message

[1026] The generative artificial intelligence generates response messages to the user according to its assigned role. For example, it might use a model such as OpenAI's GPT-3.

[1027] The server generates a prompt such as, "That sounds difficult. Let's try a few things to help you get a good night's sleep. Going to bed at the same time every night and having time to relax before bed can be effective."

[1028] Providing response messages to users

[1029] The server sends the generated response message to the terminal. This message is often sent in JSON format.

[1030] The device displays received messages in the chat box. For example, a message like, "That sounds tough. Let's try a few things to help you get a good night's sleep," might be displayed.

[1031] Specific example

[1032] Example 1: Health-related cases

[1033] The server sends the first message to the user: "Hello! What are you interested in today?"

[1034] User response: "I've been having trouble sleeping lately."

[1035] The server analyzes the text data and extracts the keyword "can't sleep".

[1036] The server assigns the role of "doctor" to the generative artificial intelligence and generates prompts.

[1037] Generative AI: "That sounds difficult. Let's try a few methods to get a good night's sleep. For example, going to bed at the same time every night and having time to relax before bed can be effective."

[1038] The server sends a response message to the user, which is then displayed on the terminal.

[1039] Example 2: Cases related to finance

[1040] The server sends the user an initial message: "What are you interested in today?"

[1041] User response: "I'm worried because I haven't been managing my household finances well lately."

[1042] The server analyzes the text data and extracts the keywords "household finances" and "anxiety."

[1043] The server assigns the role of "tax accountant" to the generative artificial intelligence and generates prompts.

[1044] Generative AI: "Household financial management is very important, isn't it? The first step is to meticulously record your income and expenses and reduce unnecessary spending. If you need more specific advice, may I ask for more details?"

[1045] The server sends a response message to the user, which is then displayed on the terminal.

[1046] This invention allows users to instantly recognize their challenges and improve their quality of life by obtaining expert and specific advice.

[1047] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1048] Step 1:

[1049] The device runs a specific application (e.g., "adviceApp.exe"). The user interface launches, displaying a chat box for the user to enter text data and a send button.

[1050] Input: User-initiated application execution

[1051] Output: Display of a user interface including a chat box and a send button.

[1052] Step 2:

[1053] The server immediately sends an initial message to the user after the application starts. For example, it might display a message like, "Hello! What are you interested in today?"

[1054] Input: Application startup signal

[1055] Output: Generation and sending of the initial message

[1056] Step 3:

[1057] The user enters text data into a text box and presses the submit button. This data is sent from the terminal to the server. For example, the user might enter "I've been having trouble sleeping lately."

[1058] Input: User text input and submission operation

[1059] Output: Convert text data to JSON format and send to the server.

[1060] Step 4:

[1061] The server analyzes the received text data using natural language processing (NLP) algorithms. For example, it uses Python libraries such as "NLTK" and "spaCy" to tokenize the text data, extract keywords, and calculate their frequencies.

[1062] Input: Text data in JSON format

[1063] Output: Extracted keywords and their frequencies

[1064] Step 5:

[1065] The server determines the appropriate role for the generative artificial intelligence based on the extracted results. For example, if the keyword "can't sleep" appears frequently, it will assign the role of "doctor."

[1066] Input: Extracted keywords and their frequency

[1067] Output: Role instructions for generative artificial intelligence

[1068] Step 6:

[1069] The generative artificial intelligence generates response messages to the user according to its assigned role. For example, using OpenAI's GPT-3 model, it might generate a prompt such as, "That sounds tough. Let's try a few things to get a good night's sleep."

[1070] Input: Role instructions and generative artificial intelligence

[1071] Output: Response message

[1072] Step 7:

[1073] The server sends a generated response message to the terminal. This message is sent in JSON format.

[1074] Input: Generated response message

[1075] Output: Send a message in JSON format to the terminal.

[1076] Step 8:

[1077] The device parses the JSON-formatted message it receives and displays it in the chat box. For example, a message like, "That sounds tough. Let's try a few things to get a good night's sleep," might be displayed.

[1078] Input: Response message in JSON format

[1079] Output: Message displayed in the chat box

[1080] (Application Example 1)

[1081] 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."

[1082] Traditional content delivery services often fail to adequately recommend content based on what users have watched previously. This results in users spending a long time finding content that interests them, leading to decreased satisfaction. Furthermore, there was a lack of systems that provided real-time, specific advice to help users address their particular challenges. This inconsistent user experience negatively impacted customer satisfaction and retention rates.

[1083] 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.

[1084] In this invention, the server includes means for receiving text data entered by a user, means for analyzing the text data and calculating the frequency of specific keywords or phrases, means for identifying the user's problem based on the analysis results and assigning an appropriate role to a generative artificial intelligence, means for generating a response message according to the role using the generative artificial intelligence, means for recommending content that the user should watch next based on the content they have watched, and means for providing the generated response message to the user. By combining the user's viewing history with real-time text data analysis, it becomes possible to provide the user with more accurate content recommendations and specific advice.

[1085] "Means for receiving text data entered by the user" refers to a function for sending and receiving text and messages entered by the user using a terminal to a server.

[1086] "Means for analyzing the text data and calculating the frequency of specific keywords or phrases" refers to a function that analyzes the input text data using a natural language processing algorithm, extracts important words and phrases, and calculates their occurrence count and frequency.

[1087] "A means of identifying the user's challenges based on the analysis results and assigning an appropriate role to the generative artificial intelligence" refers to a function that identifies the problems the user faces from the analyzed keywords and phrases and assigns a role suitable for solving those problems to the generative artificial intelligence.

[1088] "Means for generating response messages according to roles using the aforementioned generative artificial intelligence" refers to a function in which the generative artificial intelligence creates an appropriate response message based on the assigned role.

[1089] "A means of recommending content to watch next based on the content a user has watched" refers to a function that analyzes highly relevant content based on a user's past viewing history and presents recommended content to watch next.

[1090] "Means for providing the generated response message to the user" refers to a function that sends the generated response message to the user's terminal and displays it.

[1091] This invention provides an interactive system for users to receive advice on specific problems, and in particular, for making recommendations based on viewing history in content distribution services. This system consists of a server, terminals, and a generative artificial intelligence (AI) model.

[1092] System Configuration

[1093] 1. Server

[1094] The server receives text data entered by the user and analyzes it. Natural language processing (NLP) algorithms are used for the analysis, and as a result, specific keywords and phrases are extracted.

[1095] The server identifies the user's problem based on the analysis results and assigns an appropriate role to the generative artificial intelligence. The response message generated by this role will be based on the user's viewing history and input text.

[1096] The server also has a function to recommend content that users should watch next, based on the content they have watched. Specifically, it analyzes and provides new content related to movies and TV shows that the user has recently watched.

[1097] 2. Terminal

[1098] The terminal provides a user interface, such as a chat box and a send button, for the user to input text data. The data entered by the user is sent to the server.

[1099] Furthermore, the terminal displays response messages sent from the server, providing an environment where users can receive real-time advice and view recommended content.

[1100] 3. Generative Artificial Intelligence

[1101] The generative artificial intelligence generates response messages based on appropriate prompts, following the roles assigned by the server. This uses OpenAI's GPT-3 model to provide specific advice and content recommendations for solving the user's problems.

[1102] The response message is generated based on a pre-configured prompt, taking into account the text data entered by the user and their viewing history.

[1103] Specific example

[1104] For example, suppose a user recently watched the movie "Inception." In this case, the server would take the following steps:

[1105] The server sends the user an initial message: "Hi! Would you like to talk about the movies you've watched recently?"

[1106] The user enters "Inception".

[1107] The server analyzes the text data and extracts the keyword "Inception".

[1108] The server assigns the role of "film critic" to the generative artificial intelligence and generates prompts.

[1109] Generative artificial intelligence generates a response message: "Since you've watched 'Inception,' I recommend 'Interstellar' or 'The Matrix' next. These films also have compelling storylines and profound philosophical themes, so please enjoy them."

[1110] The server sends the generated response message to the user, and it is displayed on the terminal.

[1111] Example of a prompt

[1112] Movies recently watched by users: Inception

[1113] Please recommend some movies.

[1114] This system combines users' viewing history with real-time text data analysis to provide users with more accurate content recommendations and specific advice.

[1115] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1116] Step 1:

[1117] The device launches the application. At this stage, the user interface is displayed, providing a chat box for the user to enter text data and a send button. Input: Launch of the user interface. Output: Display of the chat box and send button.

[1118] Step 2:

[1119] The server sends an initial message to the user. For example, the message might say, "Hello! Shall we talk about the movies you've watched recently?" Input: Initial state of the application. Output: Display of the initial message.

[1120] Step 3:

[1121] The user enters text data about the title of a movie they watched or other related information, and then presses the submit button. This data is sent from the terminal to the server. Input: User's text input (e.g., Inception). Output: Text data sent to the server.

[1122] Step 4:

[1123] The server analyzes the received text data using a natural language processing (NLP) algorithm. This analysis extracts specific keywords and phrases, and calculates their frequency. Input: Title of a movie watched (e.g., Inception). Output: Extracted keywords and their frequencies.

[1124] Step 5:

[1125] The server identifies the user's problem based on the analysis results and assigns an appropriate role to the generative artificial intelligence. For example, if the keyword "Inception" is extracted, the role of "film critic" will be assigned to the generative artificial intelligence. Input: Keyword analysis results. Output: Role assignment to the generative artificial intelligence.

[1126] Step 6:

[1127] A generative artificial intelligence generates a response message using appropriate prompts based on the assigned role. This response message includes specific advice and recommended content for the user. Input: Role assignment and prompt. Output: Response message (e.g., "Since you watched 'Inception,' we recommend 'Interstellar' or 'The Matrix' next").

[1128] Step 7:

[1129] The server sends a generated response message to the terminal and displays it on the user interface. The user views this message and is helped to decide what content to view next. Input: Response message. Output: Message displayed on the user interface.

[1130] This series of processing steps allows users to receive appropriate content recommendations and specific advice in real time, based on their viewing history. Furthermore, the use of generative artificial intelligence provides a deeper understanding of and solutions to the user's challenges.

[1131] 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.

[1132] This invention is an interactive system for users to receive advice on specific issues and emotions. The system analyzes text data entered by the user and the user's emotional information, and uses generative artificial intelligence to provide appropriate response messages based on the content, thereby effectively resolving the user's issues and emotional problems.

[1133] System Configuration

[1134] 1. Launch the user interface

[1135] The device launches the application. The application displays a user interface, including a chat box and a send button for the user to enter text data.

[1136] 2. Initial contact and message reception

[1137] The server sends a question to the user as its initial message. For example, a message like, "Hello! What are you interested in today?"

[1138] The user enters text data as a response and presses the send button. For example, they might enter, "I've been having trouble sleeping lately."

[1139] 3. Analysis of text data

[1140] The server analyzes the received text data using natural language processing (NLP) algorithms. The analysis extracts specific keywords and phrases, and calculates their frequency. For example, it analyzes the frequency of the keyword "can't sleep."

[1141] 4. Emotion recognition by an emotion engine

[1142] During the process of analyzing text data, the server uses an emotion engine to recognize the user's emotions. It extracts emotions such as "anxiety" and "stress" from the user's input.

[1143] 5. Determining the role of generative artificial intelligence

[1144] The server identifies the user's issues and emotions based on the analysis results and emotion recognition results. From the analysis results and emotion information, "possible sleep disorder" is identified as an issue and "anxiety" as an emotion.

[1145] The server assigns an appropriate role to the generative artificial intelligence. For example, it might assign the role of "doctor."

[1146] 6. Generating a response message

[1147] The generative artificial intelligence generates response messages using appropriate prompts based on its assigned role. For example, "That sounds difficult. Let's try a few things to get a good night's sleep. Going to bed at the same time every night and having time to relax before bed can be effective."

[1148] Taking the results of emotion recognition into consideration, for example, for "anxiety," add a message to care for those emotions, such as, "Also, try breathing exercises and light exercise to help you relax. You are not alone, we support you."

[1149] 7. Providing response messages to users

[1150] The server sends a generated response message to the terminal. The message is displayed in the user interface and can be viewed by the user.

[1151] Specific example

[1152] Example 1: Health-related cases (using the emotion engine)

[1153] The server sends the first message to the user: "Hello! What are you interested in today?"

[1154] User response: "I've been having trouble sleeping lately, and it's making me a little anxious."

[1155] The server analyzes the text data. It extracts the keyword "can't sleep" and recognizes the emotion "anxiety."

[1156] The server assigns the role of "doctor" to the generative artificial intelligence and generates prompts.

[1157] Generative AI: "That sounds difficult. Let's try a few things to help you get a good night's sleep. For example, going to bed at the same time every night and having time to relax before bed can be effective. You can also try breathing exercises or light exercise to help you relax. You are not alone, we are here to support you."

[1158] The server sends a response message to the user, which is displayed on the terminal.

[1159] Example 2: Financial-related cases (using the emotion engine)

[1160] The server sends the user an initial message: "What are you interested in today?"

[1161] User response: "I'm worried because I haven't been managing my household finances well lately."

[1162] The server analyzes the text data. It extracts the keywords "household finances" and "anxiety" and recognizes the emotion of "anxiety."

[1163] The server assigns the role of "tax accountant" to the generative artificial intelligence and generates a prompt.

[1164] Generative AI: "Household financial management is very important, isn't it? The first step is to meticulously record your income and expenses and reduce unnecessary spending. If you need more specific advice, may I ask for more details? Also, please feel free to contact me anytime to reduce any anxiety you may feel during the process."

[1165] The server sends a response message to the user, which is displayed on the terminal.

[1166] In this way, users can instantly receive advice tailored to specific challenges and emotions. By adding an emotion engine, it becomes possible to provide advice that takes users' emotions into consideration, leading to higher satisfaction.

[1167] The following describes the processing flow.

[1168] Step 1:

[1169] The device launches an application via the user. The application displays a user interface, including a chat box where the user can enter text data and a send button.

[1170] Step 2:

[1171] The server sends the user an initial message asking, "Hello! What are you interested in today?"

[1172] Step 3:

[1173] The user enters text data into the chat box and presses the send button. For example, they might type, "I've been having trouble sleeping lately, and it's making me a little anxious."

[1174] Step 4:

[1175] The terminal sends the message entered by the user to the server.

[1176] Step 5:

[1177] The server saves the text data it receives.

[1178] Step 6:

[1179] The server uses natural language processing (NLP) algorithms to analyze the received text data. Specifically, it extracts keywords and phrases from the text data and calculates their frequency of occurrence. For example, it analyzes the frequency of the keyword "can't sleep."

[1180] Step 7:

[1181] The server uses an emotion engine to recognize the user's emotions from text data. Specifically, it extracts emotions such as "anxiety" and "stress."

[1182] Step 8:

[1183] The server identifies the user's issues and emotions based on the analysis results and emotion recognition results. For example, the analysis results might identify "possible sleep disorder" as an issue, and "anxiety" as the emotion.

[1184] Step 9:

[1185] The server assigns appropriate roles to the generative artificial intelligence. For example, it might assign the role of "doctor" based on the task and emotions.

[1186] Step 10:

[1187] The server generates and sends an appropriate prompt to the generative artificial intelligence. For example, it might generate a prompt that says, "The user is having trouble sleeping. Provide advice on improving sleep."

[1188] Step 11:

[1189] The generative artificial intelligence generates a response message based on the role it has been assigned. For example, it might say, "That sounds difficult. Let's try a few things to get a good night's sleep. Going to bed at the same time every night and having time to relax before bed can be effective. You can also try breathing exercises and light exercise to help you relax. You are not alone, we support you."

[1190] Step 12:

[1191] The server sends the generated response message to the terminal.

[1192] Step 13:

[1193] The device displays the response message in the user's chat box.

[1194] Step 14:

[1195] The user receives the generated response message and can enter further questions or feedback as needed.

[1196] Step 15:

[1197] The terminal sends an additional message to the server, and the process from step 5 is repeated.

[1198] This allows users to receive advice on specific issues and emotions, and the system can provide responses that are appropriate to the user's situation.

[1199] (Example 2)

[1200] 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."

[1201] In recent years, there has been a growing demand for dialogue systems that provide appropriate advice tailored to the specific challenges and emotions users face. However, conventional dialogue systems have struggled to recognize emotions in addition to analyzing user text data, making it difficult to provide appropriate advice. In particular, advice provided without considering the user's emotional information is often insufficient and results in low user satisfaction.

[1202] The identification processing performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving text data input by the user, means for analyzing the text data and calculating the frequency of specific keywords and phrases, means for classifying emotions to recognize the user's emotions from the analyzed text data, means for identifying the user's problem based on the analysis results and emotion recognition results and assigning an appropriate role to the generative artificial intelligence, means for generating a response message according to the role using the generative artificial intelligence, and means for providing the generated response message to the user. This makes it possible to provide appropriate advice that also takes emotions into consideration for the user's specific problem.

[1203] "Means for receiving text data entered by the user" refers to the function that allows the server to receive text data entered by the user using a terminal.

[1204] "Means for calculating the frequency of specific keywords or phrases" refers to a function that analyzes received text data and measures the number of occurrences of pre-set keywords or phrases.

[1205] "Emotion classification means" refers to a function that uses algorithms or models to identify and classify a user's emotions during the process of analyzing text data.

[1206] "Means for assigning appropriate roles to generative artificial intelligence" refers to a function for setting appropriate roles (e.g., doctor, tax accountant, etc.) to generative artificial intelligence based on analysis results and emotion recognition results.

[1207] "Means for generating response messages" refers to a function that uses generative artificial intelligence to create response messages according to the assigned role.

[1208] "Means of providing a response message to the user" refers to a function that sends the generated response message to the user's terminal and displays it on the user interface.

[1209] This invention is an interactive system for users to receive advice on specific issues and emotions. The system analyzes text data entered by the user and the user's emotional information, and uses generative artificial intelligence to provide appropriate response messages based on that analysis.

[1210] System Configuration

[1211] Launching the user interface

[1212] The device launches the application. The application displays a user interface, including a chat box and a send button for the user to enter text data.

[1213] Sending and receiving the first message

[1214] The server sends the first message. The message says, "Hello! What are you interested in today?"

[1215] The user responds to this message by entering text data such as, "I've been having trouble sleeping lately," and then presses the send button.

[1216] Receiving and analyzing text data

[1217] The server receives text data sent by the user. The received data is analyzed using natural language processing (NLP) algorithms. Here, NLP libraries such as Python's NLTK and spaCy are used to extract the keyword "can't sleep".

[1218] Emotion recognition by an emotion engine

[1219] During the process of analyzing text data, the server uses an emotion engine to recognize the user's emotions. Emotion classification models (such as BERT or RoBERTa) are used to extract emotions like "anxiety" and "stress." Specifically, the text is input to the emotion classification model, and emotion labels are returned as output.

[1220] Identifying the issues and emotions, and determining the role of generative artificial intelligence.

[1221] The server identifies the user's issues and emotions based on the analysis results and emotion recognition results. For example, it might identify "potential sleep disorder" and "anxiety."

[1222] The server assigns the role of "doctor" to the generative artificial intelligence. Specifically, it selects the appropriate role from an internal database and provides the role information to the generative AI model as part of a prompt.

[1223] Generating a response message

[1224] The generative artificial intelligence creates a prompt message according to its assigned role. Here, for example, GPT-3 or a similar generative AI model is used. The prompt message would be something like, "The user is feeling anxious because they can't sleep. Please provide advice from a doctor's perspective."

[1225] Generative artificial intelligence generates a response message based on the input prompt. Based on that prompt, it creates content such as, "That sounds difficult. Let's try a few methods to get a good night's sleep. For example, going to bed at the same time every night or taking some time to relax before bed can be effective."

[1226] The generative artificial intelligence also takes emotion recognition results into consideration and adds the message, "Also, try breathing exercises and light exercise to help you relax. You are not alone; we support you."

[1227] Providing a response message to the user

[1228] The server sends the generated response message to the terminal. Specifically, the message data is sent to the user interface via the HTTP response.

[1229] The device displays received messages in the user interface. Users can see the response messages in the chat box on the screen.

[1230] Specific example

[1231] Example 1: Health-related cases

[1232] The server sends the first message to the user: "Hello! What are you interested in today?"

[1233] User response: "I've been having trouble sleeping lately, and it's making me a little anxious."

[1234] The server analyzes the text data, extracts the keyword "can't sleep," and recognizes the emotion of "anxiety."

[1235] The server assigns the role of "doctor" to the generative artificial intelligence and generates prompts.

[1236] The generative artificial intelligence created the following message: "That must be difficult. Let's try a few things to get a good night's sleep. For example, going to bed at the same time every night and having time to relax before bed can be effective. You can also try breathing exercises or light exercise to help you relax. You are not alone, we are here for you."

[1237] The server sends a response message to the user, which is then displayed on the terminal.

[1238] Example 2: Cases related to finance

[1239] The server sends the user an initial message: "What are you interested in today?"

[1240] User response: "I'm worried because I haven't been managing my household finances well lately."

[1241] The server analyzes the text data, extracts "household finances" and "anxiety," and recognizes the emotion of "anxiety."

[1242] The server assigns the role of "tax accountant" to the generative artificial intelligence and generates prompts.

[1243] The generative AI created the following message: "Household financial management is very important, isn't it? The first step is to meticulously record your income and expenses and reduce unnecessary spending. If you need more specific advice, may I ask for more details? Also, please feel free to contact us anytime to alleviate any anxiety you may feel during this process."

[1244] The server sends a response message to the user, which is then displayed on the terminal.

[1245] Thus, the present invention makes it possible to provide appropriate advice that takes into account the emotional aspects of the user's specific problems.

[1246] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1247] Step 1:

[1248] The device launches the application.

[1249] Input: The terminal the user is operating from.

[1250] Action: The application launches and the user interface is displayed.

[1251] Output: Screen including chat box and send button

[1252] Step 2:

[1253] The server sends the initial message.

[1254] Input: Initial message template on the server

[1255] Action: The server generates the message "Hello! What are you interested in today?" and sends it to the terminal.

[1256] Output: Initial message displayed to the user

[1257] Step 3:

[1258] The user responds to the initial message.

[1259] Input: User's text data (e.g., "I've been having trouble sleeping lately")

[1260] Action: The user enters text and presses the submit button.

[1261] Output: Text data sent to the server

[1262] Step 4:

[1263] The server receives text data sent by the user.

[1264] Input: Text data entered and submitted by the user.

[1265] Operation: The server receives text data and saves it to the database.

[1266] Output: Saved text data

[1267] Step 5:

[1268] The server analyzes the text data using natural language processing (NLP) algorithms.

[1269] Input: Saved text data

[1270] Specific operation: Use Python's NLTK or spaCy to extract specific keywords or phrases (e.g., "can't sleep").

[1271] Output: Keyword frequency data from the analysis results

[1272] Step 6:

[1273] The server uses an emotion engine to recognize the user's emotions.

[1274] Input: User's text data

[1275] Specific operation: Use emotion classification models such as BERT or RoBERTa to extract emotion labels such as "anxiety" and "stress."

[1276] Output: Sentiment labels as recognition results

[1277] Step 7:

[1278] The server identifies the user's issues and emotions based on the analysis results and emotion recognition results.

[1279] Input: Analysis results (keyword frequency data) and sentiment labels

[1280] Specific actions: The issue identified is "possible sleep disorder," and the emotion identified is "anxiety."

[1281] Output: Identification of issues and emotions

[1282] Step 8:

[1283] The server assigns appropriate roles to the generative artificial intelligence.

[1284] Input: Identification results of issues and emotions

[1285] Specific operation: Select an appropriate role (e.g., "doctor") from the internal database and provide role information to the generative artificial intelligence model.

[1286] Output: Role information assigned to the generative artificial intelligence.

[1287] Step 9:

[1288] The generative artificial intelligence generates the response message.

[1289] Input: User's identified issues, emotions, and assigned roles

[1290] Specific operation: Uses a generative AI model such as GPT-3 to generate a response message based on the prompt.

[1291] Example prompt: "The user is experiencing anxiety due to insomnia. Please provide advice from a doctor's perspective."

[1292] Output: Generated response message ("That sounds difficult. Let's try a few things to help you get a good night's sleep...")

[1293] Step 10:

[1294] The server sends the generated response message to the terminal.

[1295] Input: Generated response message

[1296] Specific operation: Message data is sent to the user interface via an HTTP response.

[1297] Output: Response message displayed on the user terminal

[1298] Step 11:

[1299] The response message received by the terminal is displayed on the user interface.

[1300] Input: Response message sent from the server

[1301] Operation: The message is displayed in the chat box and can be viewed by the user.

[1302] Output: Response message displayed to the user

[1303] This series of processes allows users to receive appropriate advice regarding specific issues and emotions.

[1304] (Application Example 2)

[1305] 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."

[1306] Conventional interactive systems often fail to adequately address user needs, as response messages generated using algorithms that identify problems in response to user input frequently do not adequately address user needs. Furthermore, responses that take into account user emotions are not provided, making it difficult to completely resolve user dissatisfaction and anxiety. As a result, improving the user experience has been challenging. This invention aims to solve this problem and provide an interactive system that can address both user problems and emotions.

[1307] 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 receiving text data input by the user, means for analyzing the text data and calculating the frequency of specific keywords and phrases, means for identifying the user's issues and emotions based on the analysis results and assigning an appropriate role to a generative artificial intelligence, means for generating a response message according to the role using the generative artificial intelligence, means for providing the generated response message to the user, and means for adding supplementary information and reassuring messages according to the user's emotions using an emotion recognition engine. This makes it possible to provide specific and useful advice regarding the user's issues and to provide appropriate care for the user's emotions.

[1308] A "user" is a person who uses an interactive system to seek advice on specific issues or emotions.

[1309] "Text data" refers to the written information entered by the user, including the user's concerns and emotions.

[1310] "Analysis" is the process of processing text data to calculate the frequency of specific keywords or phrases.

[1311] An "emotion recognition engine" is an algorithm that extracts a user's emotions from text data.

[1312] "Generative artificial intelligence" is artificial intelligence that generates appropriate response messages in response to a given prompt.

[1313] A "response message" is a message generated by a generative artificial intelligence system that addresses the user's problems and provides emotional support.

[1314] A "server" is a core computer device in a system that receives user input data, analyzes it, assigns roles, and generates and provides response messages.

[1315] A "role" refers to the specific response characteristics and areas of expertise assigned to a generative artificial intelligence.

[1316] "Supplemental information" refers to additional explanations or advice added to a response message that are tailored to the user's emotional state.

[1317] A "reassuring message" is a message provided to care for the user's emotions and offer a sense of mental reassurance.

[1318] A "natural language processing algorithm" is a computational method used to analyze text data to extract keywords and recognize emotions.

[1319] This invention is an interactive system for users to receive advice on specific issues and emotions. The system analyzes text data entered by the user and the user's emotional information, and uses generative artificial intelligence to provide appropriate response messages based on that analysis. The embodiments for carrying out this invention are described in detail below.

[1320] System Configuration

[1321] hardware

[1322] User terminal: Smartphones, tablets, personal computers, etc. These are devices used by users to input text data and view received response messages.

[1323] Server: A computer device that performs backend processing. It receives user input data, analyzes it, assigns roles, and generates and sends response messages.

[1324] software

[1325] Programming language: Python

[1326] Natural language processing algorithms: Use TensorFlow or PyTorch to parse user input data.

[1327] Emotion recognition engine: Extracts emotions from text data using BERT models or similar emotion recognition models.

[1328] Generative AI Model: Using GPT-3 or similar generative models, it generates response messages tailored to the user's needs.

[1329] Web framework: Use Flask or Django to manage communication between the server and the user's device.

[1330] User Interface: We provide an intuitive interface for users using React or Vue.js.

[1331] Program processing

[1332] When the server receives text data from a user, it first analyzes the input data using a natural language processing algorithm. This analysis calculates the frequency of specific keywords and phrases, identifying the user's issues and emotions. At this stage, an emotion recognition engine is used to extract emotions from the user's text data.

[1333] Next, based on the analysis results and emotion recognition results, an appropriate role is assigned to the generative artificial intelligence. The generative artificial intelligence, once assigned a role, generates a response message based on the prompt text. At this time, the emotion recognition results are also taken into consideration, and supplementary information and reassuring messages corresponding to the user's emotions are added.

[1334] Finally, the generated response message is sent from the server to the user's terminal and displayed on the user interface. This allows the user to receive appropriate, emotionally sensitive responses along with specific advice.

[1335] Specific example

[1336] For example, consider a scenario where a user enters "I want a new smartphone, but I'm unsure which one to get" in a virtual store.

[1337] The server first receives this text data and begins analysis. The analysis extracts the keywords "smartphone" and "hesitant," and the emotion recognition engine recognizes the emotion "anxiety."

[1338] Next, the prompt given to the generative AI model will be as follows:

[1339] "I want a new smartphone, but I don't know which one to get. What do you recommend?"

[1340] Generative artificial intelligence generates the following response message based on this prompt:

[1341] "For the latest models, we'll introduce some specific new models and highly-rated models. Also, certain mid-range models are highly regarded."

[1342] The following supplementary information is added to the recognized emotion of "anxiety":

[1343] "Making your first expensive purchase can be nerve-wracking, right? Please feel free to ask any questions you may have. We'll work together to address any concerns you may have."

[1344] In this way, users can gain a sense of security along with specific and helpful advice. This approach significantly improves the user experience.

[1345] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1346] Step 1:

[1347] The server receives text data entered by the user. Text data entered by the user through the device is sent to the server. For example, if a user enters "I want a new smartphone, but I don't know which one to get," that text data is sent to the server.

[1348] Step 2:

[1349] The server analyzes the received text data using natural language processing algorithms. Specifically, the server extracts specific keywords and phrases from the text data and calculates their frequency. For example, the keywords "smartphone" and "hesitant" are extracted, and their frequencies are calculated.

[1350] Step 3:

[1351] The server applies an emotion recognition engine to the analyzed text data. This engine extracts the user's emotions from the text data. For example, the emotion "anxious" might be recognized from the word "hesitant." The input is the analyzed text data, and the output is the extracted emotion information.

[1352] Step 4:

[1353] The server identifies the user's problem based on the analysis results and emotion recognition results, and assigns an appropriate role to the generative artificial intelligence. For example, if the user's problem is identified as "choosing a smartphone" and the user's emotion is identified as "anxiety," the generative artificial intelligence is assigned the role of "product advisor."

[1354] Step 5:

[1355] The server provides a prompt to a generative artificial intelligence (AI) system to generate a response message. The prompt is generated based on the user's input data and emotions. For example, if the prompt "I want a new smartphone, but I don't know which one to get. What do you recommend?" is input to the generative AI system, the response message "For the latest models, we can recommend several specific new models and highly-rated models. Also, certain mid-range models are highly rated." is output.

[1356] Step 6:

[1357] The server adds supplementary information and reassuring messages aimed at emotional support to the generated response message. For example, it might add information such as, "Making your first expensive purchase can be nerve-wracking. Please feel free to ask any questions you may have. We'll work together to address your concerns."

[1358] Step 7:

[1359] The server sends the generated response message to the user's terminal and displays it on the user interface. This allows the user to receive an emotionally sensitive message along with appropriate advice. For example, the user's terminal might display a message such as, "If you're looking for the latest model, we can recommend some of the newest and most highly-rated models. We also have some highly-rated mid-range models. Making your first expensive purchase can be nerve-wracking, so please feel free to ask any questions you may have. We'll help you resolve your concerns."

[1360] 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.

[1361] 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.

[1362] 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.

[1363] [Fourth Embodiment]

[1364] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1365] 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.

[1366] 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).

[1367] 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.

[1368] 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.

[1369] 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).

[1370] 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.

[1371] 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.

[1372] 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.

[1373] 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.

[1374] 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.

[1375] 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.

[1376] 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".

[1377] This invention is an interactive system for users to receive advice on specific problems. The system analyzes text data entered by the user and uses generative artificial intelligence to provide appropriate response messages based on the data's content, thereby effectively solving the user's problems.

[1378] System Configuration

[1379] 1. Launch the user interface

[1380] The device launches the application. The application displays a user interface that includes a chat box and a send button for the user to enter text data.

[1381] 2. Initial contact and message reception

[1382] The server sends a question to the user as its initial message. For example, a message like, "Hello! What are you interested in today?"

[1383] The user enters text data as a response and presses the send button. The entered message is sent from the terminal to the server.

[1384] 3. Analysis of text data

[1385] The server analyzes the received text data using natural language processing (NLP) algorithms. The analysis extracts specific keywords and phrases, and calculates their frequency.

[1386] For example, if the input is "I haven't been able to sleep lately," the keyword "can't sleep" will be analyzed.

[1387] 4. Determining the role of generative artificial intelligence

[1388] The server identifies the user's problem based on the analysis results. It then assigns appropriate roles to generative artificial intelligence based on frequently occurring keywords.

[1389] For example, if the keyword "can't sleep" appears frequently, the role of "doctor" might be assigned to that keyword.

[1390] 5. Generating a response message

[1391] The generative artificial intelligence generates response messages using appropriate prompts according to its assigned role. For example, a message like, "That sounds difficult. Let's try a few things to get a good night's sleep. Going to bed at the same time every night and having time to relax before bed can be effective."

[1392] 6. Providing response messages to users

[1393] The server sends a generated response message to the terminal. The message is displayed in the user interface and can be viewed by the user.

[1394] Specific example

[1395] Example 1: Health-related cases

[1396] The server sends the first message to the user: "Hello! What are you interested in today?"

[1397] User response: "I've been having trouble sleeping lately."

[1398] The server analyzes the text data. It extracts the keyword "can't sleep" and checks its frequent occurrences.

[1399] The server assigns the role of "doctor" to the generative artificial intelligence and generates prompts.

[1400] Generative AI: "That sounds difficult. Let's try a few methods to get a good night's sleep. For example, going to bed at the same time every night and having time to relax before bed can be effective."

[1401] The server sends a response message to the user, which is displayed on the terminal.

[1402] Example 2: Cases related to finance

[1403] The server sends the user an initial message: "What are you interested in today?"

[1404] User response: "I'm worried because I haven't been managing my household finances well lately."

[1405] The server analyzes the text data and extracts the keywords "household finances" and "anxiety."

[1406] The server assigns the role of "tax accountant" to the generative artificial intelligence and generates a prompt.

[1407] Generative AI: "Household financial management is very important, isn't it? The first step is to meticulously record your income and expenses and reduce unnecessary spending. If you need more specific advice, may I ask for more details?"

[1408] The server sends a response message to the user, which is displayed on the terminal.

[1409] This system allows users to instantly recognize their challenges and receive expert advice. This is expected to significantly improve users' quality of life.

[1410] The following describes the processing flow.

[1411] Step 1:

[1412] The device launches an application via the user. The application displays a user interface, including a chat box where the user can enter text data and a send button.

[1413] Step 2:

[1414] The server sends the user an initial message asking, "Hello! What are you interested in today?"

[1415] Step 3:

[1416] The user enters text data into the chat box and presses the send button. For example, they might type, "I've been having trouble sleeping lately."

[1417] Step 4:

[1418] The terminal sends the message entered by the user to the server.

[1419] Step 5:

[1420] The server saves the text data it receives.

[1421] Step 6:

[1422] The server uses natural language processing (NLP) algorithms to analyze the received text data. Specifically, it extracts keywords and phrases from the text data and calculates their frequency of occurrence. For example, it analyzes the frequency of the keyword "can't sleep."

[1423] Step 7:

[1424] The server identifies the user's problem based on the analysis results. The analysis confirms that keywords related to "difficulty sleeping" appear frequently, and identifies "possible sleep disorder" as the user's problem.

[1425] Step 8:

[1426] The server assigns an appropriate role to the generative artificial intelligence. Based on the identified task, it assigns, for example, the role of "doctor."

[1427] Step 9:

[1428] The server generates and sends an appropriate prompt to the generative artificial intelligence. For example, the prompt might say, "The user is having trouble sleeping. Provide advice on improving sleep."

[1429] Step 10:

[1430] The generative artificial intelligence generates a response message based on the role it has been assigned. For example, it might say, "That sounds difficult. Let's try a few things to get a good night's sleep. Going to bed at the same time every night and having time to relax before bed can be effective."

[1431] Step 11:

[1432] The server sends the generated response message to the terminal.

[1433] Step 12:

[1434] The device displays the response message in the user's chat box.

[1435] Step 13:

[1436] The user receives the generated response message and can enter further questions or feedback as needed.

[1437] Step 14:

[1438] The terminal sends an additional message to the server, and the process from step 5 is repeated.

[1439] (Example 1)

[1440] 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".

[1441] In conventional conversational systems, it can be difficult for users to obtain appropriate advice for the specific problems they face. This is because the input text data is not effectively analyzed, and expert responses that meet the user's needs are not generated. Furthermore, the user interface may not be intuitive, and the flow of conversation may be interrupted, which is also a problem. Therefore, there is a need for a conversational system that can efficiently analyze the text data entered by the user and quickly provide expert and specific advice.

[1442] 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.

[1443] In this invention, the server includes means for receiving text data entered by a user, means for analyzing the text data and calculating the frequency of specific keywords or phrases, means for identifying the user's problem based on the analysis results and assigning an appropriate role to a generative artificial intelligence, means for generating a response message according to the role using the generative artificial intelligence, means for providing the generated response message to the user, means for activating a user interface, means for sending an initial message to the user, and means for displaying the response message on the user interface. This enables the user to immediately recognize their problem and quickly obtain the specific and expert advice they need.

[1444] A "user" is someone who uses this system to seek advice.

[1445] "Text data" refers to the words and sentences entered by the user, which are the objects that are analyzed by the system.

[1446] "Analysis" refers to the process of decomposing and extracting features from received text data using natural language processing algorithms.

[1447] "Keywords" are important words or phrases contained within the text data entered by the user.

[1448] "Frequency" refers to the number of times a particular keyword or phrase appears within text data.

[1449] A "problem" refers to an issue or difficulty identified based on text data entered by the user into the system.

[1450] "Generative artificial intelligence" refers to algorithms and machine learning models that have the ability to generate appropriate response messages based on input data and a specified role.

[1451] "Role" refers to the professional position or character that a generative artificial intelligence assumes when generating a response message.

[1452] A "response message" is a text containing solutions or advice generated by a generative artificial intelligence system in response to a user's problem.

[1453] A "user interface" is software that provides screen displays and input methods for users to interact with a system.

[1454] An "initial message" is a message that a server sends to a user via the user interface for the first time, such as a question or confirmation.

[1455] "Means of receiving" refers to the mechanisms and methods for importing text data entered by a user into the system.

[1456] "Means of providing" refers to the mechanisms and methods for displaying the generated response message to the user.

[1457] "Means of display" refers to mechanisms or methods for visually showing response messages on a user interface.

[1458] This invention is an interactive system for users to receive advice on specific problems. The system analyzes text data entered by the user and uses generative artificial intelligence to provide appropriate response messages based on the data's content, thereby effectively solving the user's problems.

[1459] System Configuration

[1460] Launching the user interface

[1461] The device runs a specific application (for example, "adviceApp.exe"). The application displays a user interface that includes a chat box and a send button for the user to enter text data.

[1462] Initial contact and message reception

[1463] The server immediately sends an initial message to the user after the application starts. For example, it might send a message like, "Hello! What are you interested in today?"

[1464] When a user enters text data into a text box and presses the submit button, that data is sent from the device to the server. For example, a user might enter, "I've been having trouble sleeping lately."

[1465] The terminal acquires the input data, converts it to JSON format, and sends it to the server.

[1466] Text data analysis

[1467] The server analyzes the received text data using natural language processing (NLP) algorithms. Python libraries such as "NLTK" and "spaCy" are used for the analysis.

[1468] Specifically, the server segments the text data, extracts keywords, and calculates their frequency. For example, it might extract the keyword "can't sleep" from the text "I've been having trouble sleeping lately."

[1469] Determining the Role of Generative Artificial Intelligence

[1470] The server determines the appropriate role for a generative artificial intelligence based on the frequency of keywords extracted through NLP analysis. For example, if the keyword "can't sleep" appears frequently, it will assign the role of "doctor."

[1471] The system compares a list of frequently occurring keywords with a database and then instructs the generative artificial intelligence to assign a role based on the results.

[1472] Generating a response message

[1473] The generative artificial intelligence generates response messages to the user according to its assigned role. For example, it might use a model such as OpenAI's GPT-3.

[1474] The server generates a prompt such as, "That sounds difficult. Let's try a few things to help you get a good night's sleep. Going to bed at the same time every night and having time to relax before bed can be effective."

[1475] Providing response messages to users

[1476] The server sends the generated response message to the terminal. This message is often sent in JSON format.

[1477] The device displays received messages in the chat box. For example, a message like, "That sounds tough. Let's try a few things to help you get a good night's sleep," might be displayed.

[1478] Specific example

[1479] Example 1: Health-related cases

[1480] The server sends the first message to the user: "Hello! What are you interested in today?"

[1481] User response: "I've been having trouble sleeping lately."

[1482] The server analyzes the text data and extracts the keyword "can't sleep".

[1483] The server assigns the role of "doctor" to the generative artificial intelligence and generates prompts.

[1484] Generative AI: "That sounds difficult. Let's try a few methods to get a good night's sleep. For example, going to bed at the same time every night and having time to relax before bed can be effective."

[1485] The server sends a response message to the user, which is then displayed on the terminal.

[1486] Example 2: Cases related to finance

[1487] The server sends the user an initial message: "What are you interested in today?"

[1488] User response: "I'm worried because I haven't been managing my household finances well lately."

[1489] The server analyzes the text data and extracts the keywords "household finances" and "anxiety."

[1490] The server assigns the role of "tax accountant" to the generative artificial intelligence and generates prompts.

[1491] Generative AI: "Household financial management is very important, isn't it? The first step is to meticulously record your income and expenses and reduce unnecessary spending. If you need more specific advice, may I ask for more details?"

[1492] The server sends a response message to the user, which is then displayed on the terminal.

[1493] This invention allows users to instantly recognize their challenges and improve their quality of life by obtaining expert and specific advice.

[1494] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1495] Step 1:

[1496] The device runs a specific application (e.g., "adviceApp.exe"). The user interface launches, displaying a chat box for the user to enter text data and a send button.

[1497] Input: User-initiated application execution

[1498] Output: Display of a user interface including a chat box and a send button.

[1499] Step 2:

[1500] The server immediately sends an initial message to the user after the application starts. For example, it might display a message like, "Hello! What are you interested in today?"

[1501] Input: Application startup signal

[1502] Output: Generation and sending of the initial message

[1503] Step 3:

[1504] The user enters text data into a text box and presses the submit button. This data is sent from the terminal to the server. For example, the user might enter "I've been having trouble sleeping lately."

[1505] Input: User text input and submission operation

[1506] Output: Convert text data to JSON format and send to the server.

[1507] Step 4:

[1508] The server analyzes the received text data using natural language processing (NLP) algorithms. For example, it uses Python libraries such as "NLTK" and "spaCy" to tokenize the text data, extract keywords, and calculate their frequencies.

[1509] Input: Text data in JSON format

[1510] Output: Extracted keywords and their frequencies

[1511] Step 5:

[1512] The server determines the appropriate role for the generative artificial intelligence based on the extracted results. For example, if the keyword "can't sleep" appears frequently, it will assign the role of "doctor."

[1513] Input: Extracted keywords and their frequency

[1514] Output: Role instructions for generative artificial intelligence

[1515] Step 6:

[1516] The generative artificial intelligence generates response messages to the user according to its assigned role. For example, using OpenAI's GPT-3 model, it might generate a prompt such as, "That sounds tough. Let's try a few things to get a good night's sleep."

[1517] Input: Role instructions and generative artificial intelligence

[1518] Output: Response message

[1519] Step 7:

[1520] The server sends a generated response message to the terminal. This message is sent in JSON format.

[1521] Input: Generated response message

[1522] Output: Send a message in JSON format to the terminal.

[1523] Step 8:

[1524] The device parses the JSON-formatted message it receives and displays it in the chat box. For example, a message like, "That sounds tough. Let's try a few things to get a good night's sleep," might be displayed.

[1525] Input: Response message in JSON format

[1526] Output: Message displayed in the chat box

[1527] (Application Example 1)

[1528] 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".

[1529] Traditional content delivery services often fail to adequately recommend content based on what users have watched previously. This results in users spending a long time finding content that interests them, leading to decreased satisfaction. Furthermore, there was a lack of systems that provided real-time, specific advice to help users address their particular challenges. This inconsistent user experience negatively impacted customer satisfaction and retention rates.

[1530] 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.

[1531] In this invention, the server includes means for receiving text data entered by a user, means for analyzing the text data and calculating the frequency of specific keywords or phrases, means for identifying the user's problem based on the analysis results and assigning an appropriate role to a generative artificial intelligence, means for generating a response message according to the role using the generative artificial intelligence, means for recommending content that the user should watch next based on the content they have watched, and means for providing the generated response message to the user. By combining the user's viewing history with real-time text data analysis, it becomes possible to provide the user with more accurate content recommendations and specific advice.

[1532] "Means for receiving text data entered by the user" refers to a function for sending and receiving text and messages entered by the user using a terminal to a server.

[1533] "Means for analyzing the text data and calculating the frequency of specific keywords or phrases" refers to a function that analyzes the input text data using a natural language processing algorithm, extracts important words and phrases, and calculates their occurrence count and frequency.

[1534] "A means of identifying the user's challenges based on the analysis results and assigning an appropriate role to the generative artificial intelligence" refers to a function that identifies the problems the user faces from the analyzed keywords and phrases and assigns a role suitable for solving those problems to the generative artificial intelligence.

[1535] "Means for generating response messages according to roles using the aforementioned generative artificial intelligence" refers to a function in which the generative artificial intelligence creates an appropriate response message based on the assigned role.

[1536] "A means of recommending content to watch next based on the content a user has watched" refers to a function that analyzes highly relevant content based on a user's past viewing history and presents recommended content to watch next.

[1537] "Means for providing the generated response message to the user" refers to a function that sends the generated response message to the user's terminal and displays it.

[1538] This invention provides an interactive system for users to receive advice on specific problems, and in particular, for making recommendations based on viewing history in content distribution services. This system consists of a server, terminals, and a generative artificial intelligence (AI) model.

[1539] System Configuration

[1540] 1. Server

[1541] The server receives text data entered by the user and analyzes it. Natural language processing (NLP) algorithms are used for the analysis, and as a result, specific keywords and phrases are extracted.

[1542] The server identifies the user's problem based on the analysis results and assigns an appropriate role to the generative artificial intelligence. The response message generated by this role will be based on the user's viewing history and input text.

[1543] The server also has a function to recommend content that users should watch next, based on the content they have watched. Specifically, it analyzes and provides new content related to movies and TV shows that the user has recently watched.

[1544] 2. Terminal

[1545] The terminal provides a user interface, such as a chat box and a send button, for the user to input text data. The data entered by the user is sent to the server.

[1546] Furthermore, the terminal displays response messages sent from the server, providing an environment where users can receive real-time advice and view recommended content.

[1547] 3. Generative Artificial Intelligence

[1548] The generative artificial intelligence generates response messages based on appropriate prompts, following the roles assigned by the server. This uses OpenAI's GPT-3 model to provide specific advice and content recommendations for solving the user's problems.

[1549] The response message is generated based on a pre-configured prompt, taking into account the text data entered by the user and their viewing history.

[1550] Specific example

[1551] For example, suppose a user recently watched the movie "Inception." In this case, the server would take the following steps:

[1552] The server sends the user an initial message: "Hi! Would you like to talk about the movies you've watched recently?"

[1553] The user enters "Inception".

[1554] The server analyzes the text data and extracts the keyword "Inception".

[1555] The server assigns the role of "film critic" to the generative artificial intelligence and generates prompts.

[1556] Generative artificial intelligence generates a response message: "Since you've watched 'Inception,' I recommend 'Interstellar' or 'The Matrix' next. These films also have compelling storylines and profound philosophical themes, so please enjoy them."

[1557] The server sends the generated response message to the user, and it is displayed on the terminal.

[1558] Example of a prompt

[1559] Movies recently watched by users: Inception

[1560] Please recommend some movies.

[1561] This system combines users' viewing history with real-time text data analysis to provide users with more accurate content recommendations and specific advice.

[1562] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1563] Step 1:

[1564] The device launches the application. At this stage, the user interface is displayed, providing a chat box for the user to enter text data and a send button. Input: Launch of the user interface. Output: Display of the chat box and send button.

[1565] Step 2:

[1566] The server sends an initial message to the user. For example, the message might say, "Hello! Shall we talk about the movies you've watched recently?" Input: Initial state of the application. Output: Display of the initial message.

[1567] Step 3:

[1568] The user enters text data about the title of a movie they watched or other related information, and then presses the submit button. This data is sent from the terminal to the server. Input: User's text input (e.g., Inception). Output: Text data sent to the server.

[1569] Step 4:

[1570] The server analyzes the received text data using a natural language processing (NLP) algorithm. This analysis extracts specific keywords and phrases, and calculates their frequency. Input: Title of a movie watched (e.g., Inception). Output: Extracted keywords and their frequencies.

[1571] Step 5:

[1572] The server identifies the user's problem based on the analysis results and assigns an appropriate role to the generative artificial intelligence. For example, if the keyword "Inception" is extracted, the role of "film critic" will be assigned to the generative artificial intelligence. Input: Keyword analysis results. Output: Role assignment to the generative artificial intelligence.

[1573] Step 6:

[1574] A generative artificial intelligence generates a response message using appropriate prompts based on the assigned role. This response message includes specific advice and recommended content for the user. Input: Role assignment and prompt. Output: Response message (e.g., "Since you watched 'Inception,' we recommend 'Interstellar' or 'The Matrix' next").

[1575] Step 7:

[1576] The server sends a generated response message to the terminal and displays it on the user interface. The user views this message and is helped to decide what content to view next. Input: Response message. Output: Message displayed on the user interface.

[1577] This series of processing steps allows users to receive appropriate content recommendations and specific advice in real time, based on their viewing history. Furthermore, the use of generative artificial intelligence provides a deeper understanding of and solutions to the user's challenges.

[1578] 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.

[1579] This invention is an interactive system for users to receive advice on specific issues and emotions. The system analyzes text data entered by the user and the user's emotional information, and uses generative artificial intelligence to provide appropriate response messages based on the content, thereby effectively resolving the user's issues and emotional problems.

[1580] System Configuration

[1581] 1. Launch the user interface

[1582] The device launches the application. The application displays a user interface, including a chat box and a send button for the user to enter text data.

[1583] 2. Initial contact and message reception

[1584] The server sends a question to the user as its initial message. For example, a message like, "Hello! What are you interested in today?"

[1585] The user enters text data as a response and presses the send button. For example, they might enter, "I've been having trouble sleeping lately."

[1586] 3. Analysis of text data

[1587] The server analyzes the received text data using natural language processing (NLP) algorithms. The analysis extracts specific keywords and phrases, and calculates their frequency. For example, it analyzes the frequency of the keyword "can't sleep."

[1588] 4. Emotion recognition by an emotion engine

[1589] During the process of analyzing text data, the server uses an emotion engine to recognize the user's emotions. It extracts emotions such as "anxiety" and "stress" from the user's input.

[1590] 5. Determining the role of generative artificial intelligence

[1591] The server identifies the user's issues and emotions based on the analysis results and emotion recognition results. From the analysis results and emotion information, "possible sleep disorder" is identified as an issue and "anxiety" as an emotion.

[1592] The server assigns an appropriate role to the generative artificial intelligence. For example, it might assign the role of "doctor."

[1593] 6. Generating a response message

[1594] The generative artificial intelligence generates response messages using appropriate prompts based on its assigned role. For example, "That sounds difficult. Let's try a few things to get a good night's sleep. Going to bed at the same time every night and having time to relax before bed can be effective."

[1595] Taking the results of emotion recognition into consideration, for example, for "anxiety," add a message to care for those emotions, such as, "Also, try breathing exercises and light exercise to help you relax. You are not alone, we support you."

[1596] 7. Providing response messages to users

[1597] The server sends a generated response message to the terminal. The message is displayed in the user interface and can be viewed by the user.

[1598] Specific example

[1599] Example 1: Health-related cases (using the emotion engine)

[1600] The server sends the first message to the user: "Hello! What are you interested in today?"

[1601] User response: "I've been having trouble sleeping lately, and it's making me a little anxious."

[1602] The server analyzes the text data. It extracts the keyword "can't sleep" and recognizes the emotion "anxiety."

[1603] The server assigns the role of "doctor" to the generative artificial intelligence and generates prompts.

[1604] Generative AI: "That sounds difficult. Let's try a few things to help you get a good night's sleep. For example, going to bed at the same time every night and having time to relax before bed can be effective. You can also try breathing exercises or light exercise to help you relax. You are not alone, we are here to support you."

[1605] The server sends a response message to the user, which is displayed on the terminal.

[1606] Example 2: Financial-related cases (using the emotion engine)

[1607] The server sends the user an initial message: "What are you interested in today?"

[1608] User response: "I'm worried because I haven't been managing my household finances well lately."

[1609] The server analyzes the text data. It extracts the keywords "household finances" and "anxiety" and recognizes the emotion of "anxiety."

[1610] The server assigns the role of "tax accountant" to the generative artificial intelligence and generates a prompt.

[1611] Generative AI: "Household financial management is very important, isn't it? The first step is to meticulously record your income and expenses and reduce unnecessary spending. If you need more specific advice, may I ask for more details? Also, please feel free to contact me anytime to reduce any anxiety you may feel during the process."

[1612] The server sends a response message to the user, which is displayed on the terminal.

[1613] In this way, users can instantly receive advice tailored to specific challenges and emotions. By adding an emotion engine, it becomes possible to provide advice that takes users' emotions into consideration, leading to higher satisfaction.

[1614] The following describes the processing flow.

[1615] Step 1:

[1616] The device launches an application via the user. The application displays a user interface, including a chat box where the user can enter text data and a send button.

[1617] Step 2:

[1618] The server sends the user an initial message asking, "Hello! What are you interested in today?"

[1619] Step 3:

[1620] The user enters text data into the chat box and presses the send button. For example, they might type, "I've been having trouble sleeping lately, and it's making me a little anxious."

[1621] Step 4:

[1622] The terminal sends the message entered by the user to the server.

[1623] Step 5:

[1624] The server saves the text data it receives.

[1625] Step 6:

[1626] The server uses natural language processing (NLP) algorithms to analyze the received text data. Specifically, it extracts keywords and phrases from the text data and calculates their frequency of occurrence. For example, it analyzes the frequency of the keyword "can't sleep."

[1627] Step 7:

[1628] The server uses an emotion engine to recognize the user's emotions from text data. Specifically, it extracts emotions such as "anxiety" and "stress."

[1629] Step 8:

[1630] The server identifies the user's issues and emotions based on the analysis results and emotion recognition results. For example, the analysis results might identify "possible sleep disorder" as an issue, and "anxiety" as the emotion.

[1631] Step 9:

[1632] The server assigns appropriate roles to the generative artificial intelligence. For example, it might assign the role of "doctor" based on the task and emotions.

[1633] Step 10:

[1634] The server generates and sends an appropriate prompt to the generative artificial intelligence. For example, it might generate a prompt that says, "The user is having trouble sleeping. Provide advice on improving sleep."

[1635] Step 11:

[1636] The generative artificial intelligence generates a response message based on the role it has been assigned. For example, it might say, "That sounds difficult. Let's try a few things to get a good night's sleep. Going to bed at the same time every night and having time to relax before bed can be effective. You can also try breathing exercises and light exercise to help you relax. You are not alone, we support you."

[1637] Step 12:

[1638] The server sends the generated response message to the terminal.

[1639] Step 13:

[1640] The device displays the response message in the user's chat box.

[1641] Step 14:

[1642] The user receives the generated response message and can enter further questions or feedback as needed.

[1643] Step 15:

[1644] The terminal sends an additional message to the server, and the process from step 5 is repeated.

[1645] This allows users to receive advice on specific issues and emotions, and the system can provide responses that are appropriate to the user's situation.

[1646] (Example 2)

[1647] 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".

[1648] In recent years, there has been a growing demand for dialogue systems that provide appropriate advice tailored to the specific challenges and emotions users face. However, conventional dialogue systems have struggled to recognize emotions in addition to analyzing user text data, making it difficult to provide appropriate advice. In particular, advice provided without considering the user's emotional information is often insufficient and results in low user satisfaction.

[1649] The identification processing performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving text data input by the user, means for analyzing the text data and calculating the frequency of specific keywords and phrases, means for classifying emotions to recognize the user's emotions from the analyzed text data, means for identifying the user's problem based on the analysis results and emotion recognition results and assigning an appropriate role to the generative artificial intelligence, means for generating a response message according to the role using the generative artificial intelligence, and means for providing the generated response message to the user. This makes it possible to provide appropriate advice that also takes emotions into consideration for the user's specific problem.

[1650] "Means for receiving text data entered by the user" refers to the function that allows the server to receive text data entered by the user using a terminal.

[1651] "Means for calculating the frequency of specific keywords or phrases" refers to a function that analyzes received text data and measures the number of occurrences of pre-set keywords or phrases.

[1652] "Emotion classification means" refers to a function that uses algorithms or models to identify and classify a user's emotions during the process of analyzing text data.

[1653] "Means for assigning appropriate roles to generative artificial intelligence" refers to a function for setting appropriate roles (e.g., doctor, tax accountant, etc.) to generative artificial intelligence based on analysis results and emotion recognition results.

[1654] "Means for generating response messages" refers to a function that uses generative artificial intelligence to create response messages according to the assigned role.

[1655] "Means of providing a response message to the user" refers to a function that sends the generated response message to the user's terminal and displays it on the user interface.

[1656] This invention is an interactive system for users to receive advice on specific issues and emotions. The system analyzes text data entered by the user and the user's emotional information, and uses generative artificial intelligence to provide appropriate response messages based on that analysis.

[1657] System Configuration

[1658] Launching the user interface

[1659] The device launches the application. The application displays a user interface, including a chat box and a send button for the user to enter text data.

[1660] Sending and receiving the first message

[1661] The server sends the first message. The message says, "Hello! What are you interested in today?"

[1662] The user responds to this message by entering text data such as, "I've been having trouble sleeping lately," and then presses the send button.

[1663] Receiving and analyzing text data

[1664] The server receives text data sent by the user. The received data is analyzed using natural language processing (NLP) algorithms. Here, NLP libraries such as Python's NLTK and spaCy are used to extract the keyword "can't sleep".

[1665] Emotion recognition by an emotion engine

[1666] During the process of analyzing text data, the server uses an emotion engine to recognize the user's emotions. Emotion classification models (such as BERT or RoBERTa) are used to extract emotions like "anxiety" and "stress." Specifically, the text is input to the emotion classification model, and emotion labels are returned as output.

[1667] Identifying the issues and emotions, and determining the role of generative artificial intelligence.

[1668] The server identifies the user's issues and emotions based on the analysis results and emotion recognition results. For example, it might identify "potential sleep disorder" and "anxiety."

[1669] The server assigns the role of "doctor" to the generative artificial intelligence. Specifically, it selects the appropriate role from an internal database and provides the role information to the generative AI model as part of a prompt.

[1670] Generating a response message

[1671] The generative artificial intelligence creates a prompt message according to its assigned role. Here, for example, GPT-3 or a similar generative AI model is used. The prompt message would be something like, "The user is feeling anxious because they can't sleep. Please provide advice from a doctor's perspective."

[1672] Generative artificial intelligence generates a response message based on the input prompt. Based on that prompt, it creates content such as, "That sounds difficult. Let's try a few methods to get a good night's sleep. For example, going to bed at the same time every night or taking some time to relax before bed can be effective."

[1673] The generative artificial intelligence also takes emotion recognition results into consideration and adds the message, "Also, try breathing exercises and light exercise to help you relax. You are not alone; we support you."

[1674] Providing a response message to the user

[1675] The server sends the generated response message to the terminal. Specifically, the message data is sent to the user interface via the HTTP response.

[1676] The device displays received messages in the user interface. Users can see the response messages in the chat box on the screen.

[1677] Specific example

[1678] Example 1: Health-related cases

[1679] The server sends the first message to the user: "Hello! What are you interested in today?"

[1680] User response: "I've been having trouble sleeping lately, and it's making me a little anxious."

[1681] The server analyzes the text data, extracts the keyword "can't sleep," and recognizes the emotion of "anxiety."

[1682] The server assigns the role of "doctor" to the generative artificial intelligence and generates prompts.

[1683] The generative artificial intelligence created the following message: "That must be difficult. Let's try a few things to get a good night's sleep. For example, going to bed at the same time every night and having time to relax before bed can be effective. You can also try breathing exercises or light exercise to help you relax. You are not alone, we are here for you."

[1684] The server sends a response message to the user, which is then displayed on the terminal.

[1685] Example 2: Cases related to finance

[1686] The server sends the user an initial message: "What are you interested in today?"

[1687] User response: "I'm worried because I haven't been managing my household finances well lately."

[1688] The server analyzes the text data, extracts "household finances" and "anxiety," and recognizes the emotion of "anxiety."

[1689] The server assigns the role of "tax accountant" to the generative artificial intelligence and generates prompts.

[1690] The generative AI created the following message: "Household financial management is very important, isn't it? The first step is to meticulously record your income and expenses and reduce unnecessary spending. If you need more specific advice, may I ask for more details? Also, please feel free to contact us anytime to alleviate any anxiety you may feel during this process."

[1691] The server sends a response message to the user, which is then displayed on the terminal.

[1692] Thus, the present invention makes it possible to provide appropriate advice that takes into account the emotional aspects of the user's specific problems.

[1693] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1694] Step 1:

[1695] The device launches the application.

[1696] Input: The terminal the user is operating from.

[1697] Action: The application launches and the user interface is displayed.

[1698] Output: Screen including chat box and send button

[1699] Step 2:

[1700] The server sends the initial message.

[1701] Input: Initial message template on the server

[1702] Action: The server generates the message "Hello! What are you interested in today?" and sends it to the terminal.

[1703] Output: Initial message displayed to the user

[1704] Step 3:

[1705] The user responds to the initial message.

[1706] Input: User's text data (e.g., "I've been having trouble sleeping lately")

[1707] Action: The user enters text and presses the submit button.

[1708] Output: Text data sent to the server

[1709] Step 4:

[1710] The server receives text data sent by the user.

[1711] Input: Text data entered and submitted by the user.

[1712] Operation: The server receives text data and saves it to the database.

[1713] Output: Saved text data

[1714] Step 5:

[1715] The server analyzes the text data using natural language processing (NLP) algorithms.

[1716] Input: Saved text data

[1717] Specific operation: Use Python's NLTK or spaCy to extract specific keywords or phrases (e.g., "can't sleep").

[1718] Output: Keyword frequency data from the analysis results

[1719] Step 6:

[1720] The server uses an emotion engine to recognize the user's emotions.

[1721] Input: User's text data

[1722] Specific operation: Use emotion classification models such as BERT or RoBERTa to extract emotion labels such as "anxiety" and "stress."

[1723] Output: Sentiment labels as recognition results

[1724] Step 7:

[1725] The server identifies the user's issues and emotions based on the analysis results and emotion recognition results.

[1726] Input: Analysis results (keyword frequency data) and sentiment labels

[1727] Specific actions: The issue identified is "possible sleep disorder," and the emotion identified is "anxiety."

[1728] Output: Identification of issues and emotions

[1729] Step 8:

[1730] The server assigns appropriate roles to the generative artificial intelligence.

[1731] Input: Identification results of issues and emotions

[1732] Specific operation: Select an appropriate role (e.g., "doctor") from the internal database and provide role information to the generative artificial intelligence model.

[1733] Output: Role information assigned to the generative artificial intelligence.

[1734] Step 9:

[1735] The generative artificial intelligence generates the response message.

[1736] Input: User's identified issues, emotions, and assigned roles

[1737] Specific operation: Uses a generative AI model such as GPT-3 to generate a response message based on the prompt.

[1738] Example prompt: "The user is experiencing anxiety due to insomnia. Please provide advice from a doctor's perspective."

[1739] Output: Generated response message ("That sounds difficult. Let's try a few things to help you get a good night's sleep...")

[1740] Step 10:

[1741] The server sends the generated response message to the terminal.

[1742] Input: Generated response message

[1743] Specific operation: Message data is sent to the user interface via an HTTP response.

[1744] Output: Response message displayed on the user terminal

[1745] Step 11:

[1746] The response message received by the terminal is displayed on the user interface.

[1747] Input: Response message sent from the server

[1748] Operation: The message is displayed in the chat box and can be viewed by the user.

[1749] Output: Response message displayed to the user

[1750] This series of processes allows users to receive appropriate advice regarding specific issues and emotions.

[1751] (Application Example 2)

[1752] 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".

[1753] Conventional interactive systems often fail to adequately address user needs, as response messages generated using algorithms that identify problems in response to user input frequently do not adequately address user needs. Furthermore, responses that take into account user emotions are not provided, making it difficult to completely resolve user dissatisfaction and anxiety. As a result, improving the user experience has been challenging. This invention aims to solve this problem and provide an interactive system that can address both user problems and emotions.

[1754] 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 receiving text data input by the user, means for analyzing the text data and calculating the frequency of specific keywords and phrases, means for identifying the user's issues and emotions based on the analysis results and assigning an appropriate role to a generative artificial intelligence, means for generating a response message according to the role using the generative artificial intelligence, means for providing the generated response message to the user, and means for adding supplementary information and reassuring messages according to the user's emotions using an emotion recognition engine. This makes it possible to provide specific and useful advice regarding the user's issues and to provide appropriate care for the user's emotions.

[1755] A "user" is a person who uses an interactive system to seek advice on specific issues or emotions.

[1756] "Text data" refers to the written information entered by the user, including the user's concerns and emotions.

[1757] "Analysis" is the process of processing text data to calculate the frequency of specific keywords or phrases.

[1758] An "emotion recognition engine" is an algorithm that extracts a user's emotions from text data.

[1759] "Generative artificial intelligence" is artificial intelligence that generates appropriate response messages in response to a given prompt.

[1760] A "response message" is a message generated by a generative artificial intelligence system that addresses the user's problems and provides emotional support.

[1761] A "server" is a core computer device in a system that receives user input data, analyzes it, assigns roles, and generates and provides response messages.

[1762] A "role" refers to the specific response characteristics and areas of expertise assigned to a generative artificial intelligence.

[1763] "Supplemental information" refers to additional explanations or advice added to a response message that are tailored to the user's emotional state.

[1764] A "reassuring message" is a message provided to care for the user's emotions and offer a sense of mental reassurance.

[1765] A "natural language processing algorithm" is a computational method used to analyze text data to extract keywords and recognize emotions.

[1766] This invention is an interactive system for users to receive advice on specific issues and emotions. The system analyzes text data entered by the user and the user's emotional information, and uses generative artificial intelligence to provide appropriate response messages based on that analysis. The embodiments for carrying out this invention are described in detail below.

[1767] System Configuration

[1768] hardware

[1769] User terminal: Smartphones, tablets, personal computers, etc. These are devices used by users to input text data and view received response messages.

[1770] Server: A computer device that performs backend processing. It receives user input data, analyzes it, assigns roles, and generates and sends response messages.

[1771] software

[1772] Programming language: Python

[1773] Natural language processing algorithms: Use TensorFlow or PyTorch to parse user input data.

[1774] Emotion recognition engine: Extracts emotions from text data using BERT models or similar emotion recognition models.

[1775] Generative AI Model: Using GPT-3 or similar generative models, it generates response messages tailored to the user's needs.

[1776] Web framework: Use Flask or Django to manage communication between the server and the user's device.

[1777] User Interface: We provide an intuitive interface for users using React or Vue.js.

[1778] Program processing

[1779] When the server receives text data from a user, it first analyzes the input data using a natural language processing algorithm. This analysis calculates the frequency of specific keywords and phrases, identifying the user's issues and emotions. At this stage, an emotion recognition engine is used to extract emotions from the user's text data.

[1780] Next, based on the analysis results and emotion recognition results, an appropriate role is assigned to the generative artificial intelligence. The generative artificial intelligence, once assigned a role, generates a response message based on the prompt text. At this time, the emotion recognition results are also taken into consideration, and supplementary information and reassuring messages corresponding to the user's emotions are added.

[1781] Finally, the generated response message is sent from the server to the user's terminal and displayed on the user interface. This allows the user to receive appropriate, emotionally sensitive responses along with specific advice.

[1782] Specific example

[1783] For example, consider a scenario where a user enters "I want a new smartphone, but I'm unsure which one to get" in a virtual store.

[1784] The server first receives this text data and begins analysis. The analysis extracts the keywords "smartphone" and "hesitant," and the emotion recognition engine recognizes the emotion "anxiety."

[1785] Next, the prompt given to the generative AI model will be as follows:

[1786] "I want a new smartphone, but I don't know which one to get. What do you recommend?"

[1787] Generative artificial intelligence generates the following response message based on this prompt:

[1788] "For the latest models, we'll introduce some specific new models and highly-rated models. Also, certain mid-range models are highly regarded."

[1789] The following supplementary information is added to the recognized emotion of "anxiety":

[1790] "Making your first expensive purchase can be nerve-wracking, right? Please feel free to ask any questions you may have. We'll work together to address any concerns you may have."

[1791] In this way, users can gain a sense of security along with specific and helpful advice. This approach significantly improves the user experience.

[1792] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1793] Step 1:

[1794] The server receives text data entered by the user. Text data entered by the user through the device is sent to the server. For example, if a user enters "I want a new smartphone, but I don't know which one to get," that text data is sent to the server.

[1795] Step 2:

[1796] The server analyzes the received text data using natural language processing algorithms. Specifically, the server extracts specific keywords and phrases from the text data and calculates their frequency. For example, the keywords "smartphone" and "hesitant" are extracted, and their frequencies are calculated.

[1797] Step 3:

[1798] The server applies an emotion recognition engine to the analyzed text data. This engine extracts the user's emotions from the text data. For example, the emotion "anxious" might be recognized from the word "hesitant." The input is the analyzed text data, and the output is the extracted emotion information.

[1799] Step 4:

[1800] The server identifies the user's problem based on the analysis results and emotion recognition results, and assigns an appropriate role to the generative artificial intelligence. For example, if the user's problem is identified as "choosing a smartphone" and the user's emotion is identified as "anxiety," the generative artificial intelligence is assigned the role of "product advisor."

[1801] Step 5:

[1802] The server provides a prompt to a generative artificial intelligence (AI) system to generate a response message. The prompt is generated based on the user's input data and emotions. For example, if the prompt "I want a new smartphone, but I don't know which one to get. What do you recommend?" is input to the generative AI system, the response message "For the latest models, we can recommend several specific new models and highly-rated models. Also, certain mid-range models are highly rated." is output.

[1803] Step 6:

[1804] The server adds supplementary information and reassuring messages aimed at emotional support to the generated response message. For example, it might add information such as, "Making your first expensive purchase can be nerve-wracking. Please feel free to ask any questions you may have. We'll work together to address your concerns."

[1805] Step 7:

[1806] The server sends the generated response message to the user's terminal and displays it on the user interface. This allows the user to receive an emotionally sensitive message along with appropriate advice. For example, the user's terminal might display a message such as, "If you're looking for the latest model, we can recommend some of the newest and most highly-rated models. We also have some highly-rated mid-range models. Making your first expensive purchase can be nerve-wracking, so please feel free to ask any questions you may have. We'll help you resolve your concerns."

[1807] 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.

[1808] 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.

[1809] 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.

[1810] 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.

[1811] 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.

[1812] 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.

[1813] 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.

[1814] 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.

[1815] 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."

[1816] 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.

[1817] 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.

[1818] 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.

[1819] 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.

[1820] 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.

[1821] 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.

[1822] 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.

[1823] 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.

[1824] 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.

[1825] 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.

[1826] 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.

[1827] 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.

[1828] The following is further disclosed regarding the embodiments described above.

[1829] (Claim 1)

[1830] A means of receiving text data entered by the user,

[1831] A means for analyzing the aforementioned text data and calculating the frequency of specific keywords or phrases,

[1832] A means for identifying the user's challenges based on the aforementioned analysis results and assigning appropriate roles to a generative artificial intelligence,

[1833] A means for generating response messages according to roles using the aforementioned generative artificial intelligence,

[1834] Means for providing the generated response message to the user,

[1835] A system that includes this.

[1836] (Claim 2)

[1837] The system according to claim 1, characterized in that the generated response message includes specific advice regarding the user's problem.

[1838] (Claim 3)

[1839] The system according to claim 1, characterized in that the means for identifying the user's problem calculates the frequency of keywords related to a specific category using a natural language processing algorithm.

[1840] (Claim 4)

[1841] The system according to claim 1, characterized in that the roles assigned to the generative artificial intelligence correspond to those of professionals such as doctors, counselors, and tax accountants.

[1842] (Claim 5)

[1843] The system according to claim 1, further comprising means for generating prompts to provide the user with optimal advice based on the analysis results.

[1844] "Example 1"

[1845] (Claim 1)

[1846] A means of receiving text data entered by the user,

[1847] A means for analyzing the aforementioned text data and calculating the frequency of specific keywords or phrases,

[1848] A means for identifying the user's challenges based on the aforementioned analysis results and assigning appropriate roles to a generative artificial intelligence,

[1849] A means for generating response messages according to roles using the aforementioned generative artificial intelligence,

[1850] Means for providing the generated response message to the user,

[1851] A means of launching the user interface,

[1852] A means of sending the initial message to the user,

[1853] Means for displaying a response message on the user interface,

[1854] A system that includes this.

[1855] (Claim 2)

[1856] The system according to claim 1, characterized in that the generated response message includes specific advice regarding the user's problem.

[1857] (Claim 3)

[1858] The system according to claim 1, characterized in that the means for identifying the user's problem calculates the frequency of keywords related to a specific category using a natural language processing algorithm.

[1859] "Application Example 1"

[1860] (Claim 1)

[1861] A means of receiving text data entered by the user,

[1862] A means for analyzing the aforementioned text data and calculating the frequency of specific keywords or phrases,

[1863] A means for identifying the user's challenges based on the aforementioned analysis results and assigning appropriate roles to a generative artificial intelligence,

[1864] A means for generating response messages according to roles using the aforementioned generative artificial intelligence,

[1865] A means of recommending content that a user should watch next based on the content they have watched,

[1866] Means for providing the generated response message to the user,

[1867] A system that includes this.

[1868] (Claim 2)

[1869] The system according to claim 1, characterized in that the generated response message includes specific advice regarding the user's problem.

[1870] (Claim 3)

[1871] The system according to claim 1, characterized in that the means for identifying the user's problem calculates the frequency of keywords related to a specific category using a natural language processing algorithm.

[1872] "Example 2 of combining an emotion engine"

[1873] (Claim 1)

[1874] A means of receiving text data entered by the user,

[1875] A means for analyzing the aforementioned text data and calculating the frequency of specific keywords or phrases,

[1876] A means of classifying emotions to recognize user emotions from analyzed text data,

[1877] A means for identifying the user's challenges based on the aforementioned analysis results and emotion recognition results, and for assigning appropriate roles to a generative artificial intelligence,

[1878] A means for generating response messages according to roles using the aforementioned generative artificial intelligence,

[1879] Means for providing the generated response message to the user,

[1880] A system that includes this.

[1881] (Claim 2)

[1882] The system according to claim 1, characterized in that the generated response message includes specific advice and emotional support for the user's problem.

[1883] (Claim 3)

[1884] The system according to claim 1, characterized in that the means for identifying the user's problem calculates the frequency of keywords and sentiment labels associated with a specific category using a natural language processing algorithm and a sentiment classification algorithm.

[1885] "Application example 2 when combining with an emotional engine"

[1886] (Claim 1)

[1887] A means of receiving text data entered by the user,

[1888] A means for analyzing the aforementioned text data and calculating the frequency of specific keywords or phrases,

[1889] A means for identifying the user's challenges and emotions based on the aforementioned analysis results and assigning appropriate roles to a generative artificial intelligence,

[1890] A means for generating response messages according to roles using the aforementioned generative artificial intelligence,

[1891] Means for providing the generated response message to the user and means for adding supplementary information or reassuring messages corresponding to the user's emotions using an emotion recognition engine,

[1892] A system that includes this.

[1893] (Claim 2)

[1894] The system according to claim 1, characterized in that the generated response message includes specific advice and emotional support regarding the user's problem.

[1895] (Claim 3)

[1896] The system according to claim 1, characterized in that the means for identifying the user's issues and emotions uses a natural language processing algorithm to calculate the frequency of keywords associated with specific categories and emotions. [Explanation of Symbols]

[1897] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of receiving text data entered by the user, A means for analyzing the aforementioned text data and calculating the frequency of specific keywords or phrases, A means for identifying the user's challenges based on the aforementioned analysis results and assigning appropriate roles to a generative artificial intelligence, A means for generating response messages according to roles using the aforementioned generative artificial intelligence, Means for providing the generated response message to the user, A system that includes this.

2. The system according to claim 1, characterized in that the generated response message includes specific advice regarding the user's problem.

3. The system according to claim 1, characterized in that the means for identifying the user's problem calculates the frequency of keywords related to a specific category using a natural language processing algorithm.

4. The system according to claim 1, characterized in that the roles assigned to the generative artificial intelligence correspond to those of professionals such as doctors, counselors, and tax accountants.

5. The system according to claim 1, further comprising means for generating prompts to provide optimal advice to the user based on the analysis results.

Images