system

Abstract

We provide the system. [Solution] A means by which the user enters a question or request via an input device, A means of sending the entered question or request to the server via the internet, A means for analyzing the received request, extracting the necessary parameters, and performing preprocessing, A means for sending pre-processed data to an artificial intelligence model and generating a corresponding response, A means for converting the generated response into a user-friendly format and sending it to the user's device, A system including means for confirming a response displayed on a user device.

Description

Technical Field

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

Background Art

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

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] Conventional customer support systems require manual responses and it is difficult to quickly respond to a large number of simultaneous users on a 24-hour basis. Also, since services can only be provided within a fixed time, there is a problem that users cannot receive support at the timing they need. Furthermore, in order to handle a wide variety of questions and troubles, specialized knowledge is required, and there is also a problem of cost for education and training therefor.

Means for Solving the Problems

[0005] The present invention provides a system that includes means for a user to input a question or request via an input device, means for transmitting the input question or request to a server via the Internet, means for analyzing the received request, extracting necessary parameters and performing preprocessing, means for transmitting the preprocessed data to an artificial intelligence model and generating a corresponding response, means for converting the generated response into a user-readable format and transmitting it to a user device, and means for confirming the response displayed on the user device. This system enables rapid and accurate responses to a large number of simultaneous users on a 24-hour basis.

[0006] A "user" refers to an individual or legal entity that uses the system to submit questions or requests.

[0007] An "input device" refers to hardware that a user uses to enter a question or request, such as a mobile device or a personal computer.

[0008] A "server" refers to a computer or group of computers that receives requests from users, analyzes them, processes them, and generates responses.

[0009] A "request" refers to a question or request for a procedure that a user submits through the system.

[0010] The "Internet" refers to a global network used for data communication between user devices and servers.

[0011] "Preprocessing" refers to the process of preparing user requests for analysis and response generation, such as cleaning and tokenizing them.

[0012] An "artificial intelligence model" refers to a machine learning model used to analyze user requests and generate appropriate responses.

[0013] "Response" refers to the answers or information generated by the artificial intelligence model and provided to the user.

[0014] "Natural language processing means" refers to the techniques and methods used to tokenize request data and prepare it for analysis.

[0015] "Tokenization" refers to the process of dividing text data into words or phrases.

[0016] A "user device" refers to a device that a user uses to access a system and check for responses, such as a smartphone or a personal computer. [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] This 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] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11]It is a sequence diagram showing the processing flow of the data processing system in Embodiment 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 Embodiment 2 when the 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 the 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 explained.

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

[0021] In the following embodiments, a RAM (Random Access Memory) with a reference number is a memory in which information is temporarily stored and is used as a 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 relates to a customer service system utilizing generative AI, which allows users to input questions and requests from mobile devices or browsers and receive quick and accurate responses. This system combines a server, user devices, the internet, and an artificial intelligence model.

[0039] User access to the system

[0040] The process begins with the user opening a dedicated app or web browser on their mobile device or computer. The user then enters a question, such as, "What causes my phone to restart frequently?" At this point, the user doesn't need to perform any special operations; they can enter the question using natural language.

[0041] The server receives the request.

[0042] The question entered by the user is sent to the server via the internet. The server receives the HTTP request, parses the request data, and extracts the necessary parameters. For example, if the server receives the question "What causes my phone to restart frequently?", it will parse this question and extract keywords such as "phone," "frequently," "restart," and "cause."

[0043] Preprocessing of request data

[0044] The server cleans the user's question, removing unnecessary spaces and special characters. It also tokenizes the question and performs preprocessing for analysis and response generation. The tokenized data is then converted into a format suitable for artificial intelligence models.

[0045] Inquiries about AI models

[0046] The pre-processed data is sent from the server to the artificial intelligence model. The AI ​​model analyzes the question, understands the user's intent, and generates an appropriate response. For example, the AI ​​model might generate a response such as, "Frequent restarts of your phone could be caused by system errors, battery issues, or app compatibility problems."

[0047] Formatting the response and sending it to the user.

[0048] The generated response is returned to the server. The server formats this response into a user-friendly format and sends it to the user's device. For example, the server might generate a message such as, "Your phone is frequently restarting, which could be caused by a system error, battery issues, or app compatibility problems," and send it to the user's device.

[0049] The user confirms the response.

[0050] The user's device receives a response from the server, which is then displayed to the user. The user can review the displayed response and take any necessary actions. For example, if the user asks, "Why is my phone battery draining so quickly?", the server might respond with information such as, "Rapid battery drain is often caused by background applications, display brightness settings, or natural battery degradation."

[0051] In this way, a customer-only support system utilizing generative AI allows users to receive prompt and accurate support 24 hours a day. Furthermore, this system is expected to improve the efficiency of customer support and reduce costs.

[0052] The following describes the processing flow.

[0053] Step 1: Access the user's system

[0054] Users use their mobile devices or computers to open a dedicated app or web browser and enter a question or request. For example, a user might type, "What causes my phone to restart frequently?"

[0055] Step 2: Send the user's request to the server.

[0056] The information entered by the user is sent to the server via the internet as an HTTP request. During this process, the user device sends the request data to the server in the appropriate format.

[0057] Step 3: The server receives the request.

[0058] The server receives HTTP requests from users and parses their contents. Specifically, it extracts the user's questions and requests from the request body.

[0059] Step 4: Cleaning the request data

[0060] The server cleans the user's question content that it receives. This process removes unnecessary spaces and special characters.

[0061] Step 5: Tokenize the request data

[0062] The server tokenizes the cleaned text. For example, it breaks down the text "What causes my phone to restart frequently?" into words like "phone," "frequently," "restart," and "cause."

[0063] Step 6: Send the tokenized data to the AI ​​model.

[0064] The server sends tokenized data to the artificial intelligence model. The request data is converted into a format that is easy for the AI ​​model to analyze.

[0065] Step 7: Generating responses using the AI ​​model

[0066] The artificial intelligence model analyzes the submitted data and generates appropriate responses to the user's questions. For example, it might generate a response such as, "Frequent restarts of your phone could be due to a system error, battery issues, or app compatibility problems."

[0067] Step 8: The server formats the response.

[0068] The server receives the response from the AI ​​model and formats it into a user-friendly format. The response is formatted to appear as natural-sounding text.

[0069] Step 9: Send the formatted response to the user device.

[0070] The server sends a formatted response to the user's device as an HTTP response. This response provides information in a format that is easy for the user to understand.

[0071] Step 10: User confirms response

[0072] The user device displays the received response, which the user then reviews. For example, it might say, "Your phone is frequently restarting. Possible reasons for this include system errors, battery issues, or app compatibility problems."

[0073] Through the steps described above, a customer-only support system utilizing generative AI can provide quick and accurate responses to user questions and requests.

[0074] (Example 1)

[0075] 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."

[0076] Traditional customer support systems suffered from long response times between users entering questions or requests and receiving a reply, hindering efficient support. Furthermore, response quality was often insufficient, frequently failing to provide users with the information they needed quickly and accurately. Additionally, the heavy burden on customer support staff made 24 / 7 support difficult.

[0077] 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.

[0078] In this invention, the server includes means for a user to input a question or request via an input device, means for transmitting the input question or request to the server via a network, means for analyzing the received request, extracting necessary parameters and performing preprocessing, means for transmitting the preprocessed data to a generating artificial intelligence model and generating a corresponding response, means for converting the generated response into a user-readable format and transmitting it to a user terminal, and means for confirming the response displayed on the user terminal. This enables users to receive prompt and accurate support 24 hours a day. Furthermore, it improves the efficiency of customer support and reduces the burden on staff.

[0079] A "user" refers to a person who uses the system to enter questions or requests.

[0080] An "input device" refers to a device or application used by a user to input questions or requests. This also includes keyboards, touchscreens, and mice.

[0081] "Network" refers to communication lines and infrastructure, including the internet, that provide the means for sending and receiving data.

[0082] A "server" refers to a computer system that has the function of receiving requests from users, analyzing and pre-processing them, generating responses, and sending those responses.

[0083] A "request" refers to a question or request that a user sends to a system via an input device.

[0084] "Analysis" refers to the breakdown and interpretation of data that a server performs in order to understand a received request.

[0085] A "parameter" refers to a specific element or data point extracted from a request.

[0086] "Preprocessing" refers to the process by which the server cleans, tokenizes, and prepares request data for analysis and response generation.

[0087] A "generative artificial intelligence model" refers to a program that uses natural language processing and machine learning to generate appropriate responses based on input data.

[0088] "Response" refers to the information that a server or generative artificial intelligence model generates in response to a request and returns to the user.

[0089] A "user terminal" refers to a device used by a user to access a system and confirm its response. Examples include personal computers, smartphones, and tablets.

[0090] This invention relates to a customer service system utilizing generative AI. This system allows users to input questions and requests via mobile devices or browsers, and receive quick and accurate responses. The system components include a server, user terminals, a network, and a generative artificial intelligence model.

[0091] Specifically, users first open a dedicated app or web browser on their mobile device or computer and enter their question in natural language. For example, a user might enter a question like, "What causes my phone to restart frequently?" At this point, users don't need to perform any special operations; they can enter their question using natural language.

[0092] The server receives user inquiries via the internet. The received inquiries are in the form of HTTP requests, and the server parses these requests to extract the necessary parameters. For example, keywords such as "mobile," "frequently," "restart," and "cause" are obtained as part of the analysis results.

[0093] Next, the server cleanses the received question content, removing unnecessary spaces and special characters. It also tokenizes the question and performs preprocessing for analysis and response generation, converting the tokenized data into a format suitable for the generated artificial intelligence model.

[0094] Once preprocessing is complete, the server sends the tokenized data to the AI ​​model. The AI ​​model analyzes the question, understands the user's intent, and generates an appropriate response. For example, it might generate a response such as, "Frequent restarts of your phone could be caused by system errors, battery issues, or app compatibility problems."

[0095] The generated response is returned to the server, which then formats it into a user-friendly format. The formatted response is then sent to the user's device as a message such as, "Your phone is frequently restarting due to a system error, battery problem, or app compatibility issue."

[0096] Finally, the user's device receives a response from the server and displays it to the user. The user can then review the displayed response and take the necessary actions. For example, if the user asks, "Why is my phone battery draining so quickly?", the server might respond with information such as, "Rapid battery drain is often caused by background applications, display brightness settings, or natural battery degradation."

[0097] As an example of a prompt, when querying a generative AI model with the question, "Why is my phone battery draining so quickly?", you would use a prompt like the following:

[0098] "Generate an answer to the following question: 'Why does my phone battery drain so quickly?'"

[0099] In this way, a customer-only support system utilizing generative AI allows users to receive prompt and accurate support 24 hours a day. Furthermore, this system is expected to improve the efficiency of customer support and reduce costs.

[0100] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0101] Step 1:

[0102] The user opens a dedicated app or web browser using a mobile device or computer. The user enters a question in natural language, such as "What causes my phone to restart frequently?". The user does not need to perform any special operations; they can simply enter the question in natural language. The input is captured on the device in text format. Specifically, the user enters a URL in the browser's address bar, or taps the app icon to launch the app, enters the question in the chat box, and clicks the "Send" button.

[0103] Step 2:

[0104] The question entered by the user is sent to the server via the internet as an HTTP request. The input is text data from the user. The server parses the received request and deciphers the content of the question. Specifically, from the question "What causes my phone to restart frequently?", it extracts keywords such as "phone," "frequently," "restart," and "cause." In terms of specific operations, the server reads the body of the HTTP request and extracts the necessary parameters.

[0105] Step 3:

[0106] The server cleans the received question content, removing unnecessary spaces and special characters. The input is text data containing the keywords extracted in step 2. The server tokenizes the question and preprocesses it for parsing and response generation. For example, the question "What causes my phone to restart frequently?" is split into words such as "phone," "frequently," "restart," and "cause," and extra spaces and special characters are removed. Specifically, the server uses a text processing library to perform cleaning and tokenization.

[0107] Step 4:

[0108] Preprocessed data is sent from the server to the AI ​​model. The input is tokenized and cleaned text data. The AI ​​model analyzes the data, understands the user's intent, and generates an appropriate response. For example, the AI ​​model might generate a response such as, "Frequent restarts of your phone could be caused by system errors, battery issues, or app compatibility issues." In practice, the server sends the data to the AI ​​model's API and retrieves the response.

[0109] Step 5:

[0110] The generated response is returned to the server, which then formats it into a user-friendly format. The input is the response data returned from the AI ​​model. The server creates a message, for example, "Your phone is frequently restarting, which could be caused by a system error, battery issues, or app compatibility problems," and sends it to the user's device. Specifically, the server applies format conversion and display templates.

[0111] Step 6:

[0112] The user's device receives a response from the server and displays it to the user. The input is formatted response data sent from the server. The user can review the displayed response and take necessary actions. Specifically, the user's device displays the received data on the screen and renders the response content using the browser or application's display engine.

[0113] This series of processes allows users to receive prompt and accurate support.

[0114] (Application Example 1)

[0115] 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."

[0116] Existing food delivery services struggle to respond quickly and accurately to customer inquiries and support requests. This results in users spending a long time resolving problems, leading to a poor customer experience. Furthermore, significant labor costs are incurred for customer support, making it inefficient. Therefore, there is a need for a system that can respond quickly and accurately to customer inquiries in food delivery services.

[0117] 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.

[0118] In this invention, the server includes means for a user to input a question or request via an input device, means for transmitting the input question or request to the server via the internet, means for analyzing the received request, extracting necessary parameters and performing preprocessing, means for transmitting the preprocessed data to an artificial intelligence model and generating a corresponding response, means for converting the generated response into a user-readable format and transmitting it to the user device, means for confirming the response displayed on the user device, means for performing tokenization and cleaning for response generation, and means for providing customer support for a food delivery service using natural language processing means. This enables quick and accurate responses to customer inquiries, resulting in improved customer experience and reduced customer support costs.

[0119] "A means by which a user enters a question or request via an input device" refers to an interface that allows a user to enter a question or request in text format using a device such as a smartphone or computer.

[0120] "Means of sending entered questions or requests to a server over the internet" refers to protocols or functions that send questions or requests from a user's device to a server over the internet.

[0121] "Means for parsing received requests, extracting necessary parameters, and performing preprocessing" refers to the process by which a server parses questions and requests received from users and extracts keywords and context.

[0122] "Means for sending pre-processed data to an artificial intelligence model and generating a corresponding response" refers to the process of inputting pre-processed data into an artificial intelligence model to generate an appropriate response to a user's question or request.

[0123] "Means for converting the generated response into a user-friendly format and sending it to the user's device" refers to the process of converting the response generated by the artificial intelligence model into a style and format that is easy for the user to understand, and then sending it to the user's device.

[0124] "Means for confirming the response displayed on the user device" refers to an interface for confirming the generated response displayed on the user's device and determining the next action.

[0125] "Means for tokenizing and cleaning for response generation" refers to a preprocessing step that tokenizes the user's question or request and removes unnecessary characters and symbols.

[0126] "Means for providing customer support for food delivery services using natural language processing" refers to a function that uses natural language processing technology to generate appropriate responses to customer inquiries and problems related to food delivery services.

[0127] This invention relates to a system for automating customer support in food delivery services. The system aims to provide quick and accurate responses to questions and requests entered by users via input devices such as smartphones and personal computers.

[0128] System configuration and operation

[0129] 1. The user accesses the system:

[0130] Users can use a dedicated application to enter questions and requests regarding food delivery in natural language. For example, they can ask questions such as, "My order hasn't arrived."

[0131] 2. The server receives the request:

[0132] The entered questions and requests are sent to the server via the internet. The server receives the HTTP request and parses its contents.

[0133] 3. Preprocessing of request data:

[0134] The server cleans the received request data, removing unnecessary spaces and special characters. Next, it tokenizes the request data, converting it into a format suitable for the artificial intelligence model. This preprocessing step uses natural language processing (NLP) techniques.

[0135] 4. Inquiries to the AI ​​model:

[0136] The pre-processed data is sent from the server to a generating AI model (e.g., OpenAI® GPT series). The AI ​​model analyzes the received data and generates appropriate responses to user questions and requests. For example, if a user asks, "My order hasn't arrived," the AI ​​model will generate a response such as, "Please check your order confirmation email. The email may contain information about the delivery status."

[0137] 5. Formatting the response and sending it to the user:

[0138] The generated response is returned to the server, which then formats it into a user-friendly format. The formatted message is then sent to the user's device, allowing the user to quickly find the appropriate solution and next steps.

[0139] 6. The user confirms the response:

[0140] The user's device receives a response from the server and displays it to the user. The user can then review the displayed response and take any necessary actions.

[0141] Hardware and software to be used

[0142] Server: Uses Amazon Web Services (AWS® registered trademark) EC2 instances.

[0143] Generative AI model: OpenAI GPT-3(registered trademark) / 4 is used.

[0144] Natural Language Processing (NLP): SpaCy will be used.

[0145] Communication protocol: Use HTTPS.

[0146] Database: PostgreSQL will be used.

[0147] Examples of specific cases and prompt statements

[0148] For example, if a user enters the question, "I'm having trouble paying my credit card," the AI ​​model will be queried in the following format:

[0149] Prompt example:

[0150] A user asked, "I'm having trouble with my credit card payment." What is the appropriate solution?

[0151] By using such prompts, the AI ​​model can generate appropriate responses to the user's situation, enabling a rapid response.

[0152] The system configuration and operation described above enable the provision of quick and accurate responses to customer inquiries in food delivery services. This improves the efficiency of customer support and enhances user satisfaction.

[0153] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0154] Step 1:

[0155] Users enter questions or requests using their smartphones or computers. They type their question (e.g., "There's a problem with my credit card payment.") into the text field on their input device and press the submit button. This input text becomes the request data sent to the server.

[0156] Step 2:

[0157] The entered question or request is sent to the server over the internet. The terminal creates an HTTP request and sends a payload containing the input text to a specific endpoint on the server. This request data is received by the server.

[0158] Step 3:

[0159] The server parses the received request, extracts the necessary parameters, and performs preprocessing. Specifically, the server analyzes the input text and extracts keywords such as "credit card," "payment," and "problem." It also removes unnecessary spaces and special characters. As a result, clean, tokenized text data is obtained.

[0160] Step 4:

[0161] The server sends pre-processed data to an artificial intelligence model (generative AI model) to generate a corresponding response. Specifically, it passes pre-processed text data to the generative AI model as a prompt, asking in the format, "The user asked, 'There is a problem with my credit card payment.' What is the appropriate solution?" The AI ​​model analyzes the received data and generates an appropriate response (e.g., "Contact your credit card company or double-check your payment information.").

[0162] Step 5:

[0163] The generated response is returned to the server, which then formats it into a user-friendly format. Specifically, the server converts the response text into HTML or JSON format and generates a message to send to the user's device. This formatted response data is then sent to the user's device.

[0164] Step 6:

[0165] The user device receives a response from the server and displays it to the user. Specifically, the response text (e.g., "Please contact your credit card company or re-verify your payment information.") is displayed on the smartphone or computer screen. The user can then review the displayed response and take appropriate action.

[0166] Through the above processing steps, the food delivery service will be able to provide quick and accurate responses to customer inquiries.

[0167] 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.

[0168] This invention relates to a customer-only support system utilizing generative AI and an emotion engine, which allows users to input questions and requests from mobile devices or browsers and receive quick and accurate responses. This system combines a server, user devices, the internet, an artificial intelligence model, and an emotion engine.

[0169] User access to the system

[0170] Users open a dedicated app or web browser using their mobile device or computer and enter their question or request. For example, a user might type, "My phone keeps restarting frequently after a recent update; could you tell me why?" In this case, the user doesn't need to perform any special operations and can enter their question in natural language.

[0171] The server receives the request.

[0172] The questions entered by the user are sent to the server as HTTP requests via the internet. The server receives these HTTP requests and parses the request data. Specifically, it extracts the user's questions and requests from the request body.

[0173] Request data preprocessing and sentiment analysis

[0174] The server cleans the received user question, removing unnecessary spaces and special characters. Next, it tokenizes the question and performs preprocessing for analysis and response generation. Furthermore, the server's built-in sentiment engine analyzes the sentiment from the user's text. For example, if the user's question is emotionally charged, such as "After the recent update, it keeps restarting frequently and it's really bothering me. Please help," the sentiment engine will recognize emotions such as "confusion" and "difficulty."

[0175] Sending data, including emotions, to AI models

[0176] Pre-processed data and recognized sentiment information are sent from the server to the artificial intelligence model. The AI ​​model generates a response that considers not only the question but also the user's emotional state. For example, it might generate a response such as, "Frequent restarts of your phone could be due to a system error, battery issues, or app compatibility. We apologize for the inconvenience."

[0177] Formatting the response and sending it to the user.

[0178] The generated response is returned to the server. The server formats this response into a user-friendly format and sends it to the user's device. For example, the server might generate a message such as, "Your phone is frequently restarting, which could be due to a system error, battery issues, or app compatibility problems. We apologize for the inconvenience," and send it to the user's device.

[0179] The user confirms the response.

[0180] The user's device receives a response from the server, which is then displayed to the user. The user can review the displayed response and take necessary actions. For example, if the user asks, "Why is my phone battery draining so quickly?", the server might respond with information such as, "Rapid battery drain is often caused by background applications, display brightness settings, or battery degradation."

[0181] In this way, a dedicated customer support system utilizing generative AI and an emotion engine allows users to receive prompt and accurate support 24 hours a day. Furthermore, this system enables the provision of responses tailored to the user's emotions, resulting in a more satisfying customer support experience.

[0182] The following describes the processing flow.

[0183] Step 1: Access the user's system

[0184] The user opens a dedicated app or web browser using a mobile device or computer. The user enters a specific question or request in the input field. For example, "After a recent update, my computer keeps restarting frequently, and I'm having trouble with it. Please tell me the cause."

[0185] Step 2: Send the user's request to the server.

[0186] The information entered by the user is sent to the server via the internet as an HTTP request. During this process, the user device sends the request data to the server in the appropriate format.

[0187] Step 3: The server receives the request.

[0188] The server receives HTTP requests from users and parses their contents. Specifically, it extracts the user's questions and requests from the request body. For example, it might receive a request that says, "After a recent update, my computer keeps restarting frequently, and I'm having trouble with it. Could you tell me the cause?"

[0189] Step 4: Cleaning the request data

[0190] The server cleans the user's question content that it receives. This process removes unnecessary spaces and special characters. For example, it removes extra spaces from the request so that the text data can be parsed accurately.

[0191] Step 5: Tokenize the request data

[0192] The server tokenizes the cleaned text. For example, it breaks down the text "After the recent update, my computer keeps restarting frequently, and I'm having trouble with it. Please tell me the cause" into words like "recently," "after the update," "frequently," "restart," "having trouble," "cause," and "please tell me."

[0193] Step 6: Sentiment analysis of request data

[0194] The emotion engine installed on the server analyzes emotions from the user's text. For example, it recognizes emotions such as "confusion" or "anxiety" from the expression "I'm troubled."

[0195] Step 7: Send the tokenized and sentiment-analyzed data to the AI ​​model.

[0196] The server sends tokenized data and sentiment information to the artificial intelligence model. The request data is converted into a format that is easy for the AI ​​model to analyze.

[0197] Step 8: Generating responses using the AI ​​model

[0198] The artificial intelligence model analyzes the transmitted data and generates appropriate responses to the user's questions. In doing so, it adjusts the response content considering the user's emotional state. For example, it might generate a response such as, "Frequent restarts of your phone could be due to a system error, battery issues, or app compatibility problems. We apologize for the inconvenience."

[0199] Step 9: Formatting the response

[0200] The server receives the response from the AI ​​model and formats it into a user-friendly format. It formats the response so that it reads like natural language. For example, it might be formatted to read, "Frequent restarts of your phone could be due to a system error, battery issues, or app compatibility problems. We apologize for the inconvenience."

[0201] Step 10: Send the formatted response to the user device.

[0202] The server sends a formatted response to the user's device as an HTTP response. This response provides information in a format that is easy for the user to understand.

[0203] Step 11: User confirms response

[0204] The user device displays the received response, which the user then acknowledges. For example, a message might appear stating, "Frequent restarts on your phone could be caused by a system error, battery issues, or app compatibility problems. We apologize for the inconvenience."

[0205] Thus, a customer-only support system utilizing generative AI and an emotion engine can provide quick and accurate responses to user questions and requests, while also enabling support that takes user emotions into consideration.

[0206] (Example 2)

[0207] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0208] Traditional customer service systems require quick and accurate responses to user inquiries, but often fail to adequately consider the user's feelings or the specific nature of their problem. This can lead to decreased user satisfaction and longer problem-solving times. Furthermore, uniform responses that disregard emotions tend to be unsatisfactory for users.

[0209] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0210] In this invention, the server includes means for a user to input a question or request via an input device, means for transmitting the input question or request to the server via the Internet, means for cleaning and tokenizing the received request and extracting necessary parameters for preprocessing, means for extracting sentiment information from the received request using a sentiment analysis engine, means for transmitting the preprocessed data and sentiment information to an artificial intelligence model and generating a corresponding response, means for converting the generated response into a user-friendly format and transmitting it to the user device, and means for confirming the response displayed on the user device. This enables accurate and highly satisfying responses to user inquiries that take sentiment into account.

[0211] A "user" refers to anyone who uses the system and enters a question or request.

[0212] "Input device" refers to a terminal used by a user to enter a question or request, and includes mobile devices and personal computers.

[0213] The "Internet" refers to a communication system that connects computer networks around the world, and is the infrastructure for data communication.

[0214] A "server" refers to a computer system that receives, analyzes, and processes requests from users to provide responses.

[0215] A "request" refers to a question or request that a user sends to a server using an input device.

[0216] "Cleaning" refers to the process of removing unnecessary spaces and special characters from received text data.

[0217] "Tokenization" refers to a natural language processing technique that divides text into words or phrases.

[0218] A "sentiment analysis engine" refers to computer software used to extract emotional information from text data.

[0219] An "artificial intelligence model" refers to an algorithm and a trained model used to generate appropriate responses to user questions and requests.

[0220] A "response" refers to a text message that a server generates in response to a user's request and ultimately sends to the user's device.

[0221] "User device" refers to a terminal used by a user to confirm a response, and includes mobile devices or personal computers.

[0222] "Preprocessing" refers to a series of data formatting tasks performed in preparation for data analysis and response generation.

[0223] A "prompt" refers to a text statement used to input specific instructions or questions to an artificial intelligence model.

[0224] This invention relates to a customer service system utilizing generative AI and an emotion analysis engine. The system aims to provide users with prompt and accurate responses to questions and requests entered via mobile devices or browsers. The system combines a server, user devices, the internet, an artificial intelligence model, and an emotion analysis engine.

[0225] Users open a dedicated app or web browser on their mobile device or computer and enter their question or request. For example, a user might type, "My phone keeps restarting frequently after a recent update; could you tell me why?" In this process, users do not need to perform any special operations and can enter their questions in natural language.

[0226] The question entered by the user is sent to the server as an HTTP request via the internet. The server receives the HTTP request and parses the request data. Specifically, it extracts the user's question and request from the request body. The request is sent to the server in JSON format, and from the data in the request body, it extracts "Question: My phone keeps restarting frequently after a recent update. Could you tell me the reason?"

[0227] Next, the server cleans the received question content, removing unnecessary spaces and special characters. Specifically, it uses Python for the cleaning process. Then, it uses a natural language processing library (e.g., NLTK or SpaCy) to tokenize the question content, splitting the sentence into words. Furthermore, it uses the server's built-in sentiment analysis engine (e.g., Microsoft® Azure® Cognitive Services) to analyze the sentiment from the user's text. If the user's question is emotionally charged, such as "After the recent update, it keeps restarting frequently and it's really bothering me. Please help," the sentiment engine will recognize emotions such as "confused" or "difficult."

[0228] Preprocessed data and recognized sentiment information are sent from the server to the artificial intelligence model. The AI ​​model generates a response considering the question and sentiment state. For example, using OpenAI's GPT-3, it might generate a response such as, "Frequent restarts of your phone could be due to a system error, battery issues, or app compatibility. We apologize for the inconvenience."

[0229] The generated response is sent back to the server, which then formats it into a user-friendly format. Specifically, it is converted to HTML format and sent to the user's device. For example, the server might generate a message such as, "Your phone is frequently restarting, which could be due to a system error, battery issues, or app compatibility problems. We apologize for the inconvenience," and send it to the user's device.

[0230] Finally, the user's device receives a response from the server, which is then displayed to the user. The user can review the displayed response and take any necessary actions. For example, if the user asks, "Why is my phone battery draining so quickly?", the server might respond with, "Rapid battery drain is often caused by background applications, display brightness settings, or battery degradation."

[0231] In this way, a dedicated customer support system utilizing generative AI and emotion analysis engines allows users to receive prompt and accurate support 24 hours a day. Furthermore, by providing responses tailored to the user's emotions, a more satisfying customer support experience is achieved.

[0232] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0233] Step 1:

[0234] User access to the system

[0235] Users use a mobile device or computer to open a dedicated app or web browser and enter a question or request. The input is in natural language and includes specific details, such as "My phone keeps restarting frequently after a recent update; could you tell me why?" This input data is saved on the device via the dedicated app or web browser.

[0236] Input: User's question or request

[0237] Output: Questions or requests stored on the input device

[0238] Step 2:

[0239] The server receives the request.

[0240] The user's input question is sent to the server as an HTTP request via the internet. The server receives this HTTP request and extracts the question content from the request body in a data format such as JSON. To prepare the parameters necessary for analysis, the request data is parsed, and the user's question and request are extracted as text data.

[0241] Input: HTTP request sent over the internet

[0242] Output: Extracted text data

[0243] Step 3:

[0244] Request data preprocessing and sentiment analysis

[0245] The server cleans the received question content, removing unnecessary spaces and special characters. Next, it uses a natural language processing library (e.g., NLTK or SpaCy) to tokenize the question content, splitting it into words. Then, it uses a sentiment analysis engine (e.g., Microsoft Azure Cognitive Services) deployed on the server to extract sentiment information from the user's text. For example, it might obtain sentiment information such as "confused" or "difficult."

[0246] Input: Extracted text data

[0247] Output: Cleaned text data and sentiment information

[0248] Step 4:

[0249] Data transmission and response generation

[0250] Pre-processed data and sentiment information are sent from the server to the artificial intelligence model. The AI ​​model uses this data to generate a response, taking into account the content of the question and the emotional state. For example, it might generate a specific response such as, "Frequent restarts of your phone could be due to a system error, battery issues, or app compatibility. We apologize for the inconvenience."

[0251] Input: Cleaned text data and sentiment information

[0252] Output: Generated response text

[0253] Step 5:

[0254] Formatting the response and sending it to the user.

[0255] The generated response is returned to the server, where it is formatted into a user-friendly format. Specifically, the response is converted to HTML or JSON format and the display format is adjusted. The formatted response is then sent to the user's device.

[0256] Input: Generated response text

[0257] Output: Formatted and formatted response message

[0258] Step 6:

[0259] The user confirms the response.

[0260] The user's device displays the response received from the server and makes it visible to the user. The user can review the displayed response and take any necessary actions. For example, if the user asks, "Why is my phone battery draining so quickly?", the provided response might say, "Rapid battery drain is often caused by background applications, display brightness settings, or battery degradation."

[0261] Input: Formatted response message

[0262] Output: Displayed response message

[0263] Through each of the above steps, users can receive prompt and accurate support, and are provided with highly satisfying responses that are tailored to their needs and feelings.

[0264] (Application Example 2)

[0265] 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".

[0266] Traditional customer support systems lacked the ability to respond in a way that considered the user's emotional state and to recommend appropriate content, limiting their potential for improving user satisfaction. Furthermore, the wide variety of information displayed on user devices necessitated a unified interface for responses.

[0267] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for analyzing the received request, extracting necessary parameters and performing preprocessing; means for transmitting the preprocessed data and data extracted from the user's emotional state through sentiment analysis to an artificial intelligence model and generating a corresponding response; and means for recommending content based on the user's past data and current emotional state. This makes it possible to provide detailed responses based on the user's emotional state and suggest content that is suitable for the user's needs.

[0268] "Means by which users input questions or requests via input devices" refers to a function that allows users to input questions or requests using natural language via input devices such as smartphones, tablets, or personal computers.

[0269] "Means for sending entered questions or requests to a server via the Internet" refers to a function that sends user questions or requests from an input device to a server via the Internet in the form of an HTTP request or similar.

[0270] "Means for parsing received requests, extracting necessary parameters, and performing preprocessing" refers to a function that allows a server to analyze questions and requests received from a user, remove unnecessary data, and extract necessary parameters.

[0271] "Means for sending pre-processed data and data extracted from the user's emotional state through sentiment analysis to an artificial intelligence model and generating a corresponding response" refers to a function for sending pre-processed data and data from the user's emotional state analyzed by a sentiment analysis engine to an artificial intelligence model and generating a response that takes the user's emotions into consideration.

[0272] "Means for converting the generated response into a user-friendly format and sending it to the user's device" refers to a function that formats the response generated by the artificial intelligence model into a form that is easy for the user to understand and displays it on the user's device.

[0273] "Means for confirming responses displayed on the user device" refers to a function that allows the user to confirm the response sent from the server on their own device.

[0274] "Means of recommending content based on a user's past data and current emotional state" refers to a function that selects and recommends the most suitable content to a user by considering the user's historical data and emotional state.

[0275] This invention is a system in which a user inputs a question or request via an input device, and appropriate responses and content recommendations are provided based on that content. The details of the system that implements this application example are described below.

[0276] Hardware and software to be used

[0277] The system utilizes servers, user devices such as smartphones and head-mounted displays, Python, TENSORFLOW®, an emotion analysis engine, and a content management system (CMS).

[0278] System Configuration

[0279] The user inputs questions or requests from a dedicated app or web browser using a smartphone or a head-mounted display. The user's input is made in natural language and does not require special operations.

[0280] The server analyzes the requests received via the Internet, extracts the necessary parameters after removing unnecessary data, and then analyzes the user's emotional state using an emotion analysis engine. For example, when the user inputs "I want to relax today", the emotion analysis engine recognizes the emotion of "relaxed".

[0281] The preprocessed data and emotional state are sent to an artificial intelligence model. This model generates optimal responses and content recommendations considering the user's emotional state. The generated responses and recommended content are returned to the server and formatted into a form that is easy for the user to understand. Finally, this is sent to the user device, and the user checks the displayed responses and recommended content.

[0282] Specific Example

[0283] For example, when the user inputs "I'm feeling down today, so I want to watch a video that can cheer me up", the server analyzes this input with an emotion analysis engine and recognizes the emotional state as "feeling down". The generation AI model considers this emotional state, extracts and recommends videos that can cheer the user up from the content management system. The server sends a list of recommended content to the user, and the user can choose the video they want to watch from it.

[0284] Example of Prompt Sentence

[0285] User Input: "I'm feeling down today, so I want to watch a video that can cheer me up."

[0286] Generation AI Input Prompt: "The input emotion is analyzed as 'feeling down'. Please recommend a video that can cheer the user up."

[0287] In this way, the system can provide optimal responses and content recommendations based on the user's emotional state and past data.

[0288] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0289] Step 1:

[0290] The user enters a question or request via an input device. The user uses a smartphone or head-mounted display to enter the question or request in natural language. This input data is transmitted to a server via the internet by the application on the input device.

[0291] Input: User's question or request in natural language.

[0292] Output: HTTP request sent to the server over the internet

[0293] Step 2:

[0294] The server parses the received request, preprocessing it by removing unnecessary spaces and special characters. It then tokenizes the received data, converting it into structured data.

[0295] Input: Received HTTP request

[0296] Output: Preprocessed structured data

[0297] Specific operation: The server parses the request body, removes unnecessary spaces and special characters using natural language processing, and tokenizes it.

[0298] Step 3:

[0299] The server uses an emotion analysis engine to analyze the user's emotional state from pre-processed data. The analysis results provide the type and intensity of the emotion.

[0300] Input: Preprocessed structured data

[0301] Output: Sentiment analysis result (type and degree of sentiment)

[0302] Specific operation: The server starts the sentiment analysis engine and analyzes the user's text data to extract the sentiment state.

[0303] Step 4:

[0304] Send the preprocessed data and the sentiment analysis result to the artificial intelligence model, and generate a response and content recommendation considering the user's sentiment state. Here, the generation AI model generates an appropriate response based on the prompt.

[0305] Input: Preprocessed structured data, sentiment analysis result

[0306] Output: Generated response and content recommendation

[0307] Specific operation: The server inputs the data into the generation AI model, generates a prompt sentence, and obtains the optimal response and content recommendation.

[0308] Step 5:

[0309] Format the generated response and content recommendation and convert them into a user-friendly format. The server sends this formatted data to the user device.

[0310] Input: Generated response and content recommendation

[0311] Output: Formatted response and content recommendation

[0312] Specific operation: The server converts the generated data into HTML or JSON format and sends it to the user device.

[0313] Step 6:

[0314] The user's device receives responses and content recommendations sent from the server and displays them to the user. The user reviews them and takes the necessary action (such as watching a video).

[0315] Input: Formatted response and content recommendations sent from the server

[0316] Output: Responses and content recommendations displayed on the user device.

[0317] Specific action: Display the data received by the user's device and allow the user to review it.

[0318] The above describes the processing flow of the system program that implements the application example.

[0319] 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.

[0320] 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.

[0321] 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.

[0322] [Second Embodiment]

[0323] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0324] 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.

[0325] 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).

[0326] 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.

[0327] 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.

[0328] 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).

[0329] 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.

[0330] 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.

[0331] 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.

[0332] 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.

[0333] 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.

[0334] 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".

[0335] This invention relates to a customer service system utilizing generative AI, which allows users to input questions and requests from mobile devices or browsers and receive quick and accurate responses. This system combines a server, user devices, the internet, and an artificial intelligence model.

[0336] User access to the system

[0337] The process begins with the user opening a dedicated app or web browser on their mobile device or computer. The user then enters a question, such as, "What causes my phone to restart frequently?" At this point, the user doesn't need to perform any special operations; they can enter the question using natural language.

[0338] The server receives the request.

[0339] The question entered by the user is sent to the server via the internet. The server receives the HTTP request, parses the request data, and extracts the necessary parameters. For example, if the server receives the question "What causes my phone to restart frequently?", it will parse this question and extract keywords such as "phone," "frequently," "restart," and "cause."

[0340] Preprocessing of request data

[0341] The server cleans the user's question, removing unnecessary spaces and special characters. It also tokenizes the question and performs preprocessing for analysis and response generation. The tokenized data is then converted into a format suitable for artificial intelligence models.

[0342] Inquiries about AI models

[0343] The pre-processed data is sent from the server to the artificial intelligence model. The AI ​​model analyzes the question, understands the user's intent, and generates an appropriate response. For example, the AI ​​model might generate a response such as, "Frequent restarts of your phone could be caused by system errors, battery issues, or app compatibility problems."

[0344] Formatting the response and sending it to the user.

[0345] The generated response is returned to the server. The server formats this response into a user-friendly format and sends it to the user's device. For example, the server might generate a message such as, "Your phone is frequently restarting, which could be caused by a system error, battery issues, or app compatibility problems," and send it to the user's device.

[0346] The user confirms the response.

[0347] The user's device receives a response from the server, which is then displayed to the user. The user can review the displayed response and take any necessary actions. For example, if the user asks, "Why is my phone battery draining so quickly?", the server might respond with information such as, "Rapid battery drain is often caused by background applications, display brightness settings, or natural battery degradation."

[0348] In this way, a customer-only support system utilizing generative AI allows users to receive prompt and accurate support 24 hours a day. Furthermore, this system is expected to improve the efficiency of customer support and reduce costs.

[0349] The following describes the processing flow.

[0350] Step 1: Access the user's system

[0351] Users use their mobile devices or computers to open a dedicated app or web browser and enter a question or request. For example, a user might type, "What causes my phone to restart frequently?"

[0352] Step 2: Send the user's request to the server.

[0353] The information entered by the user is sent to the server via the internet as an HTTP request. During this process, the user device sends the request data to the server in the appropriate format.

[0354] Step 3: The server receives the request.

[0355] The server receives HTTP requests from users and parses their contents. Specifically, it extracts the user's questions and requests from the request body.

[0356] Step 4: Cleaning the request data

[0357] The server cleans the user's question content that it receives. This process removes unnecessary spaces and special characters.

[0358] Step 5: Tokenize the request data

[0359] The server tokenizes the cleaned text. For example, it breaks down the text "What causes my phone to restart frequently?" into words like "phone," "frequently," "restart," and "cause."

[0360] Step 6: Send the tokenized data to the AI ​​model.

[0361] The server sends tokenized data to the artificial intelligence model. The request data is converted into a format that is easy for the AI ​​model to analyze.

[0362] Step 7: Generating responses using the AI ​​model

[0363] The artificial intelligence model analyzes the submitted data and generates appropriate responses to the user's questions. For example, it might generate a response such as, "Frequent restarts of your phone could be due to a system error, battery issues, or app compatibility problems."

[0364] Step 8: The server formats the response.

[0365] The server receives the response from the AI ​​model and formats it into a user-friendly format. The response is formatted to appear as natural-sounding text.

[0366] Step 9: Send the formatted response to the user device.

[0367] The server sends a formatted response to the user's device as an HTTP response. This response provides information in a format that is easy for the user to understand.

[0368] Step 10: User confirms response

[0369] The user device displays the received response, which the user then reviews. For example, it might say, "Your phone is frequently restarting. Possible reasons for this include system errors, battery issues, or app compatibility problems."

[0370] Through the steps described above, a customer-only support system utilizing generative AI can provide quick and accurate responses to user questions and requests.

[0371] (Example 1)

[0372] 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."

[0373] Traditional customer support systems suffered from long response times between users entering questions or requests and receiving a reply, hindering efficient support. Furthermore, response quality was often insufficient, frequently failing to provide users with the information they needed quickly and accurately. Additionally, the heavy burden on customer support staff made 24 / 7 support difficult.

[0374] 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.

[0375] In this invention, the server includes means for a user to input a question or request via an input device, means for transmitting the input question or request to the server via a network, means for analyzing the received request, extracting necessary parameters and performing preprocessing, means for transmitting the preprocessed data to a generating artificial intelligence model and generating a corresponding response, means for converting the generated response into a user-readable format and transmitting it to a user terminal, and means for confirming the response displayed on the user terminal. This enables users to receive prompt and accurate support 24 hours a day. Furthermore, it improves the efficiency of customer support and reduces the burden on staff.

[0376] A "user" refers to a person who uses the system to enter questions or requests.

[0377] An "input device" refers to a device or application used by a user to input questions or requests. This also includes keyboards, touchscreens, and mice.

[0378] "Network" refers to communication lines and infrastructure, including the internet, that provide the means for sending and receiving data.

[0379] A "server" refers to a computer system that has the function of receiving requests from users, analyzing and pre-processing them, generating responses, and sending those responses.

[0380] A "request" refers to a question or request that a user sends to a system via an input device.

[0381] "Analysis" refers to the breakdown and interpretation of data that a server performs in order to understand a received request.

[0382] A "parameter" refers to a specific element or data point extracted from a request.

[0383] "Preprocessing" refers to the process by which the server cleans, tokenizes, and prepares request data for analysis and response generation.

[0384] A "generative artificial intelligence model" refers to a program that uses natural language processing and machine learning to generate appropriate responses based on input data.

[0385] "Response" refers to the information that a server or generative artificial intelligence model generates in response to a request and returns to the user.

[0386] A "user terminal" refers to a device used by a user to access a system and confirm its response. Examples include personal computers, smartphones, and tablets.

[0387] This invention relates to a customer service system utilizing generative AI. This system allows users to input questions and requests via mobile devices or browsers, and receive quick and accurate responses. The system components include a server, user terminals, a network, and a generative artificial intelligence model.

[0388] Specifically, users first open a dedicated app or web browser on their mobile device or computer and enter their question in natural language. For example, a user might enter a question like, "What causes my phone to restart frequently?" At this point, users don't need to perform any special operations; they can enter their question using natural language.

[0389] The server receives user inquiries via the internet. The received inquiries are in the form of HTTP requests, and the server parses these requests to extract the necessary parameters. For example, keywords such as "mobile," "frequently," "restart," and "cause" are obtained as part of the analysis results.

[0390] Next, the server cleanses the received question content, removing unnecessary spaces and special characters. It also tokenizes the question and performs preprocessing for analysis and response generation, converting the tokenized data into a format suitable for the generated artificial intelligence model.

[0391] Once preprocessing is complete, the server sends the tokenized data to the AI ​​model. The AI ​​model analyzes the question, understands the user's intent, and generates an appropriate response. For example, it might generate a response such as, "Frequent restarts of your phone could be caused by system errors, battery issues, or app compatibility problems."

[0392] The generated response is returned to the server, which then formats it into a user-friendly format. The formatted response is then sent to the user's device as a message such as, "Your phone is frequently restarting due to a system error, battery problem, or app compatibility issue."

[0393] Finally, the user's device receives a response from the server and displays it to the user. The user can then review the displayed response and take the necessary actions. For example, if the user asks, "Why is my phone battery draining so quickly?", the server might respond with information such as, "Rapid battery drain is often caused by background applications, display brightness settings, or natural battery degradation."

[0394] As an example of a prompt, when querying a generative AI model with the question, "Why is my phone battery draining so quickly?", you would use a prompt like the following:

[0395] "Generate an answer to the following question: 'Why does my phone battery drain so quickly?'"

[0396] In this way, a customer-only support system utilizing generative AI allows users to receive prompt and accurate support 24 hours a day. Furthermore, this system is expected to improve the efficiency of customer support and reduce costs.

[0397] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0398] Step 1:

[0399] The user opens a dedicated app or web browser using a mobile device or computer. The user enters a question in natural language, such as "What causes my phone to restart frequently?". The user does not need to perform any special operations; they can simply enter the question in natural language. The input is captured on the device in text format. Specifically, the user enters a URL in the browser's address bar, or taps the app icon to launch the app, enters the question in the chat box, and clicks the "Send" button.

[0400] Step 2:

[0401] The question entered by the user is sent to the server via the internet as an HTTP request. The input is text data from the user. The server parses the received request and deciphers the content of the question. Specifically, from the question "What causes my phone to restart frequently?", it extracts keywords such as "phone," "frequently," "restart," and "cause." In terms of specific operations, the server reads the body of the HTTP request and extracts the necessary parameters.

[0402] Step 3:

[0403] The server cleans the received question content, removing unnecessary spaces and special characters. The input is text data containing the keywords extracted in step 2. The server tokenizes the question and preprocesses it for parsing and response generation. For example, the question "What causes my phone to restart frequently?" is split into words such as "phone," "frequently," "restart," and "cause," and extra spaces and special characters are removed. Specifically, the server uses a text processing library to perform cleaning and tokenization.

[0404] Step 4:

[0405] Preprocessed data is sent from the server to the AI ​​model. The input is tokenized and cleaned text data. The AI ​​model analyzes the data, understands the user's intent, and generates an appropriate response. For example, the AI ​​model might generate a response such as, "Frequent restarts of your phone could be caused by system errors, battery issues, or app compatibility issues." In practice, the server sends the data to the AI ​​model's API and retrieves the response.

[0406] Step 5:

[0407] The generated response is returned to the server, which then formats it into a user-friendly format. The input is the response data returned from the AI ​​model. The server creates a message, for example, "Your phone is frequently restarting, which could be caused by a system error, battery issues, or app compatibility problems," and sends it to the user's device. Specifically, the server applies format conversion and display templates.

[0408] Step 6:

[0409] The user's device receives a response from the server and displays it to the user. The input is formatted response data sent from the server. The user can review the displayed response and take necessary actions. Specifically, the user's device displays the received data on the screen and renders the response content using the browser or application's display engine.

[0410] This series of processes allows users to receive prompt and accurate support.

[0411] (Application Example 1)

[0412] 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."

[0413] Existing food delivery services struggle to respond quickly and accurately to customer inquiries and support requests. This results in users spending a long time resolving problems, leading to a poor customer experience. Furthermore, significant labor costs are incurred for customer support, making it inefficient. Therefore, there is a need for a system that can respond quickly and accurately to customer inquiries in food delivery services.

[0414] 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.

[0415] In this invention, the server includes means for a user to input a question or request via an input device, means for transmitting the input question or request to the server via the internet, means for analyzing the received request, extracting necessary parameters and performing preprocessing, means for transmitting the preprocessed data to an artificial intelligence model and generating a corresponding response, means for converting the generated response into a user-readable format and transmitting it to the user device, means for confirming the response displayed on the user device, means for performing tokenization and cleaning for response generation, and means for providing customer support for a food delivery service using natural language processing means. This enables quick and accurate responses to customer inquiries, resulting in improved customer experience and reduced customer support costs.

[0416] "A means by which a user enters a question or request via an input device" refers to an interface that allows a user to enter a question or request in text format using a device such as a smartphone or computer.

[0417] "Means of sending entered questions or requests to a server over the internet" refers to protocols or functions that send questions or requests from a user's device to a server over the internet.

[0418] "Means for parsing received requests, extracting necessary parameters, and performing preprocessing" refers to the process by which a server parses questions and requests received from users and extracts keywords and context.

[0419] "Means for sending pre-processed data to an artificial intelligence model and generating a corresponding response" refers to the process of inputting pre-processed data into an artificial intelligence model to generate an appropriate response to a user's question or request.

[0420] "Means for converting the generated response into a user-friendly format and sending it to the user's device" refers to the process of converting the response generated by the artificial intelligence model into a style and format that is easy for the user to understand, and then sending it to the user's device.

[0421] "Means for confirming the response displayed on the user device" refers to an interface for confirming the generated response displayed on the user's device and determining the next action.

[0422] "Means for tokenizing and cleaning for response generation" refers to a preprocessing step that tokenizes the user's question or request and removes unnecessary characters and symbols.

[0423] "Means for providing customer support for food delivery services using natural language processing" refers to a function that uses natural language processing technology to generate appropriate responses to customer inquiries and problems related to food delivery services.

[0424] This invention relates to a system for automating customer support in food delivery services. The system aims to provide quick and accurate responses to questions and requests entered by users via input devices such as smartphones and personal computers.

[0425] System configuration and operation

[0426] 1. The user accesses the system:

[0427] Users can use a dedicated application to enter questions and requests regarding food delivery in natural language. For example, they can ask questions such as, "My order hasn't arrived."

[0428] 2. The server receives the request:

[0429] The entered questions and requests are sent to the server via the internet. The server receives the HTTP request and parses its contents.

[0430] 3. Preprocessing of request data:

[0431] The server cleans the received request data, removing unnecessary spaces and special characters. Next, it tokenizes the request data, converting it into a format suitable for the artificial intelligence model. This preprocessing step uses natural language processing (NLP) techniques.

[0432] 4. Inquiries to the AI ​​model:

[0433] The pre-processed data is sent from the server to a generating AI model (e.g., the OpenAI GPT series). The AI ​​model analyzes the received data and generates appropriate responses to user questions and requests. For example, if a user asks, "My order hasn't arrived," the AI ​​model will generate a response such as, "Please check your order confirmation email. The email may contain information about the delivery status."

[0434] 5. Formatting the response and sending it to the user:

[0435] The generated response is returned to the server, which then formats it into a user-friendly format. The formatted message is then sent to the user's device, allowing the user to quickly find the appropriate solution and next steps.

[0436] 6. The user confirms the response:

[0437] The user's device receives a response from the server and displays it to the user. The user can then review the displayed response and take any necessary actions.

[0438] Hardware and software to be used

[0439] Server: Use an Amazon Web Services (AWS) EC2 instance.

[0440] Generative AI model: OpenAI GPT-3 / 4 is used.

[0441] Natural Language Processing (NLP): SpaCy will be used.

[0442] Communication protocol: Use HTTPS.

[0443] Database: PostgreSQL will be used.

[0444] Examples of specific cases and prompt statements

[0445] For example, if a user enters the question, "I'm having trouble paying my credit card," the AI ​​model will be queried in the following format:

[0446] Prompt example:

[0447] A user asked, "I'm having trouble with my credit card payment." What is the appropriate solution?

[0448] By using such prompts, the AI ​​model can generate appropriate responses to the user's situation, enabling a rapid response.

[0449] The system configuration and operation described above enable the provision of quick and accurate responses to customer inquiries in food delivery services. This improves the efficiency of customer support and enhances user satisfaction.

[0450] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0451] Step 1:

[0452] Users enter questions or requests using their smartphones or computers. They type their question (e.g., "There's a problem with my credit card payment.") into the text field on their input device and press the submit button. This input text becomes the request data sent to the server.

[0453] Step 2:

[0454] The entered question or request is sent to the server over the internet. The terminal creates an HTTP request and sends a payload containing the input text to a specific endpoint on the server. This request data is received by the server.

[0455] Step 3:

[0456] The server parses the received request, extracts the necessary parameters, and performs preprocessing. Specifically, the server analyzes the input text and extracts keywords such as "credit card," "payment," and "problem." It also removes unnecessary spaces and special characters. As a result, clean, tokenized text data is obtained.

[0457] Step 4:

[0458] The server sends pre-processed data to an artificial intelligence model (generative AI model) to generate a corresponding response. Specifically, it passes pre-processed text data to the generative AI model as a prompt, asking in the format, "The user asked, 'There is a problem with my credit card payment.' What is the appropriate solution?" The AI ​​model analyzes the received data and generates an appropriate response (e.g., "Contact your credit card company or double-check your payment information.").

[0459] Step 5:

[0460] The generated response is returned to the server, which then formats it into a user-friendly format. Specifically, the server converts the response text into HTML or JSON format and generates a message to send to the user's device. This formatted response data is then sent to the user's device.

[0461] Step 6:

[0462] The user device receives a response from the server and displays it to the user. Specifically, the response text (e.g., "Please contact your credit card company or re-verify your payment information.") is displayed on the smartphone or computer screen. The user can then review the displayed response and take appropriate action.

[0463] Through the above processing steps, the food delivery service will be able to provide quick and accurate responses to customer inquiries.

[0464] 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.

[0465] This invention relates to a customer-only support system utilizing generative AI and an emotion engine, which allows users to input questions and requests from mobile devices or browsers and receive quick and accurate responses. This system combines a server, user devices, the internet, an artificial intelligence model, and an emotion engine.

[0466] User access to the system

[0467] Users open a dedicated app or web browser using their mobile device or computer and enter their question or request. For example, a user might type, "My phone keeps restarting frequently after a recent update; could you tell me why?" In this case, the user doesn't need to perform any special operations and can enter their question in natural language.

[0468] The server receives the request.

[0469] The questions entered by the user are sent to the server as HTTP requests via the internet. The server receives these HTTP requests and parses the request data. Specifically, it extracts the user's questions and requests from the request body.

[0470] Request data preprocessing and sentiment analysis

[0471] The server cleans the received user question, removing unnecessary spaces and special characters. Next, it tokenizes the question and performs preprocessing for analysis and response generation. Furthermore, the server's built-in sentiment engine analyzes the sentiment from the user's text. For example, if the user's question is emotionally charged, such as "After the recent update, it keeps restarting frequently and it's really bothering me. Please help," the sentiment engine will recognize emotions such as "confusion" and "difficulty."

[0472] Sending data, including emotions, to AI models

[0473] Pre-processed data and recognized sentiment information are sent from the server to the artificial intelligence model. The AI ​​model generates a response that considers not only the question but also the user's emotional state. For example, it might generate a response such as, "Frequent restarts of your phone could be due to a system error, battery issues, or app compatibility. We apologize for the inconvenience."

[0474] Formatting the response and sending it to the user.

[0475] The generated response is returned to the server. The server formats this response into a user-friendly format and sends it to the user's device. For example, the server might generate a message such as, "Your phone is frequently restarting, which could be due to a system error, battery issues, or app compatibility problems. We apologize for the inconvenience," and send it to the user's device.

[0476] The user confirms the response.

[0477] The user's device receives a response from the server, which is then displayed to the user. The user can review the displayed response and take necessary actions. For example, if the user asks, "Why is my phone battery draining so quickly?", the server might respond with information such as, "Rapid battery drain is often caused by background applications, display brightness settings, or battery degradation."

[0478] In this way, a dedicated customer support system utilizing generative AI and an emotion engine allows users to receive prompt and accurate support 24 hours a day. Furthermore, this system enables the provision of responses tailored to the user's emotions, resulting in a more satisfying customer support experience.

[0479] The following describes the processing flow.

[0480] Step 1: Access the user's system

[0481] The user opens a dedicated app or web browser using a mobile device or computer. The user enters a specific question or request in the input field. For example, "After a recent update, my computer keeps restarting frequently, and I'm having trouble with it. Please tell me the cause."

[0482] Step 2: Send the user's request to the server.

[0483] The information entered by the user is sent to the server via the internet as an HTTP request. During this process, the user device sends the request data to the server in the appropriate format.

[0484] Step 3: The server receives the request.

[0485] The server receives HTTP requests from users and parses their contents. Specifically, it extracts the user's questions and requests from the request body. For example, it might receive a request that says, "After a recent update, my computer keeps restarting frequently, and I'm having trouble with it. Could you tell me the cause?"

[0486] Step 4: Cleaning the request data

[0487] The server cleans the user's question content that it receives. This process removes unnecessary spaces and special characters. For example, it removes extra spaces from the request so that the text data can be parsed accurately.

[0488] Step 5: Tokenize the request data

[0489] The server tokenizes the cleaned text. For example, it breaks down the text "After the recent update, my computer keeps restarting frequently, and I'm having trouble with it. Please tell me the cause" into words like "recently," "after the update," "frequently," "restart," "having trouble," "cause," and "please tell me."

[0490] Step 6: Sentiment analysis of request data

[0491] The emotion engine installed on the server analyzes emotions from the user's text. For example, it recognizes emotions such as "confusion" or "anxiety" from the expression "I'm troubled."

[0492] Step 7: Send the tokenized and sentiment-analyzed data to the AI ​​model.

[0493] The server sends tokenized data and sentiment information to the artificial intelligence model. The request data is converted into a format that is easy for the AI ​​model to analyze.

[0494] Step 8: Generating responses using the AI ​​model

[0495] The artificial intelligence model analyzes the transmitted data and generates appropriate responses to the user's questions. In doing so, it adjusts the response content considering the user's emotional state. For example, it might generate a response such as, "Frequent restarts of your phone could be due to a system error, battery issues, or app compatibility problems. We apologize for the inconvenience."

[0496] Step 9: Formatting the response

[0497] The server receives the response from the AI ​​model and formats it into a user-friendly format. It formats the response so that it reads like natural language. For example, it might be formatted to read, "Frequent restarts of your phone could be due to a system error, battery issues, or app compatibility problems. We apologize for the inconvenience."

[0498] Step 10: Send the formatted response to the user device.

[0499] The server sends a formatted response to the user's device as an HTTP response. This response provides information in a format that is easy for the user to understand.

[0500] Step 11: User confirms response

[0501] The user device displays the received response, which the user then acknowledges. For example, a message might appear stating, "Frequent restarts on your phone could be caused by a system error, battery issues, or app compatibility problems. We apologize for the inconvenience."

[0502] Thus, a customer-only support system utilizing generative AI and an emotion engine can provide quick and accurate responses to user questions and requests, while also enabling support that takes user emotions into consideration.

[0503] (Example 2)

[0504] 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".

[0505] Traditional customer service systems require quick and accurate responses to user inquiries, but often fail to adequately consider the user's feelings or the specific nature of their problem. This can lead to decreased user satisfaction and longer problem-solving times. Furthermore, uniform responses that disregard emotions tend to be unsatisfactory for users.

[0506] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0507] In this invention, the server includes means for a user to input a question or request via an input device, means for transmitting the input question or request to the server via the Internet, means for cleaning and tokenizing the received request and extracting necessary parameters for preprocessing, means for extracting sentiment information from the received request using a sentiment analysis engine, means for transmitting the preprocessed data and sentiment information to an artificial intelligence model and generating a corresponding response, means for converting the generated response into a user-friendly format and transmitting it to the user device, and means for confirming the response displayed on the user device. This enables accurate and highly satisfying responses to user inquiries that take sentiment into account.

[0508] A "user" refers to anyone who uses the system and enters a question or request.

[0509] "Input device" refers to a terminal used by a user to enter a question or request, and includes mobile devices and personal computers.

[0510] The "Internet" refers to a communication system that connects computer networks around the world, and is the infrastructure for data communication.

[0511] A "server" refers to a computer system that receives, analyzes, and processes requests from users to provide responses.

[0512] A "request" refers to a question or request that a user sends to a server using an input device.

[0513] "Cleaning" refers to the process of removing unnecessary spaces and special characters from received text data.

[0514] "Tokenization" refers to a natural language processing technique that divides text into words or phrases.

[0515] A "sentiment analysis engine" refers to computer software used to extract emotional information from text data.

[0516] An "artificial intelligence model" refers to an algorithm and a trained model used to generate appropriate responses to user questions and requests.

[0517] A "response" refers to a text message that a server generates in response to a user's request and ultimately sends to the user's device.

[0518] "User device" refers to a terminal used by a user to confirm a response, and includes mobile devices or personal computers.

[0519] "Preprocessing" refers to a series of data formatting tasks performed in preparation for data analysis and response generation.

[0520] A "prompt" refers to a text statement used to input specific instructions or questions to an artificial intelligence model.

[0521] This invention relates to a customer service system utilizing generative AI and an emotion analysis engine. The system aims to provide users with prompt and accurate responses to questions and requests entered via mobile devices or browsers. The system combines a server, user devices, the internet, an artificial intelligence model, and an emotion analysis engine.

[0522] Users open a dedicated app or web browser on their mobile device or computer and enter their question or request. For example, a user might type, "My phone keeps restarting frequently after a recent update; could you tell me why?" In this process, users do not need to perform any special operations and can enter their questions in natural language.

[0523] The question entered by the user is sent to the server as an HTTP request via the internet. The server receives the HTTP request and parses the request data. Specifically, it extracts the user's question and request from the request body. The request is sent to the server in JSON format, and from the data in the request body, it extracts "Question: My phone keeps restarting frequently after a recent update. Could you tell me the reason?"

[0524] Next, the server cleans the received question content, removing unnecessary spaces and special characters. Specifically, it uses Python for the cleaning process. Then, it uses a natural language processing library (e.g., NLTK or SpaCy) to tokenize the question content, splitting the sentence into individual words. Furthermore, it uses the server's built-in sentiment analysis engine (e.g., Microsoft Azure Cognitive Services) to analyze the sentiment from the user's text. If the user's question is emotionally charged, such as "After the recent update, it keeps restarting frequently and it's really bothering me. Please help," the sentiment engine will recognize emotions such as "confused" or "difficult."

[0525] Preprocessed data and recognized sentiment information are sent from the server to the artificial intelligence model. The AI ​​model generates a response considering the question and sentiment state. For example, using OpenAI's GPT-3, it might generate a response such as, "Frequent restarts of your phone could be due to a system error, battery issues, or app compatibility. We apologize for the inconvenience."

[0526] The generated response is sent back to the server, which then formats it into a user-friendly format. Specifically, it is converted to HTML format and sent to the user's device. For example, the server might generate a message such as, "Your phone is frequently restarting, which could be due to a system error, battery issues, or app compatibility problems. We apologize for the inconvenience," and send it to the user's device.

[0527] Finally, the user's device receives a response from the server, which is then displayed to the user. The user can review the displayed response and take any necessary actions. For example, if the user asks, "Why is my phone battery draining so quickly?", the server might respond with, "Rapid battery drain is often caused by background applications, display brightness settings, or battery degradation."

[0528] In this way, a dedicated customer support system utilizing generative AI and emotion analysis engines allows users to receive prompt and accurate support 24 hours a day. Furthermore, by providing responses tailored to the user's emotions, a more satisfying customer support experience is achieved.

[0529] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0530] Step 1:

[0531] User access to the system

[0532] Users use a mobile device or computer to open a dedicated app or web browser and enter a question or request. The input is in natural language and includes specific details, such as "My phone keeps restarting frequently after a recent update; could you tell me why?" This input data is saved on the device via the dedicated app or web browser.

[0533] Input: User's question or request

[0534] Output: Questions or requests stored on the input device

[0535] Step 2:

[0536] The server receives the request.

[0537] The user's input question is sent to the server as an HTTP request via the internet. The server receives this HTTP request and extracts the question content from the request body in a data format such as JSON. To prepare the parameters necessary for analysis, the request data is parsed, and the user's question and request are extracted as text data.

[0538] Input: HTTP request sent over the internet

[0539] Output: Extracted text data

[0540] Step 3:

[0541] Request data preprocessing and sentiment analysis

[0542] The server cleans the received question content, removing unnecessary spaces and special characters. Next, it uses a natural language processing library (e.g., NLTK or SpaCy) to tokenize the question content, splitting it into words. Then, it uses a sentiment analysis engine (e.g., Microsoft Azure Cognitive Services) deployed on the server to extract sentiment information from the user's text. For example, it might obtain sentiment information such as "confused" or "difficult."

[0543] Input: Extracted text data

[0544] Output: Cleaned text data and sentiment information

[0545] Step 4:

[0546] Data transmission and response generation

[0547] Pre-processed data and sentiment information are sent from the server to the artificial intelligence model. The AI ​​model uses this data to generate a response, taking into account the content of the question and the emotional state. For example, it might generate a specific response such as, "Frequent restarts of your phone could be due to a system error, battery issues, or app compatibility. We apologize for the inconvenience."

[0548] Input: Cleaned text data and sentiment information

[0549] Output: Generated response text

[0550] Step 5:

[0551] Formatting the response and sending it to the user.

[0552] The generated response is returned to the server, where it is formatted into a user-friendly format. Specifically, the response is converted to HTML or JSON format and the display format is adjusted. The formatted response is then sent to the user's device.

[0553] Input: Generated response text

[0554] Output: Formatted and formatted response message

[0555] Step 6:

[0556] The user confirms the response.

[0557] The user's device displays the response received from the server and makes it visible to the user. The user can review the displayed response and take any necessary actions. For example, if the user asks, "Why is my phone battery draining so quickly?", the provided response might say, "Rapid battery drain is often caused by background applications, display brightness settings, or battery degradation."

[0558] Input: Formatted response message

[0559] Output: Displayed response message

[0560] Through each of the above steps, users can receive prompt and accurate support, and are provided with highly satisfying responses that are tailored to their needs and feelings.

[0561] (Application Example 2)

[0562] 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."

[0563] Traditional customer support systems lacked the ability to respond in a way that considered the user's emotional state and to recommend appropriate content, limiting their potential for improving user satisfaction. Furthermore, the wide variety of information displayed on user devices necessitated a unified interface for responses.

[0564] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for analyzing the received request, extracting necessary parameters and performing preprocessing; means for transmitting the preprocessed data and data extracted from the user's emotional state through sentiment analysis to an artificial intelligence model and generating a corresponding response; and means for recommending content based on the user's past data and current emotional state. This makes it possible to provide detailed responses based on the user's emotional state and suggest content that is suitable for the user's needs.

[0565] "Means by which users input questions or requests via input devices" refers to a function that allows users to input questions or requests using natural language via input devices such as smartphones, tablets, or personal computers.

[0566] "Means for sending entered questions or requests to a server via the Internet" refers to a function that sends user questions or requests from an input device to a server via the Internet in the form of an HTTP request or similar.

[0567] "Means for parsing received requests, extracting necessary parameters, and performing preprocessing" refers to a function that allows a server to analyze questions and requests received from a user, remove unnecessary data, and extract necessary parameters.

[0568] "Means for sending pre-processed data and data extracted from the user's emotional state through sentiment analysis to an artificial intelligence model and generating a corresponding response" refers to a function for sending pre-processed data and data from the user's emotional state analyzed by a sentiment analysis engine to an artificial intelligence model and generating a response that takes the user's emotions into consideration.

[0569] "Means for converting the generated response into a user-friendly format and sending it to the user's device" refers to a function that formats the response generated by the artificial intelligence model into a form that is easy for the user to understand and displays it on the user's device.

[0570] "Means for confirming responses displayed on the user device" refers to a function that allows the user to confirm the response sent from the server on their own device.

[0571] "Means of recommending content based on a user's past data and current emotional state" refers to a function that selects and recommends the most suitable content to a user by considering the user's historical data and emotional state.

[0572] This invention is a system in which a user inputs a question or request via an input device, and appropriate responses and content recommendations are provided based on that content. The details of the system that implements this application example are described below.

[0573] Hardware and software to be used

[0574] The system utilizes servers, user devices such as smartphones and head-mounted displays, Python, TensorFlow, an emotion analysis engine, and a content management system (CMS).

[0575] System Configuration

[0576] Users input questions and requests using a smartphone or head-mounted display via a dedicated app or web browser. User input is in natural language and requires no special operation.

[0577] The server analyzes requests received via the internet, removes unnecessary data, and extracts the necessary parameters. Next, it uses an emotion analysis engine to analyze the user's emotional state. For example, if the user enters "I want to relax today," the emotion analysis engine recognizes the emotion as "relaxed."

[0578] Pre-processed data and emotional states are sent to an artificial intelligence model. This model considers the user's emotional state to generate optimal responses and content recommendations. The generated responses and recommended content are returned to the server and formatted into a user-friendly format. Finally, this is sent to the user's device, where the user reviews the displayed responses and recommendations.

[0579] Specific example

[0580] For example, if a user inputs "I'm feeling down today, so I'd like to watch a video that will cheer me up," the server analyzes this input using its sentiment analysis engine and recognizes the emotional state as "feeling down." The generative AI model takes this emotional state into account and extracts and recommends uplifting videos from its content management system. The server sends a list of recommended content to the user, who can then choose a video they want to watch.

[0581] Example of a prompt

[0582] User input: "I'm feeling down today, so I'd like to watch a video that will cheer me up."

[0583] Generated AI input prompt: "The entered emotion has been analyzed as 'feeling down.' Please recommend a video that will cheer you up."

[0584] In this way, the system can provide optimal responses and content recommendations based on the user's emotional state and past data.

[0585] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0586] Step 1:

[0587] The user enters a question or request via an input device. The user uses a smartphone or head-mounted display to enter the question or request in natural language. This input data is transmitted to a server via the internet by the application on the input device.

[0588] Input: User's question or request in natural language.

[0589] Output: HTTP request sent to the server over the internet

[0590] Step 2:

[0591] The server parses the received request, preprocessing it by removing unnecessary spaces and special characters. It then tokenizes the received data, converting it into structured data.

[0592] Input: Received HTTP request

[0593] Output: Preprocessed structured data

[0594] Specific operation: The server parses the request body, removes unnecessary spaces and special characters using natural language processing, and tokenizes it.

[0595] Step 3:

[0596] The server uses an emotion analysis engine to analyze the user's emotional state from pre-processed data. The analysis results provide the type and intensity of the emotion.

[0597] Input: Preprocessed structured data

[0598] Output: Emotion analysis results (type and degree of emotion)

[0599] Specific operation: The server starts the sentiment analysis engine and analyzes the user's text data to extract their emotional state.

[0600] Step 4:

[0601] Preprocessed data and sentiment analysis results are sent to an artificial intelligence model to generate responses and content recommendations that take the user's emotional state into account. The generated AI model then produces appropriate responses based on the prompts.

[0602] Input: Preprocessed structured data, sentiment analysis results

[0603] Output: Generated responses and content recommendations

[0604] Specific operation: The server inputs data into the generated AI model, generates prompt sentences, and obtains the optimal response and content recommendation.

[0605] Step 5:

[0606] The generated responses and content recommendations are formatted and converted into a user-friendly format. The server then sends this formatted data to the user's device.

[0607] Input: Generated responses and content recommendations

[0608] Output: Formatted responses and content recommendations

[0609] Specific operation: The server converts the generated data into HTML or JSON format and sends it to the user's device.

[0610] Step 6:

[0611] The user's device receives responses and content recommendations sent from the server and displays them to the user. The user reviews them and takes the necessary action (such as watching a video).

[0612] Input: Formatted response and content recommendations sent from the server

[0613] Output: Responses and content recommendations displayed on the user device.

[0614] Specific action: Display the data received by the user's device and allow the user to review it.

[0615] The above describes the processing flow of the system program that implements the application example.

[0616] 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.

[0617] 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.

[0618] 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.

[0619] [Third Embodiment]

[0620] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0621] 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.

[0622] 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).

[0623] 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.

[0624] 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.

[0625] 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).

[0626] 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.

[0627] 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.

[0628] 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.

[0629] 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.

[0630] 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.

[0631] 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".

[0632] This invention relates to a customer service system utilizing generative AI, which allows users to input questions and requests from mobile devices or browsers and receive quick and accurate responses. This system combines a server, user devices, the internet, and an artificial intelligence model.

[0633] User access to the system

[0634] The process begins with the user opening a dedicated app or web browser on their mobile device or computer. The user then enters a question, such as, "What causes my phone to restart frequently?" At this point, the user doesn't need to perform any special operations; they can enter the question using natural language.

[0635] The server receives the request.

[0636] The question entered by the user is sent to the server via the internet. The server receives the HTTP request, parses the request data, and extracts the necessary parameters. For example, if the server receives the question "What causes my phone to restart frequently?", it will parse this question and extract keywords such as "phone," "frequently," "restart," and "cause."

[0637] Preprocessing of request data

[0638] The server cleans the user's question, removing unnecessary spaces and special characters. It also tokenizes the question and performs preprocessing for analysis and response generation. The tokenized data is then converted into a format suitable for artificial intelligence models.

[0639] Inquiries about AI models

[0640] The pre-processed data is sent from the server to the artificial intelligence model. The AI ​​model analyzes the question, understands the user's intent, and generates an appropriate response. For example, the AI ​​model might generate a response such as, "Frequent restarts of your phone could be caused by system errors, battery issues, or app compatibility problems."

[0641] Formatting the response and sending it to the user.

[0642] The generated response is returned to the server. The server formats this response into a user-friendly format and sends it to the user's device. For example, the server might generate a message such as, "Your phone is frequently restarting, which could be caused by a system error, battery issues, or app compatibility problems," and send it to the user's device.

[0643] The user confirms the response.

[0644] The user's device receives a response from the server, which is then displayed to the user. The user can review the displayed response and take any necessary actions. For example, if the user asks, "Why is my phone battery draining so quickly?", the server might respond with information such as, "Rapid battery drain is often caused by background applications, display brightness settings, or natural battery degradation."

[0645] In this way, a customer-only support system utilizing generative AI allows users to receive prompt and accurate support 24 hours a day. Furthermore, this system is expected to improve the efficiency of customer support and reduce costs.

[0646] The following describes the processing flow.

[0647] Step 1: Access the user's system

[0648] Users use their mobile devices or computers to open a dedicated app or web browser and enter a question or request. For example, a user might type, "What causes my phone to restart frequently?"

[0649] Step 2: Send the user's request to the server.

[0650] The information entered by the user is sent to the server via the internet as an HTTP request. During this process, the user device sends the request data to the server in the appropriate format.

[0651] Step 3: The server receives the request.

[0652] The server receives HTTP requests from users and parses their contents. Specifically, it extracts the user's questions and requests from the request body.

[0653] Step 4: Cleaning the request data

[0654] The server cleans the user's question content that it receives. This process removes unnecessary spaces and special characters.

[0655] Step 5: Tokenize the request data

[0656] The server tokenizes the cleaned text. For example, it breaks down the text "What causes my phone to restart frequently?" into words like "phone," "frequently," "restart," and "cause."

[0657] Step 6: Send the tokenized data to the AI ​​model.

[0658] The server sends tokenized data to the artificial intelligence model. The request data is converted into a format that is easy for the AI ​​model to analyze.

[0659] Step 7: Generating responses using the AI ​​model

[0660] The artificial intelligence model analyzes the submitted data and generates appropriate responses to the user's questions. For example, it might generate a response such as, "Frequent restarts of your phone could be due to a system error, battery issues, or app compatibility problems."

[0661] Step 8: The server formats the response.

[0662] The server receives the response from the AI ​​model and formats it into a user-friendly format. The response is formatted to appear as natural-sounding text.

[0663] Step 9: Send the formatted response to the user device.

[0664] The server sends a formatted response to the user's device as an HTTP response. This response provides information in a format that is easy for the user to understand.

[0665] Step 10: User confirms response

[0666] The user device displays the received response, which the user then reviews. For example, it might say, "Your phone is frequently restarting. Possible reasons for this include system errors, battery issues, or app compatibility problems."

[0667] Through the steps described above, a customer-only support system utilizing generative AI can provide quick and accurate responses to user questions and requests.

[0668] (Example 1)

[0669] 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."

[0670] Traditional customer support systems suffered from long response times between users entering questions or requests and receiving a reply, hindering efficient support. Furthermore, response quality was often insufficient, frequently failing to provide users with the information they needed quickly and accurately. Additionally, the heavy burden on customer support staff made 24 / 7 support difficult.

[0671] 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.

[0672] In this invention, the server includes means for a user to input a question or request via an input device, means for transmitting the input question or request to the server via a network, means for analyzing the received request, extracting necessary parameters and performing preprocessing, means for transmitting the preprocessed data to a generating artificial intelligence model and generating a corresponding response, means for converting the generated response into a user-readable format and transmitting it to a user terminal, and means for confirming the response displayed on the user terminal. This enables users to receive prompt and accurate support 24 hours a day. Furthermore, it improves the efficiency of customer support and reduces the burden on staff.

[0673] A "user" refers to a person who uses the system to enter questions or requests.

[0674] An "input device" refers to a device or application used by a user to input questions or requests. This also includes keyboards, touchscreens, and mice.

[0675] "Network" refers to communication lines and infrastructure, including the internet, that provide the means for sending and receiving data.

[0676] A "server" refers to a computer system that has the function of receiving requests from users, analyzing and pre-processing them, generating responses, and sending those responses.

[0677] A "request" refers to a question or request that a user sends to a system via an input device.

[0678] "Analysis" refers to the breakdown and interpretation of data that a server performs in order to understand a received request.

[0679] A "parameter" refers to a specific element or data point extracted from a request.

[0680] "Preprocessing" refers to the process by which the server cleans, tokenizes, and prepares request data for analysis and response generation.

[0681] A "generative artificial intelligence model" refers to a program that uses natural language processing and machine learning to generate appropriate responses based on input data.

[0682] "Response" refers to the information that a server or generative artificial intelligence model generates in response to a request and returns to the user.

[0683] A "user terminal" refers to a device used by a user to access a system and confirm its response. Examples include personal computers, smartphones, and tablets.

[0684] This invention relates to a customer service system utilizing generative AI. This system allows users to input questions and requests via mobile devices or browsers, and receive quick and accurate responses. The system components include a server, user terminals, a network, and a generative artificial intelligence model.

[0685] Specifically, users first open a dedicated app or web browser on their mobile device or computer and enter their question in natural language. For example, a user might enter a question like, "What causes my phone to restart frequently?" At this point, users don't need to perform any special operations; they can enter their question using natural language.

[0686] The server receives user inquiries via the internet. The received inquiries are in the form of HTTP requests, and the server parses these requests to extract the necessary parameters. For example, keywords such as "mobile," "frequently," "restart," and "cause" are obtained as part of the analysis results.

[0687] Next, the server cleanses the received question content, removing unnecessary spaces and special characters. It also tokenizes the question and performs preprocessing for analysis and response generation, converting the tokenized data into a format suitable for the generated artificial intelligence model.

[0688] Once preprocessing is complete, the server sends the tokenized data to the AI ​​model. The AI ​​model analyzes the question, understands the user's intent, and generates an appropriate response. For example, it might generate a response such as, "Frequent restarts of your phone could be caused by system errors, battery issues, or app compatibility problems."

[0689] The generated response is returned to the server, which then formats it into a user-friendly format. The formatted response is then sent to the user's device as a message such as, "Your phone is frequently restarting due to a system error, battery problem, or app compatibility issue."

[0690] Finally, the user's device receives a response from the server and displays it to the user. The user can then review the displayed response and take the necessary actions. For example, if the user asks, "Why is my phone battery draining so quickly?", the server might respond with information such as, "Rapid battery drain is often caused by background applications, display brightness settings, or natural battery degradation."

[0691] As an example of a prompt, when querying a generative AI model with the question, "Why is my phone battery draining so quickly?", you would use a prompt like the following:

[0692] "Generate an answer to the following question: 'Why does my phone battery drain so quickly?'"

[0693] In this way, a customer-only support system utilizing generative AI allows users to receive prompt and accurate support 24 hours a day. Furthermore, this system is expected to improve the efficiency of customer support and reduce costs.

[0694] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0695] Step 1:

[0696] The user opens a dedicated app or web browser using a mobile device or computer. The user enters a question in natural language, such as "What causes my phone to restart frequently?". The user does not need to perform any special operations; they can simply enter the question in natural language. The input is captured on the device in text format. Specifically, the user enters a URL in the browser's address bar, or taps the app icon to launch the app, enters the question in the chat box, and clicks the "Send" button.

[0697] Step 2:

[0698] The question entered by the user is sent to the server via the internet as an HTTP request. The input is text data from the user. The server parses the received request and deciphers the content of the question. Specifically, from the question "What causes my phone to restart frequently?", it extracts keywords such as "phone," "frequently," "restart," and "cause." In terms of specific operations, the server reads the body of the HTTP request and extracts the necessary parameters.

[0699] Step 3:

[0700] The server cleans the received question content, removing unnecessary spaces and special characters. The input is text data containing the keywords extracted in step 2. The server tokenizes the question and preprocesses it for parsing and response generation. For example, the question "What causes my phone to restart frequently?" is split into words such as "phone," "frequently," "restart," and "cause," and extra spaces and special characters are removed. Specifically, the server uses a text processing library to perform cleaning and tokenization.

[0701] Step 4:

[0702] Preprocessed data is sent from the server to the AI ​​model. The input is tokenized and cleaned text data. The AI ​​model analyzes the data, understands the user's intent, and generates an appropriate response. For example, the AI ​​model might generate a response such as, "Frequent restarts of your phone could be caused by system errors, battery issues, or app compatibility issues." In practice, the server sends the data to the AI ​​model's API and retrieves the response.

[0703] Step 5:

[0704] The generated response is returned to the server, which then formats it into a user-friendly format. The input is the response data returned from the AI ​​model. The server creates a message, for example, "Your phone is frequently restarting, which could be caused by a system error, battery issues, or app compatibility problems," and sends it to the user's device. Specifically, the server applies format conversion and display templates.

[0705] Step 6:

[0706] The user's device receives a response from the server and displays it to the user. The input is formatted response data sent from the server. The user can review the displayed response and take necessary actions. Specifically, the user's device displays the received data on the screen and renders the response content using the browser or application's display engine.

[0707] This series of processes allows users to receive prompt and accurate support.

[0708] (Application Example 1)

[0709] 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."

[0710] Existing food delivery services struggle to respond quickly and accurately to customer inquiries and support requests. This results in users spending a long time resolving problems, leading to a poor customer experience. Furthermore, significant labor costs are incurred for customer support, making it inefficient. Therefore, there is a need for a system that can respond quickly and accurately to customer inquiries in food delivery services.

[0711] 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.

[0712] In this invention, the server includes means for a user to input a question or request via an input device, means for transmitting the input question or request to the server via the internet, means for analyzing the received request, extracting necessary parameters and performing preprocessing, means for transmitting the preprocessed data to an artificial intelligence model and generating a corresponding response, means for converting the generated response into a user-readable format and transmitting it to the user device, means for confirming the response displayed on the user device, means for performing tokenization and cleaning for response generation, and means for providing customer support for a food delivery service using natural language processing means. This enables quick and accurate responses to customer inquiries, resulting in improved customer experience and reduced customer support costs.

[0713] "A means by which a user enters a question or request via an input device" refers to an interface that allows a user to enter a question or request in text format using a device such as a smartphone or computer.

[0714] "Means of sending entered questions or requests to a server over the internet" refers to protocols or functions that send questions or requests from a user's device to a server over the internet.

[0715] "Means for parsing received requests, extracting necessary parameters, and performing preprocessing" refers to the process by which a server parses questions and requests received from users and extracts keywords and context.

[0716] "Means for sending pre-processed data to an artificial intelligence model and generating a corresponding response" refers to the process of inputting pre-processed data into an artificial intelligence model to generate an appropriate response to a user's question or request.

[0717] "Means for converting the generated response into a user-friendly format and sending it to the user's device" refers to the process of converting the response generated by the artificial intelligence model into a style and format that is easy for the user to understand, and then sending it to the user's device.

[0718] "Means for confirming the response displayed on the user device" refers to an interface for confirming the generated response displayed on the user's device and determining the next action.

[0719] "Means for tokenizing and cleaning for response generation" refers to a preprocessing step that tokenizes the user's question or request and removes unnecessary characters and symbols.

[0720] "Means for providing customer support for food delivery services using natural language processing" refers to a function that uses natural language processing technology to generate appropriate responses to customer inquiries and problems related to food delivery services.

[0721] This invention relates to a system for automating customer support in food delivery services. The system aims to provide quick and accurate responses to questions and requests entered by users via input devices such as smartphones and personal computers.

[0722] System configuration and operation

[0723] 1. The user accesses the system:

[0724] Users can use a dedicated application to enter questions and requests regarding food delivery in natural language. For example, they can ask questions such as, "My order hasn't arrived."

[0725] 2. The server receives the request:

[0726] The entered questions and requests are sent to the server via the internet. The server receives the HTTP request and parses its contents.

[0727] 3. Preprocessing of request data:

[0728] The server cleans the received request data, removing unnecessary spaces and special characters. Next, it tokenizes the request data, converting it into a format suitable for the artificial intelligence model. This preprocessing step uses natural language processing (NLP) techniques.

[0729] 4. Inquiries to the AI ​​model:

[0730] The pre-processed data is sent from the server to a generating AI model (e.g., the OpenAI GPT series). The AI ​​model analyzes the received data and generates appropriate responses to user questions and requests. For example, if a user asks, "My order hasn't arrived," the AI ​​model will generate a response such as, "Please check your order confirmation email. The email may contain information about the delivery status."

[0731] 5. Formatting the response and sending it to the user:

[0732] The generated response is returned to the server, which then formats it into a user-friendly format. The formatted message is then sent to the user's device, allowing the user to quickly find the appropriate solution and next steps.

[0733] 6. The user confirms the response:

[0734] The user's device receives a response from the server and displays it to the user. The user can then review the displayed response and take any necessary actions.

[0735] Hardware and software to be used

[0736] Server: Use an Amazon Web Services (AWS) EC2 instance.

[0737] Generative AI model: OpenAI GPT-3 / 4 is used.

[0738] Natural Language Processing (NLP): SpaCy will be used.

[0739] Communication protocol: Use HTTPS.

[0740] Database: PostgreSQL will be used.

[0741] Examples of specific cases and prompt statements

[0742] For example, if a user enters the question, "I'm having trouble paying my credit card," the AI ​​model will be queried in the following format:

[0743] Prompt example:

[0744] A user asked, "I'm having trouble with my credit card payment." What is the appropriate solution?

[0745] By using such prompts, the AI ​​model can generate appropriate responses to the user's situation, enabling a rapid response.

[0746] The system configuration and operation described above enable the provision of quick and accurate responses to customer inquiries in food delivery services. This improves the efficiency of customer support and enhances user satisfaction.

[0747] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0748] Step 1:

[0749] Users enter questions or requests using their smartphones or computers. They type their question (e.g., "There's a problem with my credit card payment.") into the text field on their input device and press the submit button. This input text becomes the request data sent to the server.

[0750] Step 2:

[0751] The entered question or request is sent to the server over the internet. The terminal creates an HTTP request and sends a payload containing the input text to a specific endpoint on the server. This request data is received by the server.

[0752] Step 3:

[0753] The server parses the received request, extracts the necessary parameters, and performs preprocessing. Specifically, the server analyzes the input text and extracts keywords such as "credit card," "payment," and "problem." It also removes unnecessary spaces and special characters. As a result, clean, tokenized text data is obtained.

[0754] Step 4:

[0755] The server sends pre-processed data to an artificial intelligence model (generative AI model) to generate a corresponding response. Specifically, it passes pre-processed text data to the generative AI model as a prompt, asking in the format, "The user asked, 'There is a problem with my credit card payment.' What is the appropriate solution?" The AI ​​model analyzes the received data and generates an appropriate response (e.g., "Contact your credit card company or double-check your payment information.").

[0756] Step 5:

[0757] The generated response is returned to the server, which then formats it into a user-friendly format. Specifically, the server converts the response text into HTML or JSON format and generates a message to send to the user's device. This formatted response data is then sent to the user's device.

[0758] Step 6:

[0759] The user device receives a response from the server and displays it to the user. Specifically, the response text (e.g., "Please contact your credit card company or re-verify your payment information.") is displayed on the smartphone or computer screen. The user can then review the displayed response and take appropriate action.

[0760] Through the above processing steps, the food delivery service will be able to provide quick and accurate responses to customer inquiries.

[0761] 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.

[0762] This invention relates to a customer-only support system utilizing generative AI and an emotion engine, which allows users to input questions and requests from mobile devices or browsers and receive quick and accurate responses. This system combines a server, user devices, the internet, an artificial intelligence model, and an emotion engine.

[0763] User access to the system

[0764] Users open a dedicated app or web browser using their mobile device or computer and enter their question or request. For example, a user might type, "My phone keeps restarting frequently after a recent update; could you tell me why?" In this case, the user doesn't need to perform any special operations and can enter their question in natural language.

[0765] The server receives the request.

[0766] The questions entered by the user are sent to the server as HTTP requests via the internet. The server receives these HTTP requests and parses the request data. Specifically, it extracts the user's questions and requests from the request body.

[0767] Request data preprocessing and sentiment analysis

[0768] The server cleans the received user question, removing unnecessary spaces and special characters. Next, it tokenizes the question and performs preprocessing for analysis and response generation. Furthermore, the server's built-in sentiment engine analyzes the sentiment from the user's text. For example, if the user's question is emotionally charged, such as "After the recent update, it keeps restarting frequently and it's really bothering me. Please help," the sentiment engine will recognize emotions such as "confusion" and "difficulty."

[0769] Sending data, including emotions, to AI models

[0770] Pre-processed data and recognized sentiment information are sent from the server to the artificial intelligence model. The AI ​​model generates a response that considers not only the question but also the user's emotional state. For example, it might generate a response such as, "Frequent restarts of your phone could be due to a system error, battery issues, or app compatibility. We apologize for the inconvenience."

[0771] Formatting the response and sending it to the user.

[0772] The generated response is returned to the server. The server formats this response into a user-friendly format and sends it to the user's device. For example, the server might generate a message such as, "Your phone is frequently restarting, which could be due to a system error, battery issues, or app compatibility problems. We apologize for the inconvenience," and send it to the user's device.

[0773] The user confirms the response.

[0774] The user's device receives a response from the server, which is then displayed to the user. The user can review the displayed response and take necessary actions. For example, if the user asks, "Why is my phone battery draining so quickly?", the server might respond with information such as, "Rapid battery drain is often caused by background applications, display brightness settings, or battery degradation."

[0775] In this way, a dedicated customer support system utilizing generative AI and an emotion engine allows users to receive prompt and accurate support 24 hours a day. Furthermore, this system enables the provision of responses tailored to the user's emotions, resulting in a more satisfying customer support experience.

[0776] The following describes the processing flow.

[0777] Step 1: Access the user's system

[0778] The user opens a dedicated app or web browser using a mobile device or computer. The user enters a specific question or request in the input field. For example, "After a recent update, my computer keeps restarting frequently, and I'm having trouble with it. Please tell me the cause."

[0779] Step 2: Send the user's request to the server.

[0780] The information entered by the user is sent to the server via the internet as an HTTP request. During this process, the user device sends the request data to the server in the appropriate format.

[0781] Step 3: The server receives the request.

[0782] The server receives HTTP requests from users and parses their contents. Specifically, it extracts the user's questions and requests from the request body. For example, it might receive a request that says, "After a recent update, my computer keeps restarting frequently, and I'm having trouble with it. Could you tell me the cause?"

[0783] Step 4: Cleaning the request data

[0784] The server cleans the user's question content that it receives. This process removes unnecessary spaces and special characters. For example, it removes extra spaces from the request so that the text data can be parsed accurately.

[0785] Step 5: Tokenize the request data

[0786] The server tokenizes the cleaned text. For example, it breaks down the text "After the recent update, my computer keeps restarting frequently, and I'm having trouble with it. Please tell me the cause" into words like "recently," "after the update," "frequently," "restart," "having trouble," "cause," and "please tell me."

[0787] Step 6: Sentiment analysis of request data

[0788] The emotion engine installed on the server analyzes emotions from the user's text. For example, it recognizes emotions such as "confusion" or "anxiety" from the expression "I'm troubled."

[0789] Step 7: Send the tokenized and sentiment-analyzed data to the AI ​​model.

[0790] The server sends tokenized data and sentiment information to the artificial intelligence model. The request data is converted into a format that is easy for the AI ​​model to analyze.

[0791] Step 8: Generating responses using the AI ​​model

[0792] The artificial intelligence model analyzes the transmitted data and generates appropriate responses to the user's questions. In doing so, it adjusts the response content considering the user's emotional state. For example, it might generate a response such as, "Frequent restarts of your phone could be due to a system error, battery issues, or app compatibility problems. We apologize for the inconvenience."

[0793] Step 9: Formatting the response

[0794] The server receives the response from the AI ​​model and formats it into a user-friendly format. It formats the response so that it reads like natural language. For example, it might be formatted to read, "Frequent restarts of your phone could be due to a system error, battery issues, or app compatibility problems. We apologize for the inconvenience."

[0795] Step 10: Send the formatted response to the user device.

[0796] The server sends a formatted response to the user's device as an HTTP response. This response provides information in a format that is easy for the user to understand.

[0797] Step 11: User confirms response

[0798] The user device displays the received response, which the user then acknowledges. For example, a message might appear stating, "Frequent restarts on your phone could be caused by a system error, battery issues, or app compatibility problems. We apologize for the inconvenience."

[0799] Thus, a customer-only support system utilizing generative AI and an emotion engine can provide quick and accurate responses to user questions and requests, while also enabling support that takes user emotions into consideration.

[0800] (Example 2)

[0801] 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."

[0802] Traditional customer service systems require quick and accurate responses to user inquiries, but often fail to adequately consider the user's feelings or the specific nature of their problem. This can lead to decreased user satisfaction and longer problem-solving times. Furthermore, uniform responses that disregard emotions tend to be unsatisfactory for users.

[0803] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0804] In this invention, the server includes means for a user to input a question or request via an input device, means for transmitting the input question or request to the server via the Internet, means for cleaning and tokenizing the received request and extracting necessary parameters for preprocessing, means for extracting sentiment information from the received request using a sentiment analysis engine, means for transmitting the preprocessed data and sentiment information to an artificial intelligence model and generating a corresponding response, means for converting the generated response into a user-friendly format and transmitting it to the user device, and means for confirming the response displayed on the user device. This enables accurate and highly satisfying responses to user inquiries that take sentiment into account.

[0805] A "user" refers to anyone who uses the system and enters a question or request.

[0806] "Input device" refers to a terminal used by a user to enter a question or request, and includes mobile devices and personal computers.

[0807] The "Internet" refers to a communication system that connects computer networks around the world, and is the infrastructure for data communication.

[0808] A "server" refers to a computer system that receives, analyzes, and processes requests from users to provide responses.

[0809] A "request" refers to a question or request that a user sends to a server using an input device.

[0810] "Cleaning" refers to the process of removing unnecessary spaces and special characters from received text data.

[0811] "Tokenization" refers to a natural language processing technique that divides text into words or phrases.

[0812] A "sentiment analysis engine" refers to computer software used to extract emotional information from text data.

[0813] An "artificial intelligence model" refers to an algorithm and a trained model used to generate appropriate responses to user questions and requests.

[0814] A "response" refers to a text message that a server generates in response to a user's request and ultimately sends to the user's device.

[0815] "User device" refers to a terminal used by a user to confirm a response, and includes mobile devices or personal computers.

[0816] "Preprocessing" refers to a series of data formatting tasks performed in preparation for data analysis and response generation.

[0817] A "prompt" refers to a text statement used to input specific instructions or questions to an artificial intelligence model.

[0818] This invention relates to a customer service system utilizing generative AI and an emotion analysis engine. The system aims to provide users with prompt and accurate responses to questions and requests entered via mobile devices or browsers. The system combines a server, user devices, the internet, an artificial intelligence model, and an emotion analysis engine.

[0819] Users open a dedicated app or web browser on their mobile device or computer and enter their question or request. For example, a user might type, "My phone keeps restarting frequently after a recent update; could you tell me why?" In this process, users do not need to perform any special operations and can enter their questions in natural language.

[0820] The question entered by the user is sent to the server as an HTTP request via the internet. The server receives the HTTP request and parses the request data. Specifically, it extracts the user's question and request from the request body. The request is sent to the server in JSON format, and from the data in the request body, it extracts "Question: My phone keeps restarting frequently after a recent update. Could you tell me the reason?"

[0821] Next, the server cleans the received question content, removing unnecessary spaces and special characters. Specifically, it uses Python for the cleaning process. Then, it uses a natural language processing library (e.g., NLTK or SpaCy) to tokenize the question content, splitting the sentence into individual words. Furthermore, it uses the server's built-in sentiment analysis engine (e.g., Microsoft Azure Cognitive Services) to analyze the sentiment from the user's text. If the user's question is emotionally charged, such as "After the recent update, it keeps restarting frequently and it's really bothering me. Please help," the sentiment engine will recognize emotions such as "confused" or "difficult."

[0822] Preprocessed data and recognized sentiment information are sent from the server to the artificial intelligence model. The AI ​​model generates a response considering the question and sentiment state. For example, using OpenAI's GPT-3, it might generate a response such as, "Frequent restarts of your phone could be due to a system error, battery issues, or app compatibility. We apologize for the inconvenience."

[0823] The generated response is sent back to the server, which then formats it into a user-friendly format. Specifically, it is converted to HTML format and sent to the user's device. For example, the server might generate a message such as, "Your phone is frequently restarting, which could be due to a system error, battery issues, or app compatibility problems. We apologize for the inconvenience," and send it to the user's device.

[0824] Finally, the user's device receives a response from the server, which is then displayed to the user. The user can review the displayed response and take any necessary actions. For example, if the user asks, "Why is my phone battery draining so quickly?", the server might respond with, "Rapid battery drain is often caused by background applications, display brightness settings, or battery degradation."

[0825] In this way, a dedicated customer support system utilizing generative AI and emotion analysis engines allows users to receive prompt and accurate support 24 hours a day. Furthermore, by providing responses tailored to the user's emotions, a more satisfying customer support experience is achieved.

[0826] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0827] Step 1:

[0828] User access to the system

[0829] Users use a mobile device or computer to open a dedicated app or web browser and enter a question or request. The input is in natural language and includes specific details, such as "My phone keeps restarting frequently after a recent update; could you tell me why?" This input data is saved on the device via the dedicated app or web browser.

[0830] Input: User's question or request

[0831] Output: Questions or requests stored on the input device

[0832] Step 2:

[0833] The server receives the request.

[0834] The user's input question is sent to the server as an HTTP request via the internet. The server receives this HTTP request and extracts the question content from the request body in a data format such as JSON. To prepare the parameters necessary for analysis, the request data is parsed, and the user's question and request are extracted as text data.

[0835] Input: HTTP request sent over the internet

[0836] Output: Extracted text data

[0837] Step 3:

[0838] Request data preprocessing and sentiment analysis

[0839] The server cleans the received question content, removing unnecessary spaces and special characters. Next, it uses a natural language processing library (e.g., NLTK or SpaCy) to tokenize the question content, splitting it into words. Then, it uses a sentiment analysis engine (e.g., Microsoft Azure Cognitive Services) deployed on the server to extract sentiment information from the user's text. For example, it might obtain sentiment information such as "confused" or "difficult."

[0840] Input: Extracted text data

[0841] Output: Cleaned text data and sentiment information

[0842] Step 4:

[0843] Data transmission and response generation

[0844] Pre-processed data and sentiment information are sent from the server to the artificial intelligence model. The AI ​​model uses this data to generate a response, taking into account the content of the question and the emotional state. For example, it might generate a specific response such as, "Frequent restarts of your phone could be due to a system error, battery issues, or app compatibility. We apologize for the inconvenience."

[0845] Input: Cleaned text data and sentiment information

[0846] Output: Generated response text

[0847] Step 5:

[0848] Formatting the response and sending it to the user.

[0849] The generated response is returned to the server, where it is formatted into a user-friendly format. Specifically, the response is converted to HTML or JSON format and the display format is adjusted. The formatted response is then sent to the user's device.

[0850] Input: Generated response text

[0851] Output: Formatted and formatted response message

[0852] Step 6:

[0853] The user confirms the response.

[0854] The user's device displays the response received from the server and makes it visible to the user. The user can review the displayed response and take any necessary actions. For example, if the user asks, "Why is my phone battery draining so quickly?", the provided response might say, "Rapid battery drain is often caused by background applications, display brightness settings, or battery degradation."

[0855] Input: Formatted response message

[0856] Output: Displayed response message

[0857] Through each of the above steps, users can receive prompt and accurate support, and are provided with highly satisfying responses that are tailored to their needs and feelings.

[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 headset-type terminal 314 will be referred to as the "terminal."

[0860] Traditional customer support systems lacked the ability to respond in a way that considered the user's emotional state and to recommend appropriate content, limiting their potential for improving user satisfaction. Furthermore, the wide variety of information displayed on user devices necessitated a unified interface for responses.

[0861] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for analyzing the received request, extracting necessary parameters and performing preprocessing; means for transmitting the preprocessed data and data extracted from the user's emotional state through sentiment analysis to an artificial intelligence model and generating a corresponding response; and means for recommending content based on the user's past data and current emotional state. This makes it possible to provide detailed responses based on the user's emotional state and suggest content that is suitable for the user's needs.

[0862] "Means by which users input questions or requests via input devices" refers to a function that allows users to input questions or requests using natural language via input devices such as smartphones, tablets, or personal computers.

[0863] "Means for sending entered questions or requests to a server via the Internet" refers to a function that sends user questions or requests from an input device to a server via the Internet in the form of an HTTP request or similar.

[0864] "Means for parsing received requests, extracting necessary parameters, and performing preprocessing" refers to a function that allows a server to analyze questions and requests received from a user, remove unnecessary data, and extract necessary parameters.

[0865] "Means for sending pre-processed data and data extracted from the user's emotional state through sentiment analysis to an artificial intelligence model and generating a corresponding response" refers to a function for sending pre-processed data and data from the user's emotional state analyzed by a sentiment analysis engine to an artificial intelligence model and generating a response that takes the user's emotions into consideration.

[0866] "Means for converting the generated response into a user-friendly format and sending it to the user's device" refers to a function that formats the response generated by the artificial intelligence model into a form that is easy for the user to understand and displays it on the user's device.

[0867] "Means for confirming responses displayed on the user device" refers to a function that allows the user to confirm the response sent from the server on their own device.

[0868] "Means of recommending content based on a user's past data and current emotional state" refers to a function that selects and recommends the most suitable content to a user by considering the user's historical data and emotional state.

[0869] This invention is a system in which a user inputs a question or request via an input device, and appropriate responses and content recommendations are provided based on that content. The details of the system that implements this application example are described below.

[0870] Hardware and software to be used

[0871] The system utilizes servers, user devices such as smartphones and head-mounted displays, Python, TensorFlow, an emotion analysis engine, and a content management system (CMS).

[0872] System Configuration

[0873] Users input questions and requests using a smartphone or head-mounted display via a dedicated app or web browser. User input is in natural language and requires no special operation.

[0874] The server analyzes requests received via the internet, removes unnecessary data, and extracts the necessary parameters. Next, it uses an emotion analysis engine to analyze the user's emotional state. For example, if the user enters "I want to relax today," the emotion analysis engine recognizes the emotion as "relaxed."

[0875] Pre-processed data and emotional states are sent to an artificial intelligence model. This model considers the user's emotional state to generate optimal responses and content recommendations. The generated responses and recommended content are returned to the server and formatted into a user-friendly format. Finally, this is sent to the user's device, where the user reviews the displayed responses and recommendations.

[0876] Specific example

[0877] For example, if a user inputs "I'm feeling down today, so I'd like to watch a video that will cheer me up," the server analyzes this input using its sentiment analysis engine and recognizes the emotional state as "feeling down." The generative AI model takes this emotional state into account and extracts and recommends uplifting videos from its content management system. The server sends a list of recommended content to the user, who can then choose a video they want to watch.

[0878] Example of a prompt

[0879] User input: "I'm feeling down today, so I'd like to watch a video that will cheer me up."

[0880] Generated AI input prompt: "The entered emotion has been analyzed as 'feeling down.' Please recommend a video that will cheer you up."

[0881] In this way, the system can provide optimal responses and content recommendations based on the user's emotional state and past data.

[0882] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0883] Step 1:

[0884] The user enters a question or request via an input device. The user uses a smartphone or head-mounted display to enter the question or request in natural language. This input data is transmitted to a server via the internet by the application on the input device.

[0885] Input: User's question or request in natural language.

[0886] Output: HTTP request sent to the server over the internet

[0887] Step 2:

[0888] The server parses the received request, preprocessing it by removing unnecessary spaces and special characters. It then tokenizes the received data, converting it into structured data.

[0889] Input: Received HTTP request

[0890] Output: Preprocessed structured data

[0891] Specific operation: The server parses the request body, removes unnecessary spaces and special characters using natural language processing, and tokenizes it.

[0892] Step 3:

[0893] The server uses an emotion analysis engine to analyze the user's emotional state from pre-processed data. The analysis results provide the type and intensity of the emotion.

[0894] Input: Preprocessed structured data

[0895] Output: Emotion analysis results (type and degree of emotion)

[0896] Specific operation: The server starts the sentiment analysis engine and analyzes the user's text data to extract their emotional state.

[0897] Step 4:

[0898] Preprocessed data and sentiment analysis results are sent to an artificial intelligence model to generate responses and content recommendations that take the user's emotional state into account. The generated AI model then produces appropriate responses based on the prompts.

[0899] Input: Preprocessed structured data, sentiment analysis results

[0900] Output: Generated responses and content recommendations

[0901] Specific operation: The server inputs data into the generated AI model, generates prompt sentences, and obtains the optimal response and content recommendation.

[0902] Step 5:

[0903] The generated responses and content recommendations are formatted and converted into a user-friendly format. The server then sends this formatted data to the user's device.

[0904] Input: Generated responses and content recommendations

[0905] Output: Formatted responses and content recommendations

[0906] Specific operation: The server converts the generated data into HTML or JSON format and sends it to the user's device.

[0907] Step 6:

[0908] The user's device receives responses and content recommendations sent from the server and displays them to the user. The user reviews them and takes the necessary action (such as watching a video).

[0909] Input: Formatted response and content recommendations sent from the server

[0910] Output: Responses and content recommendations displayed on the user device.

[0911] Specific action: Display the data received by the user's device and allow the user to review it.

[0912] The above describes the processing flow of the system program that implements the application example.

[0913] 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.

[0914] 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.

[0915] 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.

[0916] [Fourth Embodiment]

[0917] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0918] 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.

[0919] 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).

[0920] 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.

[0921] 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.

[0922] 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).

[0923] 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.

[0924] 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.

[0925] 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.

[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 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.

[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 robot 414 as the "terminal".

[0930] This invention relates to a customer service system utilizing generative AI, which allows users to input questions and requests from mobile devices or browsers and receive quick and accurate responses. This system combines a server, user devices, the internet, and an artificial intelligence model.

[0931] User access to the system

[0932] The process begins with the user opening a dedicated app or web browser on their mobile device or computer. The user then enters a question, such as, "What causes my phone to restart frequently?" At this point, the user doesn't need to perform any special operations; they can enter the question using natural language.

[0933] The server receives the request.

[0934] The question entered by the user is sent to the server via the internet. The server receives the HTTP request, parses the request data, and extracts the necessary parameters. For example, if the server receives the question "What causes my phone to restart frequently?", it will parse this question and extract keywords such as "phone," "frequently," "restart," and "cause."

[0935] Preprocessing of request data

[0936] The server cleans the user's question, removing unnecessary spaces and special characters. It also tokenizes the question and performs preprocessing for analysis and response generation. The tokenized data is then converted into a format suitable for artificial intelligence models.

[0937] Inquiries about AI models

[0938] The pre-processed data is sent from the server to the artificial intelligence model. The AI ​​model analyzes the question, understands the user's intent, and generates an appropriate response. For example, the AI ​​model might generate a response such as, "Frequent restarts of your phone could be caused by system errors, battery issues, or app compatibility problems."

[0939] Formatting the response and sending it to the user.

[0940] The generated response is returned to the server. The server formats this response into a user-friendly format and sends it to the user's device. For example, the server might generate a message such as, "Your phone is frequently restarting, which could be caused by a system error, battery issues, or app compatibility problems," and send it to the user's device.

[0941] The user confirms the response.

[0942] The user's device receives a response from the server, which is then displayed to the user. The user can review the displayed response and take any necessary actions. For example, if the user asks, "Why is my phone battery draining so quickly?", the server might respond with information such as, "Rapid battery drain is often caused by background applications, display brightness settings, or natural battery degradation."

[0943] In this way, a customer-only support system utilizing generative AI allows users to receive prompt and accurate support 24 hours a day. Furthermore, this system is expected to improve the efficiency of customer support and reduce costs.

[0944] The following describes the processing flow.

[0945] Step 1: Access the user's system

[0946] Users use their mobile devices or computers to open a dedicated app or web browser and enter a question or request. For example, a user might type, "What causes my phone to restart frequently?"

[0947] Step 2: Send the user's request to the server.

[0948] The information entered by the user is sent to the server via the internet as an HTTP request. During this process, the user device sends the request data to the server in the appropriate format.

[0949] Step 3: The server receives the request.

[0950] The server receives HTTP requests from users and parses their contents. Specifically, it extracts the user's questions and requests from the request body.

[0951] Step 4: Cleaning the request data

[0952] The server cleans the user's question content that it receives. This process removes unnecessary spaces and special characters.

[0953] Step 5: Tokenize the request data

[0954] The server tokenizes the cleaned text. For example, it breaks down the text "What causes my phone to restart frequently?" into words like "phone," "frequently," "restart," and "cause."

[0955] Step 6: Send the tokenized data to the AI ​​model.

[0956] The server sends tokenized data to the artificial intelligence model. The request data is converted into a format that is easy for the AI ​​model to analyze.

[0957] Step 7: Generating responses using the AI ​​model

[0958] The artificial intelligence model analyzes the submitted data and generates appropriate responses to the user's questions. For example, it might generate a response such as, "Frequent restarts of your phone could be due to a system error, battery issues, or app compatibility problems."

[0959] Step 8: The server formats the response.

[0960] The server receives the response from the AI ​​model and formats it into a user-friendly format. The response is formatted to appear as natural-sounding text.

[0961] Step 9: Send the formatted response to the user device.

[0962] The server sends a formatted response to the user's device as an HTTP response. This response provides information in a format that is easy for the user to understand.

[0963] Step 10: User confirms response

[0964] The user device displays the received response, which the user then reviews. For example, it might say, "Your phone is frequently restarting. Possible reasons for this include system errors, battery issues, or app compatibility problems."

[0965] Through the steps described above, a customer-only support system utilizing generative AI can provide quick and accurate responses to user questions and requests.

[0966] (Example 1)

[0967] 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".

[0968] Traditional customer support systems suffered from long response times between users entering questions or requests and receiving a reply, hindering efficient support. Furthermore, response quality was often insufficient, frequently failing to provide users with the information they needed quickly and accurately. Additionally, the heavy burden on customer support staff made 24 / 7 support difficult.

[0969] 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.

[0970] In this invention, the server includes means for a user to input a question or request via an input device, means for transmitting the input question or request to the server via a network, means for analyzing the received request, extracting necessary parameters and performing preprocessing, means for transmitting the preprocessed data to a generating artificial intelligence model and generating a corresponding response, means for converting the generated response into a user-readable format and transmitting it to a user terminal, and means for confirming the response displayed on the user terminal. This enables users to receive prompt and accurate support 24 hours a day. Furthermore, it improves the efficiency of customer support and reduces the burden on staff.

[0971] A "user" refers to a person who uses the system to enter questions or requests.

[0972] An "input device" refers to a device or application used by a user to input questions or requests. This also includes keyboards, touchscreens, and mice.

[0973] "Network" refers to communication lines and infrastructure, including the internet, that provide the means for sending and receiving data.

[0974] A "server" refers to a computer system that has the function of receiving requests from users, analyzing and pre-processing them, generating responses, and sending those responses.

[0975] A "request" refers to a question or request that a user sends to a system via an input device.

[0976] "Analysis" refers to the breakdown and interpretation of data that a server performs in order to understand a received request.

[0977] A "parameter" refers to a specific element or data point extracted from a request.

[0978] "Preprocessing" refers to the process by which the server cleans, tokenizes, and prepares request data for analysis and response generation.

[0979] A "generative artificial intelligence model" refers to a program that uses natural language processing and machine learning to generate appropriate responses based on input data.

[0980] "Response" refers to the information that a server or generative artificial intelligence model generates in response to a request and returns to the user.

[0981] A "user terminal" refers to a device used by a user to access a system and confirm its response. Examples include personal computers, smartphones, and tablets.

[0982] This invention relates to a customer service system utilizing generative AI. This system allows users to input questions and requests via mobile devices or browsers, and receive quick and accurate responses. The system components include a server, user terminals, a network, and a generative artificial intelligence model.

[0983] Specifically, users first open a dedicated app or web browser on their mobile device or computer and enter their question in natural language. For example, a user might enter a question like, "What causes my phone to restart frequently?" At this point, users don't need to perform any special operations; they can enter their question using natural language.

[0984] The server receives user inquiries via the internet. The received inquiries are in the form of HTTP requests, and the server parses these requests to extract the necessary parameters. For example, keywords such as "mobile," "frequently," "restart," and "cause" are obtained as part of the analysis results.

[0985] Next, the server cleanses the received question content, removing unnecessary spaces and special characters. It also tokenizes the question and performs preprocessing for analysis and response generation, converting the tokenized data into a format suitable for the generated artificial intelligence model.

[0986] Once preprocessing is complete, the server sends the tokenized data to the AI ​​model. The AI ​​model analyzes the question, understands the user's intent, and generates an appropriate response. For example, it might generate a response such as, "Frequent restarts of your phone could be caused by system errors, battery issues, or app compatibility problems."

[0987] The generated response is returned to the server, which then formats it into a user-friendly format. The formatted response is then sent to the user's device as a message such as, "Your phone is frequently restarting due to a system error, battery problem, or app compatibility issue."

[0988] Finally, the user's device receives a response from the server and displays it to the user. The user can then review the displayed response and take the necessary actions. For example, if the user asks, "Why is my phone battery draining so quickly?", the server might respond with information such as, "Rapid battery drain is often caused by background applications, display brightness settings, or natural battery degradation."

[0989] As an example of a prompt, when querying a generative AI model with the question, "Why is my phone battery draining so quickly?", you would use a prompt like the following:

[0990] "Generate an answer to the following question: 'Why does my phone battery drain so quickly?'"

[0991] In this way, a customer-only support system utilizing generative AI allows users to receive prompt and accurate support 24 hours a day. Furthermore, this system is expected to improve the efficiency of customer support and reduce costs.

[0992] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0993] Step 1:

[0994] The user opens a dedicated app or web browser using a mobile device or computer. The user enters a question in natural language, such as "What causes my phone to restart frequently?". The user does not need to perform any special operations; they can simply enter the question in natural language. The input is captured on the device in text format. Specifically, the user enters a URL in the browser's address bar, or taps the app icon to launch the app, enters the question in the chat box, and clicks the "Send" button.

[0995] Step 2:

[0996] The question entered by the user is sent to the server via the internet as an HTTP request. The input is text data from the user. The server parses the received request and deciphers the content of the question. Specifically, from the question "What causes my phone to restart frequently?", it extracts keywords such as "phone," "frequently," "restart," and "cause." In terms of specific operations, the server reads the body of the HTTP request and extracts the necessary parameters.

[0997] Step 3:

[0998] The server cleans the received question content, removing unnecessary spaces and special characters. The input is text data containing the keywords extracted in step 2. The server tokenizes the question and preprocesses it for parsing and response generation. For example, the question "What causes my phone to restart frequently?" is split into words such as "phone," "frequently," "restart," and "cause," and extra spaces and special characters are removed. Specifically, the server uses a text processing library to perform cleaning and tokenization.

[0999] Step 4:

[1000] Preprocessed data is sent from the server to the AI ​​model. The input is tokenized and cleaned text data. The AI ​​model analyzes the data, understands the user's intent, and generates an appropriate response. For example, the AI ​​model might generate a response such as, "Frequent restarts of your phone could be caused by system errors, battery issues, or app compatibility issues." In practice, the server sends the data to the AI ​​model's API and retrieves the response.

[1001] Step 5:

[1002] The generated response is returned to the server, which then formats it into a user-friendly format. The input is the response data returned from the AI ​​model. The server creates a message, for example, "Your phone is frequently restarting, which could be caused by a system error, battery issues, or app compatibility problems," and sends it to the user's device. Specifically, the server applies format conversion and display templates.

[1003] Step 6:

[1004] The user's device receives a response from the server and displays it to the user. The input is formatted response data sent from the server. The user can review the displayed response and take necessary actions. Specifically, the user's device displays the received data on the screen and renders the response content using the browser or application's display engine.

[1005] This series of processes allows users to receive prompt and accurate support.

[1006] (Application Example 1)

[1007] 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".

[1008] Existing food delivery services struggle to respond quickly and accurately to customer inquiries and support requests. This results in users spending a long time resolving problems, leading to a poor customer experience. Furthermore, significant labor costs are incurred for customer support, making it inefficient. Therefore, there is a need for a system that can respond quickly and accurately to customer inquiries in food delivery services.

[1009] 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.

[1010] In this invention, the server includes means for a user to input a question or request via an input device, means for transmitting the input question or request to the server via the internet, means for analyzing the received request, extracting necessary parameters and performing preprocessing, means for transmitting the preprocessed data to an artificial intelligence model and generating a corresponding response, means for converting the generated response into a user-readable format and transmitting it to the user device, means for confirming the response displayed on the user device, means for performing tokenization and cleaning for response generation, and means for providing customer support for a food delivery service using natural language processing means. This enables quick and accurate responses to customer inquiries, resulting in improved customer experience and reduced customer support costs.

[1011] "A means by which a user enters a question or request via an input device" refers to an interface that allows a user to enter a question or request in text format using a device such as a smartphone or computer.

[1012] "Means of sending entered questions or requests to a server over the internet" refers to protocols or functions that send questions or requests from a user's device to a server over the internet.

[1013] "Means for parsing received requests, extracting necessary parameters, and performing preprocessing" refers to the process by which a server parses questions and requests received from users and extracts keywords and context.

[1014] "Means for sending pre-processed data to an artificial intelligence model and generating a corresponding response" refers to the process of inputting pre-processed data into an artificial intelligence model to generate an appropriate response to a user's question or request.

[1015] "Means for converting the generated response into a user-friendly format and sending it to the user's device" refers to the process of converting the response generated by the artificial intelligence model into a style and format that is easy for the user to understand, and then sending it to the user's device.

[1016] "Means for confirming the response displayed on the user device" refers to an interface for confirming the generated response displayed on the user's device and determining the next action.

[1017] "Means for tokenizing and cleaning for response generation" refers to a preprocessing step that tokenizes the user's question or request and removes unnecessary characters and symbols.

[1018] "Means for providing customer support for food delivery services using natural language processing" refers to a function that uses natural language processing technology to generate appropriate responses to customer inquiries and problems related to food delivery services.

[1019] This invention relates to a system for automating customer support in food delivery services. The system aims to provide quick and accurate responses to questions and requests entered by users via input devices such as smartphones and personal computers.

[1020] System configuration and operation

[1021] 1. The user accesses the system:

[1022] Users can use a dedicated application to enter questions and requests regarding food delivery in natural language. For example, they can ask questions such as, "My order hasn't arrived."

[1023] 2. The server receives the request:

[1024] The entered questions and requests are sent to the server via the internet. The server receives the HTTP request and parses its contents.

[1025] 3. Preprocessing of request data:

[1026] The server cleans the received request data, removing unnecessary spaces and special characters. Next, it tokenizes the request data, converting it into a format suitable for the artificial intelligence model. This preprocessing step uses natural language processing (NLP) techniques.

[1027] 4. Inquiries to the AI ​​model:

[1028] The pre-processed data is sent from the server to a generating AI model (e.g., the OpenAI GPT series). The AI ​​model analyzes the received data and generates appropriate responses to user questions and requests. For example, if a user asks, "My order hasn't arrived," the AI ​​model will generate a response such as, "Please check your order confirmation email. The email may contain information about the delivery status."

[1029] 5. Formatting the response and sending it to the user:

[1030] The generated response is returned to the server, which then formats it into a user-friendly format. The formatted message is then sent to the user's device, allowing the user to quickly find the appropriate solution and next steps.

[1031] 6. The user confirms the response:

[1032] The user's device receives a response from the server and displays it to the user. The user can then review the displayed response and take any necessary actions.

[1033] Hardware and software to be used

[1034] Server: Use an Amazon Web Services (AWS) EC2 instance.

[1035] Generative AI model: OpenAI GPT-3 / 4 is used.

[1036] Natural Language Processing (NLP): SpaCy will be used.

[1037] Communication protocol: Use HTTPS.

[1038] Database: PostgreSQL will be used.

[1039] Examples of specific cases and prompt statements

[1040] For example, if a user enters the question, "I'm having trouble paying my credit card," the AI ​​model will be queried in the following format:

[1041] Prompt example:

[1042] A user asked, "I'm having trouble with my credit card payment." What is the appropriate solution?

[1043] By using such prompts, the AI ​​model can generate appropriate responses to the user's situation, enabling a rapid response.

[1044] The system configuration and operation described above enable the provision of quick and accurate responses to customer inquiries in food delivery services. This improves the efficiency of customer support and enhances user satisfaction.

[1045] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1046] Step 1:

[1047] Users enter questions or requests using their smartphones or computers. They type their question (e.g., "There's a problem with my credit card payment.") into the text field on their input device and press the submit button. This input text becomes the request data sent to the server.

[1048] Step 2:

[1049] The entered question or request is sent to the server over the internet. The terminal creates an HTTP request and sends a payload containing the input text to a specific endpoint on the server. This request data is received by the server.

[1050] Step 3:

[1051] The server parses the received request, extracts the necessary parameters, and performs preprocessing. Specifically, the server analyzes the input text and extracts keywords such as "credit card," "payment," and "problem." It also removes unnecessary spaces and special characters. As a result, clean, tokenized text data is obtained.

[1052] Step 4:

[1053] The server sends pre-processed data to an artificial intelligence model (generative AI model) to generate a corresponding response. Specifically, it passes pre-processed text data to the generative AI model as a prompt, asking in the format, "The user asked, 'There is a problem with my credit card payment.' What is the appropriate solution?" The AI ​​model analyzes the received data and generates an appropriate response (e.g., "Contact your credit card company or double-check your payment information.").

[1054] Step 5:

[1055] The generated response is returned to the server, which then formats it into a user-friendly format. Specifically, the server converts the response text into HTML or JSON format and generates a message to send to the user's device. This formatted response data is then sent to the user's device.

[1056] Step 6:

[1057] The user device receives a response from the server and displays it to the user. Specifically, the response text (e.g., "Please contact your credit card company or re-verify your payment information.") is displayed on the smartphone or computer screen. The user can then review the displayed response and take appropriate action.

[1058] Through the above processing steps, the food delivery service will be able to provide quick and accurate responses to customer inquiries.

[1059] 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.

[1060] This invention relates to a customer-only support system utilizing generative AI and an emotion engine, which allows users to input questions and requests from mobile devices or browsers and receive quick and accurate responses. This system combines a server, user devices, the internet, an artificial intelligence model, and an emotion engine.

[1061] User access to the system

[1062] Users open a dedicated app or web browser using their mobile device or computer and enter their question or request. For example, a user might type, "My phone keeps restarting frequently after a recent update; could you tell me why?" In this case, the user doesn't need to perform any special operations and can enter their question in natural language.

[1063] The server receives the request.

[1064] The questions entered by the user are sent to the server as HTTP requests via the internet. The server receives these HTTP requests and parses the request data. Specifically, it extracts the user's questions and requests from the request body.

[1065] Request data preprocessing and sentiment analysis

[1066] The server cleans the received user question, removing unnecessary spaces and special characters. Next, it tokenizes the question and performs preprocessing for analysis and response generation. Furthermore, the server's built-in sentiment engine analyzes the sentiment from the user's text. For example, if the user's question is emotionally charged, such as "After the recent update, it keeps restarting frequently and it's really bothering me. Please help," the sentiment engine will recognize emotions such as "confusion" and "difficulty."

[1067] Sending data, including emotions, to AI models

[1068] Pre-processed data and recognized sentiment information are sent from the server to the artificial intelligence model. The AI ​​model generates a response that considers not only the question but also the user's emotional state. For example, it might generate a response such as, "Frequent restarts of your phone could be due to a system error, battery issues, or app compatibility. We apologize for the inconvenience."

[1069] Formatting the response and sending it to the user.

[1070] The generated response is returned to the server. The server formats this response into a user-friendly format and sends it to the user's device. For example, the server might generate a message such as, "Your phone is frequently restarting, which could be due to a system error, battery issues, or app compatibility problems. We apologize for the inconvenience," and send it to the user's device.

[1071] The user confirms the response.

[1072] The user's device receives a response from the server, which is then displayed to the user. The user can review the displayed response and take necessary actions. For example, if the user asks, "Why is my phone battery draining so quickly?", the server might respond with information such as, "Rapid battery drain is often caused by background applications, display brightness settings, or battery degradation."

[1073] In this way, a dedicated customer support system utilizing generative AI and an emotion engine allows users to receive prompt and accurate support 24 hours a day. Furthermore, this system enables the provision of responses tailored to the user's emotions, resulting in a more satisfying customer support experience.

[1074] The following describes the processing flow.

[1075] Step 1: Access the user's system

[1076] The user opens a dedicated app or web browser using a mobile device or computer. The user enters a specific question or request in the input field. For example, "After a recent update, my computer keeps restarting frequently, and I'm having trouble with it. Please tell me the cause."

[1077] Step 2: Send the user's request to the server.

[1078] The information entered by the user is sent to the server via the internet as an HTTP request. During this process, the user device sends the request data to the server in the appropriate format.

[1079] Step 3: The server receives the request.

[1080] The server receives HTTP requests from users and parses their contents. Specifically, it extracts the user's questions and requests from the request body. For example, it might receive a request that says, "After a recent update, my computer keeps restarting frequently, and I'm having trouble with it. Could you tell me the cause?"

[1081] Step 4: Cleaning the request data

[1082] The server cleans the user's question content that it receives. This process removes unnecessary spaces and special characters. For example, it removes extra spaces from the request so that the text data can be parsed accurately.

[1083] Step 5: Tokenize the request data

[1084] The server tokenizes the cleaned text. For example, it breaks down the text "After the recent update, my computer keeps restarting frequently, and I'm having trouble with it. Please tell me the cause" into words like "recently," "after the update," "frequently," "restart," "having trouble," "cause," and "please tell me."

[1085] Step 6: Sentiment analysis of request data

[1086] The emotion engine installed on the server analyzes emotions from the user's text. For example, it recognizes emotions such as "confusion" or "anxiety" from the expression "I'm troubled."

[1087] Step 7: Send the tokenized and sentiment-analyzed data to the AI ​​model.

[1088] The server sends tokenized data and sentiment information to the artificial intelligence model. The request data is converted into a format that is easy for the AI ​​model to analyze.

[1089] Step 8: Generating responses using the AI ​​model

[1090] The artificial intelligence model analyzes the transmitted data and generates appropriate responses to the user's questions. In doing so, it adjusts the response content considering the user's emotional state. For example, it might generate a response such as, "Frequent restarts of your phone could be due to a system error, battery issues, or app compatibility problems. We apologize for the inconvenience."

[1091] Step 9: Formatting the response

[1092] The server receives the response from the AI ​​model and formats it into a user-friendly format. It formats the response so that it reads like natural language. For example, it might be formatted to read, "Frequent restarts of your phone could be due to a system error, battery issues, or app compatibility problems. We apologize for the inconvenience."

[1093] Step 10: Send the formatted response to the user device.

[1094] The server sends a formatted response to the user's device as an HTTP response. This response provides information in a format that is easy for the user to understand.

[1095] Step 11: User confirms response

[1096] The user device displays the received response, which the user then acknowledges. For example, a message might appear stating, "Frequent restarts on your phone could be caused by a system error, battery issues, or app compatibility problems. We apologize for the inconvenience."

[1097] Thus, a customer-only support system utilizing generative AI and an emotion engine can provide quick and accurate responses to user questions and requests, while also enabling support that takes user emotions into consideration.

[1098] (Example 2)

[1099] 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".

[1100] Traditional customer service systems require quick and accurate responses to user inquiries, but often fail to adequately consider the user's feelings or the specific nature of their problem. This can lead to decreased user satisfaction and longer problem-solving times. Furthermore, uniform responses that disregard emotions tend to be unsatisfactory for users.

[1101] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1102] In this invention, the server includes means for a user to input a question or request via an input device, means for transmitting the input question or request to the server via the Internet, means for cleaning and tokenizing the received request and extracting necessary parameters for preprocessing, means for extracting sentiment information from the received request using a sentiment analysis engine, means for transmitting the preprocessed data and sentiment information to an artificial intelligence model and generating a corresponding response, means for converting the generated response into a user-friendly format and transmitting it to the user device, and means for confirming the response displayed on the user device. This enables accurate and highly satisfying responses to user inquiries that take sentiment into account.

[1103] A "user" refers to anyone who uses the system and enters a question or request.

[1104] "Input device" refers to a terminal used by a user to enter a question or request, and includes mobile devices and personal computers.

[1105] The "Internet" refers to a communication system that connects computer networks around the world, and is the infrastructure for data communication.

[1106] A "server" refers to a computer system that receives, analyzes, and processes requests from users to provide responses.

[1107] A "request" refers to a question or request that a user sends to a server using an input device.

[1108] "Cleaning" refers to the process of removing unnecessary spaces and special characters from received text data.

[1109] "Tokenization" refers to a natural language processing technique that divides text into words or phrases.

[1110] A "sentiment analysis engine" refers to computer software used to extract emotional information from text data.

[1111] An "artificial intelligence model" refers to an algorithm and a trained model used to generate appropriate responses to user questions and requests.

[1112] A "response" refers to a text message that a server generates in response to a user's request and ultimately sends to the user's device.

[1113] "User device" refers to a terminal used by a user to confirm a response, and includes mobile devices or personal computers.

[1114] "Preprocessing" refers to a series of data formatting tasks performed in preparation for data analysis and response generation.

[1115] A "prompt" refers to a text statement used to input specific instructions or questions to an artificial intelligence model.

[1116] This invention relates to a customer service system utilizing generative AI and an emotion analysis engine. The system aims to provide users with prompt and accurate responses to questions and requests entered via mobile devices or browsers. The system combines a server, user devices, the internet, an artificial intelligence model, and an emotion analysis engine.

[1117] Users open a dedicated app or web browser on their mobile device or computer and enter their question or request. For example, a user might type, "My phone keeps restarting frequently after a recent update; could you tell me why?" In this process, users do not need to perform any special operations and can enter their questions in natural language.

[1118] The question entered by the user is sent to the server as an HTTP request via the internet. The server receives the HTTP request and parses the request data. Specifically, it extracts the user's question and request from the request body. The request is sent to the server in JSON format, and from the data in the request body, it extracts "Question: My phone keeps restarting frequently after a recent update. Could you tell me the reason?"

[1119] Next, the server cleans the received question content, removing unnecessary spaces and special characters. Specifically, it uses Python for the cleaning process. Then, it uses a natural language processing library (e.g., NLTK or SpaCy) to tokenize the question content, splitting the sentence into individual words. Furthermore, it uses the server's built-in sentiment analysis engine (e.g., Microsoft Azure Cognitive Services) to analyze the sentiment from the user's text. If the user's question is emotionally charged, such as "After the recent update, it keeps restarting frequently and it's really bothering me. Please help," the sentiment engine will recognize emotions such as "confused" or "difficult."

[1120] Preprocessed data and recognized sentiment information are sent from the server to the artificial intelligence model. The AI ​​model generates a response considering the question and sentiment state. For example, using OpenAI's GPT-3, it might generate a response such as, "Frequent restarts of your phone could be due to a system error, battery issues, or app compatibility. We apologize for the inconvenience."

[1121] The generated response is sent back to the server, which then formats it into a user-friendly format. Specifically, it is converted to HTML format and sent to the user's device. For example, the server might generate a message such as, "Your phone is frequently restarting, which could be due to a system error, battery issues, or app compatibility problems. We apologize for the inconvenience," and send it to the user's device.

[1122] Finally, the user's device receives a response from the server, which is then displayed to the user. The user can review the displayed response and take any necessary actions. For example, if the user asks, "Why is my phone battery draining so quickly?", the server might respond with, "Rapid battery drain is often caused by background applications, display brightness settings, or battery degradation."

[1123] In this way, a dedicated customer support system utilizing generative AI and emotion analysis engines allows users to receive prompt and accurate support 24 hours a day. Furthermore, by providing responses tailored to the user's emotions, a more satisfying customer support experience is achieved.

[1124] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1125] Step 1:

[1126] User access to the system

[1127] Users use a mobile device or computer to open a dedicated app or web browser and enter a question or request. The input is in natural language and includes specific details, such as "My phone keeps restarting frequently after a recent update; could you tell me why?" This input data is saved on the device via the dedicated app or web browser.

[1128] Input: User's question or request

[1129] Output: Questions or requests stored on the input device

[1130] Step 2:

[1131] The server receives the request.

[1132] The user's input question is sent to the server as an HTTP request via the internet. The server receives this HTTP request and extracts the question content from the request body in a data format such as JSON. To prepare the parameters necessary for analysis, the request data is parsed, and the user's question and request are extracted as text data.

[1133] Input: HTTP request sent over the internet

[1134] Output: Extracted text data

[1135] Step 3:

[1136] Request data preprocessing and sentiment analysis

[1137] The server cleans the received question content, removing unnecessary spaces and special characters. Next, it uses a natural language processing library (e.g., NLTK or SpaCy) to tokenize the question content, splitting it into words. Then, it uses a sentiment analysis engine (e.g., Microsoft Azure Cognitive Services) deployed on the server to extract sentiment information from the user's text. For example, it might obtain sentiment information such as "confused" or "difficult."

[1138] Input: Extracted text data

[1139] Output: Cleaned text data and sentiment information

[1140] Step 4:

[1141] Data transmission and response generation

[1142] Pre-processed data and sentiment information are sent from the server to the artificial intelligence model. The AI ​​model uses this data to generate a response, taking into account the content of the question and the emotional state. For example, it might generate a specific response such as, "Frequent restarts of your phone could be due to a system error, battery issues, or app compatibility. We apologize for the inconvenience."

[1143] Input: Cleaned text data and sentiment information

[1144] Output: Generated response text

[1145] Step 5:

[1146] Formatting the response and sending it to the user.

[1147] The generated response is returned to the server, where it is formatted into a user-friendly format. Specifically, the response is converted to HTML or JSON format and the display format is adjusted. The formatted response is then sent to the user's device.

[1148] Input: Generated response text

[1149] Output: Formatted and formatted response message

[1150] Step 6:

[1151] The user confirms the response.

[1152] The user's device displays the response received from the server and makes it visible to the user. The user can review the displayed response and take any necessary actions. For example, if the user asks, "Why is my phone battery draining so quickly?", the provided response might say, "Rapid battery drain is often caused by background applications, display brightness settings, or battery degradation."

[1153] Input: Formatted response message

[1154] Output: Displayed response message

[1155] Through each of the above steps, users can receive prompt and accurate support, and are provided with highly satisfying responses that are tailored to their needs and feelings.

[1156] (Application Example 2)

[1157] 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".

[1158] Traditional customer support systems lacked the ability to respond in a way that considered the user's emotional state and to recommend appropriate content, limiting their potential for improving user satisfaction. Furthermore, the wide variety of information displayed on user devices necessitated a unified interface for responses.

[1159] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for analyzing the received request, extracting necessary parameters and performing preprocessing; means for transmitting the preprocessed data and data extracted from the user's emotional state through sentiment analysis to an artificial intelligence model and generating a corresponding response; and means for recommending content based on the user's past data and current emotional state. This makes it possible to provide detailed responses based on the user's emotional state and suggest content that is suitable for the user's needs.

[1160] "Means by which users input questions or requests via input devices" refers to a function that allows users to input questions or requests using natural language via input devices such as smartphones, tablets, or personal computers.

[1161] "Means for sending entered questions or requests to a server via the Internet" refers to a function that sends user questions or requests from an input device to a server via the Internet in the form of an HTTP request or similar.

[1162] "Means for parsing received requests, extracting necessary parameters, and performing preprocessing" refers to a function that allows a server to analyze questions and requests received from a user, remove unnecessary data, and extract necessary parameters.

[1163] "Means for sending pre-processed data and data extracted from the user's emotional state through sentiment analysis to an artificial intelligence model and generating a corresponding response" refers to a function for sending pre-processed data and data from the user's emotional state analyzed by a sentiment analysis engine to an artificial intelligence model and generating a response that takes the user's emotions into consideration.

[1164] "Means for converting the generated response into a user-friendly format and sending it to the user's device" refers to a function that formats the response generated by the artificial intelligence model into a form that is easy for the user to understand and displays it on the user's device.

[1165] "Means for confirming responses displayed on the user device" refers to a function that allows the user to confirm the response sent from the server on their own device.

[1166] "Means of recommending content based on a user's past data and current emotional state" refers to a function that selects and recommends the most suitable content to a user by considering the user's historical data and emotional state.

[1167] This invention is a system in which a user inputs a question or request via an input device, and appropriate responses and content recommendations are provided based on that content. The details of the system that implements this application example are described below.

[1168] Hardware and software to be used

[1169] The system utilizes servers, user devices such as smartphones and head-mounted displays, Python, TensorFlow, an emotion analysis engine, and a content management system (CMS).

[1170] System Configuration

[1171] Users input questions and requests using a smartphone or head-mounted display via a dedicated app or web browser. User input is in natural language and requires no special operation.

[1172] The server analyzes requests received via the internet, removes unnecessary data, and extracts the necessary parameters. Next, it uses an emotion analysis engine to analyze the user's emotional state. For example, if the user enters "I want to relax today," the emotion analysis engine recognizes the emotion as "relaxed."

[1173] Pre-processed data and emotional states are sent to an artificial intelligence model. This model considers the user's emotional state to generate optimal responses and content recommendations. The generated responses and recommended content are returned to the server and formatted into a user-friendly format. Finally, this is sent to the user's device, where the user reviews the displayed responses and recommendations.

[1174] Specific example

[1175] For example, if a user inputs "I'm feeling down today, so I'd like to watch a video that will cheer me up," the server analyzes this input using its sentiment analysis engine and recognizes the emotional state as "feeling down." The generative AI model takes this emotional state into account and extracts and recommends uplifting videos from its content management system. The server sends a list of recommended content to the user, who can then choose a video they want to watch.

[1176] Example of a prompt

[1177] User input: "I'm feeling down today, so I'd like to watch a video that will cheer me up."

[1178] Generated AI input prompt: "The entered emotion has been analyzed as 'feeling down.' Please recommend a video that will cheer you up."

[1179] In this way, the system can provide optimal responses and content recommendations based on the user's emotional state and past data.

[1180] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1181] Step 1:

[1182] The user enters a question or request via an input device. The user uses a smartphone or head-mounted display to enter the question or request in natural language. This input data is transmitted to a server via the internet by the application on the input device.

[1183] Input: User's question or request in natural language.

[1184] Output: HTTP request sent to the server over the internet

[1185] Step 2:

[1186] The server parses the received request, preprocessing it by removing unnecessary spaces and special characters. It then tokenizes the received data, converting it into structured data.

[1187] Input: Received HTTP request

[1188] Output: Preprocessed structured data

[1189] Specific operation: The server parses the request body, removes unnecessary spaces and special characters using natural language processing, and tokenizes it.

[1190] Step 3:

[1191] The server uses an emotion analysis engine to analyze the user's emotional state from pre-processed data. The analysis results provide the type and intensity of the emotion.

[1192] Input: Preprocessed structured data

[1193] Output: Emotion analysis results (type and degree of emotion)

[1194] Specific operation: The server starts the sentiment analysis engine and analyzes the user's text data to extract their emotional state.

[1195] Step 4:

[1196] Preprocessed data and sentiment analysis results are sent to an artificial intelligence model to generate responses and content recommendations that take the user's emotional state into account. The generated AI model then produces appropriate responses based on the prompts.

[1197] Input: Preprocessed structured data, sentiment analysis results

[1198] Output: Generated responses and content recommendations

[1199] Specific operation: The server inputs data into the generated AI model, generates prompt sentences, and obtains the optimal response and content recommendation.

[1200] Step 5:

[1201] The generated responses and content recommendations are formatted and converted into a user-friendly format. The server then sends this formatted data to the user's device.

[1202] Input: Generated responses and content recommendations

[1203] Output: Formatted responses and content recommendations

[1204] Specific operation: The server converts the generated data into HTML or JSON format and sends it to the user's device.

[1205] Step 6:

[1206] The user's device receives responses and content recommendations sent from the server and displays them to the user. The user reviews them and takes the necessary action (such as watching a video).

[1207] Input: Formatted response and content recommendations sent from the server

[1208] Output: Responses and content recommendations displayed on the user device.

[1209] Specific action: Display the data received by the user's device and allow the user to review it.

[1210] The above describes the processing flow of the system program that implements the application example.

[1211] 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.

[1212] 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.

[1213] 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.

[1214] 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.

[1215] 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.

[1216] 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.

[1217] 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.

[1218] 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.

[1219] 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."

[1220] 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.

[1221] 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.

[1222] 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.

[1223] 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.

[1224] 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.

[1225] 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.

[1226] 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.

[1227] 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.

[1228] 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.

[1229] 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.

[1230] 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.

[1231] 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.

[1232] The following is further disclosed regarding the embodiments described above.

[1233] (Claim 1)

[1234] A means by which the user enters a question or request via an input device,

[1235] A means of sending the entered question or request to the server via the internet,

[1236] A means for analyzing the received request, extracting the necessary parameters, and performing preprocessing,

[1237] A means for sending pre-processed data to an artificial intelligence model and generating a corresponding response,

[1238] A means for converting the generated response into a user-friendly format and sending it to the user's device,

[1239] A system including means for confirming a response displayed on a user device.

[1240] (Claim 2)

[1241] The system according to claim 1, wherein the artificial intelligence model generates an appropriate response to a user request, including product specification information, troubleshooting information, reservation procedure information, or service guidance information.

[1242] (Claim 3)

[1243] The system according to claim 1, wherein the server tokenizes the request data using natural language processing means and transmits the tokenized data to an artificial intelligence model.

[1244] "Example 1"

[1245] (Claim 1)

[1246] A means by which the user enters a question or request via an input device,

[1247] Means for sending the entered question or request to the server over the network,

[1248] A means for analyzing the received request, extracting the necessary parameters, and performing preprocessing,

[1249] A means for sending pre-processed data to a generating artificial intelligence model and generating a corresponding response,

[1250] A means for converting the generated response into a user-friendly format and sending it to the user's terminal,

[1251] A system including means for confirming the response displayed on the user terminal.

[1252] (Claim 2)

[1253] The system according to claim 1, wherein the generating artificial intelligence model generates appropriate responses to user requests, including product specification information, problem-solving information, reservation procedure information, or service guidance information, and provides prompt and accurate support on a 24-hour basis.

[1254] (Claim 3)

[1255] The system according to claim 1, wherein the server tokenizes and cleans the request data using natural language processing means, removes unnecessary spaces and special characters to perform preprocessing for analysis and response generation, and transmits the tokenized and cleaned data to a generating artificial intelligence model.

[1256] "Application Example 1"

[1257] (Claim 1)

[1258] A means by which the user enters a question or request via an input device,

[1259] A means of sending the entered question or request to the server via the internet,

[1260] A means for analyzing the received request, extracting the necessary parameters, and performing preprocessing,

[1261] A means for sending pre-processed data to an artificial intelligence model and generating a corresponding response,

[1262] A means for converting the generated response into a user-friendly format and sending it to the user's device,

[1263] A means of confirming the response displayed on the user device,

[1264] Means for tokenizing and cleaning in order to generate a response,

[1265] A system including means for providing customer support for a food delivery service using natural language processing means.

[1266] (Claim 2)

[1267] The system according to claim 1, wherein the artificial intelligence model generates an appropriate response to a user request, including product specification information, troubleshooting information, reservation procedure information, service guidance information, or problem-solving information related to food delivery services.

[1268] (Claim 3)

[1269] The system according to claim 1, wherein the server tokenizes and cleans the request data using natural language processing means, and transmits the tokenized and cleaned data to an artificial intelligence model.

[1270] "Example 2 of combining an emotion engine"

[1271] (Claim 1)

[1272] A means by which the user enters a question or request via an input device,

[1273] A means of sending the entered question or request to the server via the internet,

[1274] A means for cleaning and tokenizing received requests, extracting necessary parameters, and performing preprocessing,

[1275] A means of extracting emotional information from a received request using an emotional analysis engine,

[1276] A means for transmitting pre-processed data and emotional information to an artificial intelligence model and generating a corresponding response,

[1277] A means for converting the generated response into a user-friendly format and sending it to the user's device,

[1278] A system including means for confirming a response displayed on a user device.

[1279] (Claim 2)

[1280] The system according to claim 1, wherein the artificial intelligence model generates an appropriate response to a user request, including product specification information, troubleshooting information, reservation procedure information, or service guidance information.

[1281] (Claim 3)

[1282] The system according to claim 1, wherein the server tokenizes the request data using natural language processing means and transmits the tokenized data to an artificial intelligence model.

[1283] "Application example 2 when combining with an emotional engine"

[1284] (Claim 1)

[1285] A means by which the user enters a question or request via an input device,

[1286] A means of sending the entered question or request to the server via the internet,

[1287] A means for analyzing the received request, extracting the necessary parameters, and performing preprocessing,

[1288] A means for sending pre-processed data and data extracted from the user's emotional state through sentiment analysis to an artificial intelligence model and generating a corresponding response,

[1289] A means for converting the generated response into a user-friendly format and sending it to the user's device,

[1290] A means of confirming the response displayed on the user device,

[1291] A means of recommending content based on the user's past data and current emotional state,

[1292] A system that includes this.

[1293] (Claim 2)

[1294] The system according to claim 1, wherein the artificial intelligence model generates an appropriate response to a user request, including product specification information, troubleshooting information, reservation procedure information, service guidance information, or content recommendation information.

[1295] (Claim 3)

[1296] The system according to claim 1, wherein the server tokenizes the request data using natural language processing means and transmits the tokenized data to an artificial intelligence model. [Explanation of Symbols]

[1297] 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 by which the user enters a question or request via an input device, A means of sending the entered question or request to the server via the internet, A means for analyzing the received request, extracting the necessary parameters, and performing preprocessing, A means for sending pre-processed data to an artificial intelligence model and generating a corresponding response, A means for converting the generated response into a user-friendly format and sending it to the user's device, A system including means for confirming a response displayed on a user device.

2. The system according to claim 1, wherein the artificial intelligence model generates an appropriate response to a user request, including product specification information, troubleshooting information, reservation procedure information, or service guidance information.

3. The system according to claim 1, wherein the server tokenizes the request data using natural language processing means and transmits the tokenized data to an artificial intelligence model.

Images