system

Abstract

This system provides input from a wide range of users, analyzes it on a server, and retrieves appropriate database information. [Solution] A system comprising: means for inputting a request from a user; means for sending the request to a server; means for performing natural language analysis on the request on the server; means for obtaining appropriate information from a database based on the natural language analysis results; means for generating an optimal suggestion for the user using generative artificial intelligence; means for sending the generated suggestion to the user's terminal; means for displaying the suggestion to the user; means for the user to select an action based on the displayed suggestion; means for sending the user's action to the server; means for receiving and analyzing feedback from the user on the server; and means for updating the generative artificial intelligence based on the feedback.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, the method including 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] In current customer support systems, it is difficult to provide quick and accurate proposals for various user requirements. In addition, since support for data transfer between different operating systems and selection of an optimal pricing plan is insufficient, users have to conduct investigations and procedures on their own, which is burdensome. As a result, there is a problem that the user experience deteriorates and it is difficult to improve customer satisfaction.

Means for Solving the Problems

[0005] This invention provides a system that takes a wide range of user requests as input, analyzes them on a server, and retrieves appropriate database information. Specifically, it includes means for analyzing user requests in natural language and generating optimal suggestions using generative artificial intelligence. It also includes means for displaying the suggestions to the user and prompting actions based on the user's selection. Furthermore, it includes means for suggesting the optimal choice of pricing plan and means for supporting the transfer of image data between different operating systems. This makes it possible to reduce the burden on the user and provide prompt and accurate support.

[0006] A "user" refers to an entity that uses a system to input requests or receive suggestions.

[0007] A "request" refers to a condition or question that a user asks the system to provide.

[0008] A "terminal" refers to a device used by a user to access a system and enter requests.

[0009] A "server" refers to a computer system that analyzes user requests and provides appropriate information in conjunction with a database.

[0010] "Natural language processing" refers to understanding user requests using natural language processing techniques.

[0011] A "database" refers to a storage system that stores related information and allows it to be retrieved as needed.

[0012] "Generative artificial intelligence" refers to technology that generates optimal suggestions based on user requests and database information.

[0013] "Suggestions" refer to the optimal options or solutions provided to the user by generative artificial intelligence.

[0014] "Feedback" refers to the opinions and evaluations that users provide regarding suggestions or system operation.

[0015] "Charge plan" refers to the price setting of communication services that users can select.

[0016] "Data transfer" refers to the process of transferring a user's image data, etc. between different operating systems.

[0017] "Action" refers to the choices and operations that a user performs after receiving a proposal.

Brief Explanation of Drawings

[0018] [Figure 1] It is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] It is a conceptual diagram showing an example of the main functions of a data processing device and a smart device according to the first embodiment. [Figure 3] It is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] It is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] It is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] It is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] It is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] It is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] It shows an emotion map to which multiple emotions are mapped. [Figure 10] It shows an emotion map to which multiple emotions are mapped. [Figure 11] It is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12]It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Embodiment 2 when combined with an emotion engine. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when combined with an emotion engine.

Mode for Carrying Out the Invention

[0019] 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.

[0020] First, the terms used in the following description will be explained.

[0021] In the following embodiments, the numbered processor (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.

[0022] In the following embodiments, the numbered RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0023] 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.

[0024] 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).

[0025] 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."

[0026] [First Embodiment]

[0027] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0028] 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.

[0029] 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).

[0030] 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.

[0031] 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.

[0032] 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.

[0033] 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.

[0034] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0035] 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.

[0036] 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.

[0037] 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.

[0038] 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".

[0039] This invention is a customer support system that provides appropriate suggestions to meet the diverse needs of users, and is implemented as follows.

[0040] System Configuration

[0041] This system consists of a terminal for inputting user requests, a server that analyzes the requests and provides appropriate information, and a database. Furthermore, generative artificial intelligence processes the data between these components to generate optimal suggestions.

[0042] Program processing

[0043] Enter the user's request.

[0044] Terminal: The user enters their request through the terminal application. For example, the user might enter "I want the latest ANDROID® model, but I want to use it as cheaply as possible" into the terminal application.

[0045] Sending and receiving requests

[0046] Terminal: The entered request is sent from the terminal to the server. HTTP requests are used to send the request data to the server.

[0047] Server: The server receives this request data and proceeds to the next step.

[0048] Request Natural Language Processing

[0049] Server: The server analyzes the received request using a natural language processing engine to extract the user's intent. For example, it might understand that the user is "looking for the latest model at a low price."

[0050] Database matching

[0051] Server: Based on the analysis results, retrieves information on the latest Android models and pricing plans from the database. It uses SQL queries such as "SELECT model, price FROM android_devices WHERE release_date > '2023-01-01'".

[0052] AI-powered proposal generation

[0053] Server: Based on data acquired and analysis results using generative artificial intelligence, it generates optimal suggestions for the user. For example, it utilizes the TENSORFLOW® model to suggest the optimal model and pricing plan, taking into account price, performance, and the user's past usage history.

[0054] Format and send the proposal.

[0055] Server: Formats the generated proposals into an output format that is easy for the user to understand. Converts the formatted proposal content into JSON format and sends it to the terminal as an HTTP response.

[0056] Display of proposed content

[0057] Terminal: Displays suggestions received from the server to the user. Users can check things like "the latest Android models, their prices, and promotional information."

[0058] User selections and actions

[0059] User: Review the suggestions and choose the most suitable option. For example, the user might "select a specific Android model and proceed with the purchase."

[0060] Device: Triggers the following action based on the user's selection: redirect to a purchase link or initiate a change of price plan.

[0061] Feedback collection and analysis

[0062] User: Enter feedback on the proposal. Provide comments such as, "This plan does not provide enough data capacity."

[0063] Terminal: Sends feedback data to the server.

[0064] Server: Analyzes received feedback and uses it as data to improve generative artificial intelligence and the system. This is reflected in future suggestions, improving the user experience.

[0065] Specific example

[0066] Specifically, if a user enters "I want to transfer 50,000 images from iCloud to Android," the system will operate as follows:

[0067] 1. Natural Language Processing: The server extracts the request content, "Migrate iCloud images."

[0068] 2. Database matching: Retrieve information about data migration from iCloud to Android from the database.

[0069] 3. Proposal Generation: Using generative artificial intelligence, we propose the optimal data migration application and procedure.

[0070] 4. Display of proposed content: Display the proposed content on the user's device and provide specific instructions and links.

[0071] 5. Feedback Collection: After users complete the migration process, they will provide feedback about their experience.

[0072] This system allows users to solve problems quickly and accurately, leading to increased satisfaction.

[0073] The following describes the processing flow.

[0074] Step 1:

[0075] Terminal: The user launches the application on their terminal and enters their request. For example, they might enter, "I want the latest Android model, but I want to use it as cheaply as possible."

[0076] Step 2:

[0077] Terminal: Sends the entered request to the server in an appropriate data format such as JSON. Sends it to the server as an HTTP request.

[0078] Step 3:

[0079] Server: Passes the received request to the natural language processing engine for analysis. For example, it might use the Python NLTK library to extract the user's intent.

[0080] Step 4:

[0081] Server: Based on the results of natural language processing, it retrieves the latest Android model and pricing plan information from the database using SQL queries. For example, it executes a query such as "SELECT model, price FROM android_devices WHERE release_date > '2023-01-01'".

[0082] Step 5:

[0083] Server: Inputs acquired data into a generative artificial intelligence system to generate optimal suggestions for the user. For example, it uses a TensorFlow model to suggest models and plans considering price, performance, and the user's past usage history.

[0084] Step 6:

[0085] Server: Formats the generated proposal content and converts it into a JSON response. The generated proposal includes details such as "model name," "price," and "campaign information."

[0086] Step 7:

[0087] Server: Sends the formatted proposal content to the terminal as an HTTP response.

[0088] Step 8:

[0089] Device: Displays received suggestions to the user. For example, it displays detailed information and pricing for new Android models, as well as related promotional information.

[0090] Step 9:

[0091] User: Review the suggestions and choose the most suitable option. For example, select a specific Android model and proceed with the purchase.

[0092] Step 10:

[0093] Device: Based on the user's selection, the following actions are triggered. For example, redirecting to the purchase page for the selected Android model, or initiating the process of changing the price plan.

[0094] Step 11:

[0095] User: Enter feedback on the proposal. For example, provide comments such as, "This plan does not provide enough data capacity."

[0096] Step 12:

[0097] Terminal: Sends the input feedback data to the server.

[0098] Step 13:

[0099] Server: Analyzes received feedback and uses it to improve the generative artificial intelligence and the system. For example, it analyzes the content of the feedback and incorporates it into future suggestions.

[0100] (Example 1)

[0101] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0102] There is a need for customer support systems that can provide quick and appropriate solutions to the diverse needs of today's users. In particular, when user requests are complex, it is not easy to accurately understand those requests and provide appropriate information. Traditional systems often fail to adequately analyze user requests and can only provide insufficient solutions, leading to decreased user satisfaction.

[0103] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0104] In this invention, the server includes means for performing natural language analysis on user requests, means for obtaining appropriate information from a database based on the natural language analysis results, and means for generating optimal suggestions for the user using generative artificial intelligence. This makes it possible to accurately understand user requests and quickly provide optimal suggestions.

[0105] "Means for inputting user requests" refers to devices or software that provide an interface that allows users to input arbitrary requests or questions in text or voice.

[0106] "Means for sending the request to the server" refers to communication protocols, devices, and software for sending the input request to the server via the Internet.

[0107] "Means for performing natural language analysis on the server" refers to natural language processing technology and software used by the server to analyze the received request and understand the user's intent.

[0108] "Means for obtaining appropriate information from a database based on the results of natural language analysis" refers to a process and software for searching and obtaining appropriate data from a database based on the information obtained as a result of natural language analysis.

[0109] "Means of generating optimal suggestions for users using generative artificial intelligence" refers to the process and software that uses generative artificial intelligence (e.g., machine learning models) to generate optimal suggestions based on analyzed requests and acquired data.

[0110] "Means for sending the generated proposal to the user's terminal" refers to the communication protocol, device, and software that send the generated proposal from the server to the user's terminal.

[0111] "Means for displaying the proposal to the user" refers to an interface and software for visually displaying the received proposal content on the user's terminal.

[0112] "Means by which the user selects an action based on the displayed suggestion" refers to an input device and software for the user to make a selection or take action based on the suggested content.

[0113] "Means for transmitting the user's actions to the server" refers to communication protocols, devices, and software that transmit the user's selected actions and feedback to the server.

[0114] "Means for receiving and analyzing user feedback on the server" refers to the process and software for receiving feedback sent by the user and analyzing its contents.

[0115] "Means for updating the generative artificial intelligence based on the aforementioned feedback" refers to the process and software for improving and updating the algorithms and models of the generative artificial intelligence based on feedback obtained from users.

[0116] This invention is a customer support system that provides appropriate suggestions to meet diverse user requests, and is implemented using the following hardware and software. The system consists of a terminal for inputting user requests, a server that analyzes the requests and provides appropriate information, and a database. Furthermore, a generative artificial intelligence processes data between these components to generate optimal suggestions.

[0117] Hardware and software

[0118] Terminal: A device used by users to input requests, including smartphones, tablets, and personal computers. This terminal has applications installed that provide a user interface.

[0119] Server: A central computer that analyzes requests and provides appropriate information. The server has natural language processing engines (e.g., spaCy and NLTK), generative artificial intelligence (e.g., TensorFlow and PyTorch), and database management systems (e.g., MySQL® and PostgreSQL) installed.

[0120] Database: A management system for storing information related to user requests, including the latest Android models and pricing plan information.

[0121] System operation

[0122] The user enters a request through an application on their device, such as "I want the latest Android model, but I want to use it as cheaply as possible." This input is sent to the server using an HTTP request. The server receives this request data and parses it using a natural language processing engine. Based on the analysis results, the server retrieves relevant information from its database. Using this retrieved information, a generative artificial intelligence generates the most suitable suggestion for the user.

[0123] The generated suggestions are formatted, converted to JSON format, and sent to the device as an HTTP response. The device then displays the suggestions to the user. Based on the suggested information, the user can choose an action and proceed with the next steps. For example, the user can select the most suitable Android model and proceed with the purchase.

[0124] Specific example

[0125] Specifically, if a user wants to know "the latest Android models, their prices, and promotional information," the system works as follows: The user enters "I want the latest Android model, but I want to use it as cheaply as possible" through an application on their device. This request is sent to the server, which performs natural language processing to understand the user's intent. Next, it retrieves relevant information from the database and generates optimal suggestions using generative artificial intelligence. These suggestions are sent to the device, and the user can review them.

[0126] Example of a prompt

[0127] "Please tell me how to buy the latest Android model cheaply."

[0128] "I want to transfer 50,000 images from iCloud to my Android device."

[0129] "Please suggest which device is best suited to our new pricing plan."

[0130] This system allows users to resolve problems quickly and accurately and receive highly satisfactory service.

[0131] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0132] Step 1:

[0133] User: The user enters their request through the application on their device. For example, they might enter, "I want the latest Android model, but I want to use it as cheaply as possible." The entered text is saved to the application's internal memory.

[0134] Input: User request written in natural language

[0135] Output: Request text stored in the terminal's internal memory

[0136] Example of operation: The user enters "I want the latest Android model, but I want to use it as cheaply as possible" into the application's text box and presses the submit button.

[0137] Step 2:

[0138] Terminal: Sends user-entered requests to the server. Specifically, request data is sent to the server as an HTTP POST request. The request body contains user input in JSON format.

[0139] Input: Request text

[0140] Output: HTTP request sent to the server

[0141] Example of operation: The device sends an HTTP POST request containing "request data" to the " / api / requests" endpoint.

[0142] Step 3:

[0143] Server: The server parses the received request data. First, it extracts the JSON data from the request body and then retrieves the request text from within it.

[0144] Input: HTTP Request

[0145] Output: Request text

[0146] Example of operation: The server extracts a request from the body of an HTTP request it receives, such as "I want the latest Android model, but I want to use it as cheaply as possible."

[0147] Step 4:

[0148] Server: The server analyzes the acquired text using a natural language processing engine (e.g., spaCy or NLTK) to extract the user's intent. Specifically, it analyzes keywords and phrases in the text to extract intents such as "latest model" or "buy cheaply."

[0149] Input: Request text

[0150] Output: Analyzed intent (e.g., keywords or phrases)

[0151] Specific example of operation: The server uses spaCy to analyze the text "I want the latest Android model, but I want to use it as cheaply as possible" and extracts the intents "latest model" and "buy cheaply".

[0152] Step 5:

[0153] Server: Based on the analysis results, retrieve relevant information from the database. Specifically, it generates SQL queries and sends requests to the database. For example, it executes an SQL query such as "SELECT model, price FROM android_devices WHERE release_date > '2023-01-01'".

[0154] Input: Analyzed intent

[0155] Output: Information retrieved from the database (e.g., the latest Android models and their prices)

[0156] Example of operation: The server executes the query "SELECT model, price FROM android_devices WHERE release_date > '2023-01-01'" on the MySQL database to retrieve the latest Android models and their prices.

[0157] Step 6:

[0158] Server: Generates optimal suggestions for the user based on information and analysis results obtained using generative artificial intelligence (e.g., TensorFlow or PyTorch). Suggestion generation considers both the user's requests and information obtained from the database.

[0159] Input: Information retrieved from the database, analyzed intent

[0160] Output: Generated proposals

[0161] A concrete example of operation: The server uses a TensorFlow model to generate suggestions such as, "The latest Android model X can be purchased for 30,000 yen."

[0162] Step 7:

[0163] Server: Formats the generated proposal, converts it to JSON format, and sends it to the user's terminal as an HTTP response.

[0164] Input: Generated proposal

[0165] Output: HTTP response (JSON format) sent to the terminal.

[0166] Specific example of operation: The server formats the proposed content into JSON format such as "Model X: 30,000 yen, Campaign Y" and sends it to the terminal via HTTP response.

[0167] Step 8:

[0168] Terminal: Displays the received suggestions to the user. Specifically, the application's UI analyzes the suggestions and displays them in a visually easy-to-understand format.

[0169] Input: HTTP response (JSON format)

[0170] Output: Displayed suggestions

[0171] Example of operation: The device parses the JSON data it receives and displays a message on the screen such as, "The latest Android model X is available for purchase for 30,000 yen."

[0172] Step 9:

[0173] User: Review the suggestions and choose the most suitable option. For example, the user selects a specific Android model and proceeds with the purchase.

[0174] Input: Displayed suggestion

[0175] Output: User selection (e.g., purchase decision)

[0176] Example of the operation: The user clicks the "Purchase button" and is redirected to a screen to proceed with the purchase process.

[0177] Step 10:

[0178] Device: Triggers the following action based on the user's selection: redirect to a purchase link or initiate a change of price plan.

[0179] Input: User Selection

[0180] Output: Next triggered action

[0181] Example of the operation: The user clicks the "Purchase button" and is redirected to a screen to proceed with the purchase process.

[0182] Step 11:

[0183] User: Enter feedback on the proposal. For example, provide comments such as, "This plan does not provide enough data capacity."

[0184] Input: Feedback on the proposed content

[0185] Output: Feedback entered into the terminal

[0186] Example of the operation: The user enters "This plan does not provide enough data capacity" into the feedback text box.

[0187] Step 12:

[0188] Terminal: Sends feedback data to the server. Specifically, it sends feedback data to the server as an HTTP POST request. The request body contains the feedback in JSON format.

[0189] Input: Feedback text

[0190] Output: HTTP request sent to the server

[0191] Specific example of operation: The terminal sends an HTTP POST request containing "feedback data" to " / api / feedback".

[0192] Step 13:

[0193] Server: Analyzes received feedback data and uses it to improve generative artificial intelligence and systems. This allows for the incorporation of feedback into future suggestions, thereby improving the user experience.

[0194] Input: HTTP request (feedback data)

[0195] Output: Updated generative AI models and system improvement data

[0196] Specific example of operation: The server analyzes the feedback data it receives and uses it to refine the proposed algorithm for the next time.

[0197] (Application Example 1)

[0198] 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."

[0199] In modern e-commerce sites, users often struggle to quickly find the right product from a vast selection. Furthermore, customer support is required to provide prompt and accurate responses to diverse user needs. However, conventional systems have been unable to effectively utilize natural language processing and generative artificial intelligence, making it difficult to improve the user experience. Solving this problem is the objective of this invention.

[0200] 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.

[0201] In this invention, the server includes means for analyzing intent using a natural language processing engine, means for retrieving products and FAQs from a database, and means for returning suggestions in JSON format. This makes it possible to accurately analyze user requests and provide optimal product suggestions and support information quickly and accurately.

[0202] A "user" is a person who uses a system to input requests.

[0203] "Requests" refer to all kinds of wishes and questions that users input into the system.

[0204] A "server" is a computer that receives and analyzes user requests, and generates and provides appropriate information in conjunction with a database.

[0205] "Natural language processing" is a technology that analyzes the requests received from users and understands their intentions.

[0206] "Generative artificial intelligence" refers to an algorithm or system that generates optimal suggestions based on analyzed information.

[0207] A "terminal" is an electronic device used by a user to input requests and display suggestions from a server.

[0208] A "database" is a storage device that stores information such as product information and FAQs.

[0209] "Feedback" refers to the opinions and evaluations that users provide regarding a proposal.

[0210] A "natural language processing engine" is software or an algorithm that analyzes user requests and understands their intent.

[0211] A "proposal" is information or a choice provided to the user, generated by a generative artificial intelligence system.

[0212] "Products and FAQs" refer to product information and answers to frequently asked questions stored in the database.

[0213] JSON format is a lightweight data exchange format for structuring, storing, and transmitting data.

[0214] This invention provides appropriate product suggestions and support information in response to user requests in a customer support system for an e-commerce site. The system analyzes the user's request using natural language processing, generates optimal suggestions using generative artificial intelligence, and provides them to the user. Specific embodiments are described below.

[0215] System Configuration

[0216] The system will be implemented with the following configuration.

[0217] Device: An electronic device (e.g., smartphone, tablet) where the user enters a request and receives a suggestion.

[0218] Server: A computer that analyzes user requests and generates appropriate information.

[0219] Database: A storage device that stores product information and FAQs.

[0220] Natural language processing engine: Software used to analyze user requests.

[0221] Generative artificial intelligence: An algorithm that generates optimal suggestions based on analysis results.

[0222] Program processing

[0223] Entering and submitting requests

[0224] The user enters the request using a terminal. The request is sent to the server using an HTTP request.

[0225] natural language analysis

[0226] The server analyzes the received request using a natural language processing engine. Specifically, it uses the Hugging Face Transformers library to understand the user's intent.

[0227] Retrieving information from the database

[0228] Based on the analysis results, the server retrieves product information and FAQs from the database using SQL queries. For example, it might execute a query like "SELECT name, price, description FROM products WHERE description LIKE '%Summer Cooling Mattress%'".

[0229] Proposal generation

[0230] The server uses generative artificial intelligence (e.g., TensorFlow models) to generate optimal suggestions based on the acquired data. The suggestions are formatted in JSON format and sent to the terminal as an HTTP response.

[0231] Display of proposed content

[0232] The user's device displays the received suggestions. This allows the user to view specific product information and answers to frequently asked questions.

[0233] Feedback collection and analysis

[0234] Users input feedback on the suggestions and send it from their device to the server. The server analyzes this feedback and updates the generative artificial intelligence to improve the accuracy of future suggestions.

[0235] Specific example

[0236] If a user enters "I want a cooling mattress for summer," the system will operate as follows:

[0237] 1. The server uses a natural language processing engine to analyze the request, "cooling mattress for summer."

[0238] 2. Retrieve the relevant product information from the database.

[0239] 3. Generate optimal product suggestions using generative artificial intelligence.

[0240] 4. Format the generated proposals in JSON format and send them to the terminal.

[0241] 5. The user reviews the suggestion and chooses an action.

[0242] Example of a prompt

[0243] User: I want a cooling mattress for summer.

[0244] System: We found several mattresses made with cooling materials for summer. The following products are recommended.

[0245] Product name: "Summer Sleep Mattress"

[0246] Price: 5,000 yen

[0247] Description: This mattress uses cooling material for a comfortable sleeping experience.

[0248] In this way, we can provide quick and accurate solutions to meet user requests.

[0249] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0250] Step 1:

[0251] The user enters a request into the terminal and sends it. The user specifically enters, "I want a cooling mattress for summer." The input data is sent to the server as an HTTP request.

[0252] Step 2:

[0253] The server analyzes the received request using a natural language processing engine (e.g., Hugging Face's Transformers library). It extracts key keywords and intent from the request data and recognizes the intent as "cooling mattress for summer." This analysis result is then used for subsequent database matching.

[0254] Step 3:

[0255] Based on the analysis results, the server retrieves appropriate product information from the database. Specifically, it executes an SQL query and sends a query like "SELECT name, price, description FROM products WHERE description LIKE '%Summer Cooling Mattress%'". The database returns the corresponding product information.

[0256] Step 4:

[0257] The server processes the acquired data using generative artificial intelligence (e.g., TensorFlow models) to create optimal recommendations. It selects the best product for the user, taking into account factors such as price, performance, and the user's past search history. This information will later be returned to the user as recommendations.

[0258] Step 5:

[0259] The server formats the generated proposal in JSON format and sends it to the terminal as an HTTP response. The formatted proposal includes product name, price, and description.

[0260] Step 6:

[0261] The terminal displays the received suggestions to the user. The user can check specific product information and suggestions. Information such as "Summer Sleep Mattress" and "Comfortable Mattress with Cooling Material" will be displayed.

[0262] Step 7:

[0263] The user reviews the displayed suggestions and chooses an action, such as clicking a purchase button.

[0264] Step 8:

[0265] The device sends user actions to the server. The action data arrives at the server as an HTTP request.

[0266] Step 9:

[0267] The server receives and analyzes user feedback. It updates the generative artificial intelligence based on the feedback data and incorporates it into future suggestions. For example, if feedback states, "This mattress is comfortable, but it's expensive," the system will improve its suggestions to prioritize price.

[0268] 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.

[0269] This invention is a customer support system that provides more personalized suggestions by incorporating an emotion engine that recognizes the user's emotions, and is implemented as follows.

[0270] System Configuration

[0271] This system consists of a terminal for inputting user requests and emotions, a server that analyzes requests and emotions and provides appropriate information, a database, and generative artificial intelligence. Furthermore, an emotion engine works in conjunction with these components to provide responses based on the user's emotions.

[0272] Program processing

[0273] Enter the user's requests and feelings.

[0274] Terminal: When a user launches an application on their terminal and enters a request, their emotions are also recognized. For example, if a user enters "I want the latest Android model, but I want to use it as cheaply as possible," facial recognition and voice analysis will determine that the user is feeling dissatisfied.

[0275] Sending requests and emotions

[0276] Terminal: Sends the entered request and sentiment data to the server in an appropriate data format such as JSON. Sends it to the server as an HTTP request.

[0277] Receiving and analyzing requests and emotions

[0278] Server: Pass the received request and sentiment data to the natural language processing engine and sentiment engine for analysis. For example, use the NLTK library in Python to extract the user's intent and recognize the user's sentiment as "dissatisfaction" with the sentiment engine.

[0279] Database collation

[0280] Server: Based on the analysis results, obtain the information of the latest Android models and pricing plans from the database using SQL queries. For example, execute a query like "SELECT model, price FROM android_devices WHERE release_date > '2023-01-01'".

[0281] Proposal generation by AI

[0282] Server: Input the obtained data and the results of sentiment analysis into a generative artificial intelligence to generate an optimal proposal considering the user's sentiment. For example, include special discounts or additional benefits in the proposal to address the dissatisfaction.

[0283] Formatting and sending of proposal content

[0284] Server: Format the generated proposal content, add a message according to the user's sentiment, and convert it into a JSON-formatted response. The generated proposal includes details such as "model name", "price", "campaign information", etc.

[0285] Display of proposal content

[0286] Terminal: Display the received proposal content to the user. For example, in addition to the detailed information and price of the new Android model and related campaign information, display a special offer to address the user's dissatisfaction.

[0287] User selection and action

[0288] User: Review the suggestions and choose the most suitable option. For example, select a specific Android model and proceed with the purchase.

[0289] Device: Based on the user's selection, the following actions are triggered. For example, redirecting to the purchase page for the selected Android model, or initiating the process of changing the price plan.

[0290] Feedback collection and analysis

[0291] User: Enter feedback on the proposal. In addition to opinions such as "This plan does not have enough data capacity," also enter your feelings about the proposal.

[0292] Terminal: Acquires feedback data, analyzes it with an emotion engine, and sends it to the server.

[0293] Feedback analysis and AI updates

[0294] Server: Analyzes received feedback and sentiment data to update the generative artificial intelligence and sentiment engine models. This results in more personalized suggestions for the next session, improving user satisfaction.

[0295] Specific example

[0296] Specifically, this occurs when the emotion engine determines that a user is stressed after typing, "I want to transfer 50,000 images from iCloud to my Android device."

[0297] 1. Natural Language Processing and Sentiment Recognition: The server extracts the request "migrate iCloud images" and the emotion "stress".

[0298] 2. Database matching: Retrieve information about data migration from iCloud to Android from the database.

[0299] 3. Proposal Generation: Use generative artificial intelligence to propose the optimal data migration applications and procedures. Also, include free support offers to reduce the user's stress.

[0300] 4. Display of Proposal Content: Display the proposal content on the user's terminal and also include guidance on special support.

[0301] 5. Feedback Collection and Analysis: After the user completes the migration work, input emotions along with feedback on that experience. The server analyzes this and uses it to improve future proposals.

[0302] This system provides more personalized support considering the user's emotions, improving satisfaction.

[0303] The processing flow is described below.

[0304] Step 1:

[0305] Terminal: The user launches the application on the terminal and enters their request. For example, enter text such as "I want the latest Android model but want to use it as cheaply as possible".

[0306] Step 2:

[0307] Terminal: While the request is being entered, use the terminal's camera and microphone to analyze the user's facial expressions and voice tone in real time by the emotion engine. For example, use face recognition technology to detect emotions such as "dissatisfaction" or "excitement".

[0308] Step 3:

[0309] Terminal: Send the entered request data and emotion data to the server in an appropriate data format such as JSON. Use an HTTP request to send the data.

[0310] Step 4:

[0311] Server: Analyzes received request data and sentiment data. The natural language processing engine analyzes the text request, and the sentiment engine analyzes the sentiment data. For example, it extracts the request "I'm looking for the latest model at a low price" and the sentiment "dissatisfied."

[0312] Step 5:

[0313] Server: Based on the request analysis results, retrieve appropriate information from the database. Use SQL queries to retrieve information on the latest Android models and pricing plans. For example, execute the query "SELECT model, price FROM android_devices WHERE release_date > '2023-01-01'".

[0314] Step 6:

[0315] Server: Based on the acquired information, a generative artificial intelligence generates optimal suggestions. Furthermore, it adjusts the suggestions considering the sentiment analysis results. For example, if the user has the emotion of "dissatisfaction," it will include special discounts or additional benefits in its suggestions.

[0316] Step 7:

[0317] Server: Formats the generated suggestions, adds sentiment-based messages, and converts them into a JSON response. The generated suggestions include details such as "model name," "price," and "campaign information."

[0318] Step 8:

[0319] Server: Sends the formatted proposal content to the terminal as an HTTP response.

[0320] Step 9:

[0321] Device: Displays received suggestions to the user. For example, it displays detailed information and pricing for new Android models, related promotional information, and special offers to address user complaints.

[0322] Step 10:

[0323] User: Review the suggestions and choose the most suitable option. For example, select a specific Android model and proceed with the purchase.

[0324] Step 11:

[0325] Device: Based on the user's selection, trigger the following actions. For example, redirect to the purchase page for the selected Android model, or initiate the process of changing the price plan.

[0326] Step 12:

[0327] User: Enter feedback on the proposed plan. This includes comments such as "This plan does not provide sufficient data capacity," as well as your feelings at the time of giving feedback.

[0328] Step 13:

[0329] Terminal: Acquires feedback data and sentiment data and sends it to the server.

[0330] Step 14:

[0331] Server: Analyzes received feedback and emotion data to update the generative artificial intelligence and emotion engine models. This results in more personalized suggestions for the next session, improving user satisfaction.

[0332] This system provides personalized suggestions that take user emotions into consideration, thereby improving user satisfaction.

[0333] (Example 2)

[0334] 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".

[0335] Traditional customer support systems, while providing appropriate information in response to user requests, failed to consider user emotions, resulting in lower user satisfaction. Furthermore, they lacked systems for effectively utilizing user feedback and improving suggestions. This made it difficult to provide optimal service to users.

[0336] 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.

[0337] In this invention, the server includes means for inputting user requests and sentiment data; means for transmitting the requests and sentiment data to the server; means for performing natural language processing and sentiment analysis on the requests and sentiment data on the server; means for obtaining appropriate information from a database based on the natural language processing results and sentiment analysis results; means for generating optimal suggestions for the user using generative artificial intelligence; means for transmitting the generated suggestions to the user's terminal; means for displaying the suggestions to the user; means for the user to select an action based on the displayed suggestions; means for transmitting the user's actions to the server; means for receiving and analyzing user feedback on the server; and means for updating the generative artificial intelligence and sentiment analysis engine based on the feedback. This enables the provision of personalized suggestions that take user sentiment into consideration, thereby improving user satisfaction. Furthermore, it enables continuous improvement of the system based on feedback.

[0338] "Means for inputting user requests and emotional data" refers to functions and devices that allow users to input their desired requests and their emotional state at the time using a terminal.

[0339] "Means for sending request and sentiment data to the server" refers to a communication function for sending request and sentiment data entered on the terminal to the server.

[0340] "Means for natural language processing and sentiment analysis of request and sentiment data" refers to a function that analyzes request data received on a server using machine learning models, etc., and analyzes sentiment data using sentiment recognition engines, etc.

[0341] "Means for obtaining appropriate information" refers to the function of retrieving necessary information from a database based on the results of natural language processing and sentiment analysis.

[0342] "A means of generating optimal suggestions for users using generative artificial intelligence" refers to a function that uses a generative artificial intelligence model to create optimal suggestions for users based on acquired information and sentiment data.

[0343] "Means for sending generated suggestions to the user's terminal" refers to a communication function for sending suggestions created by generative artificial intelligence to the user's terminal.

[0344] "Means of displaying suggestions to the user" refers to a function that displays the content of suggestions received on the terminal to the user visually or audibly.

[0345] "Means for users to choose their actions" refers to interface functions that allow users to make choices in response to displayed suggestions.

[0346] "Means for sending user actions to the server" refers to a communication function for sending user-selected action data to the server.

[0347] "Means for receiving and analyzing feedback" refers to a function that receives user feedback data on a server and analyzes that data.

[0348] "Means for updating generative artificial intelligence and sentiment analysis engines" refers to functions for continuously improving generative artificial intelligence models and sentiment analysis engines based on analyzed feedback data.

[0349] This invention relates to a customer support system that provides more personalized suggestions by incorporating an emotion engine that recognizes user emotions. The system consists of a terminal for inputting user requests and emotions, a server that analyzes requests and emotions and provides appropriate information, a database, and generative artificial intelligence.

[0350] Hardware and software configuration

[0351] terminal

[0352] A terminal is a device for users to input requests and emotional data, and includes smartphones, tablets, and PCs. The terminal's application uses the camera and microphone to acquire user input and perform facial recognition and voice analysis. A common facial recognition API (e.g., Google® Cloud Vision API) is used for facial recognition, and a common voice analysis API (e.g., Amazon Transcribe) is used for voice analysis.

[0353] server

[0354] The server analyzes received requests and sentiment data and processes them to generate appropriate suggestions. The server uses Python's Natural Language Toolkit (NLTK) and spaCy for its natural language processing engine, and machine learning models for its sentiment analysis engine. A database (e.g., an SQL database) is also integrated to retrieve relevant information based on the analysis results and sentiment data. For generative artificial intelligence, advanced generative models such as GPT-3 (registered trademark) are used.

[0355] Specific processing of the program

[0356] The program of this system performs the following specific processing.

[0357] Request and emotional input

[0358] The user launches an application on their device and enters a request, while simultaneously capturing emotional data through the camera and microphone. For example, if the user enters "I want the latest Android model, but I want to use it as cheaply as possible," the device's camera will capture a dissatisfied expression.

[0359] Sending requests and emotions

[0360] The device converts the acquired request and sentiment data into JSON format and sends it to the server as an HTTP POST request.

[0361] Receiving and analyzing requests and emotions

[0362] The server analyzes received requests and sentiment data using a natural language processing engine and a sentiment analysis engine. The natural language processing engine extracts the user's requests, and the sentiment analysis engine recognizes the user's emotions.

[0363] Database matching

[0364] Based on the analysis results, the server uses SQL queries to retrieve appropriate information from the database. For example, it might execute a query like "SELECT model, price FROM android_devices WHERE release_date > '2023-01-01'" to retrieve the latest Android model information.

[0365] AI-powered proposal generation

[0366] The server inputs prompts into a generative artificial intelligence model based on the acquired information and sentiment analysis results, generating the most suitable suggestions for the user. For example, it uses the following prompts:

[0367] "This user wants the latest Android device, but is dissatisfied with the lack of a suitable solution to their needs, as they want to use it as cheaply as possible. Please provide them with the best possible offer."

[0368] Submitting and displaying proposals

[0369] The generated suggestions are reformatted and sent to the user's device as a JSON response. The user can then visually or audibly review the suggestions on their device. For example, a suggestion such as "Latest Android model: Android XYZ, Price: 50,000 JPY, Special discount 10%" might be displayed.

[0370] Feedback collection and analysis

[0371] The user enters feedback on the suggestion into their device and sends it back to the server. The server analyzes the feedback and sentiment data, and uses the results to update the generative artificial intelligence model and sentiment analysis engine, making the next suggestion more accurate.

[0372] Specific example

[0373] For example, if a user enters "I want to transfer 50,000 images from iCloud to Android" and the sentiment analysis engine determines that the user is feeling stressed:

[0374] 1. The server extracts the request "migrate iCloud images" and the emotion "stress".

[0375] 2. Retrieve information about data migration from iCloud to Android from the database.

[0376] 3. Use generative artificial intelligence to suggest the optimal data migration app and procedure, and include a free support offer.

[0377] 4. Display the proposed content on the user's device and include information about special support.

[0378] 5. After the user completes the migration process, they input their emotions along with feedback about their experience. The server analyzes this data to help improve future suggestions.

[0379] This system provides more personalized support that takes user emotions into account, leading to increased satisfaction.

[0380] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0381] Step 1:

[0382] User: Launch the application on the device and enter a request in the text field. For example, enter a request such as, "I want the latest Android model, but I want to use it as cheaply as possible." Generate input data (request text).

[0383] Terminal: Acquires input data and generates emotion data using face recognition APIs and voice analysis APIs with the camera and microphone. For example, it can use the Google Cloud Vision API to recognize an unhappy expression and obtain emotion data.

[0384] Input: User's requested text, user's facial image, or voice data

[0385] Output: Request text, dissatisfaction data

[0386] Step 2:

[0387] Terminal: Converts the acquired request and sentiment data into JSON format and sends it to the server using an HTTP POST request.

[0388] Input: Request text, dissatisfaction data

[0389] Output: HTTP POST request (JSON format)

[0390] Step 3:

[0391] Server: Receives HTTP requests and extracts requests and sentiments from JSON data. The extracted data is then passed to a natural language processing engine (e.g., NLTK or spaCy) and a sentiment analysis engine for analysis.

[0392] Input: HTTP POST request (JSON format)

[0393] Output: Request data as analysis results (e.g., "latest Android model"), sentiment data ("dissatisfied")

[0394] Step 4:

[0395] Server: Based on the analysis results, it executes an SQL query to retrieve the necessary information from the database. For example, to retrieve "information about the latest Android models," it executes "SELECT model, price FROM android_devices WHERE release_date > '2023-01-01'".

[0396] Input: Requested data for analysis results ("latest Android model")

[0397] Output: Database query results (e.g., model name and price of a specific device)

[0398] Step 5:

[0399] Server: Based on the acquired information, it inputs prompts into a generative artificial intelligence (e.g., GPT-3) to generate the best possible suggestions for the user. For example, it uses the following prompts:

[0400] "This user wants the latest Android device, but is dissatisfied with the lack of a suitable solution to their needs, as they want to use it as cheaply as possible. Please provide them with the best possible offer."

[0401] Input: Database query results, analyzed sentiment data

[0402] Output: Generated proposal details (e.g., model name, price, special discount information)

[0403] Step 6:

[0404] Server: Converts the generated proposal into a JSON response and sends it to the terminal as an HTTP response.

[0405] Input: Generated proposal content

[0406] Output: HTTP response (JSON format)

[0407] Step 7:

[0408] Terminal: Displays suggestions received from the server to the user. For example, it displays detailed information and pricing for the latest Android models, as well as related promotional information and special offers.

[0409] Input: HTTP response (JSON format)

[0410] Output: Suggestions displayed to the user

[0411] Step 8:

[0412] User: Review the suggestions and choose the best option. For example, select a recommended Android model and proceed with the purchase.

[0413] Input: Suggestions displayed to the user

[0414] Output: User's selected action

[0415] Step 9:

[0416] Device: Based on the user's selection, the following actions are triggered. For example, redirecting to the purchase page for the selected Android model, or initiating the process of changing the price plan.

[0417] Input: User's selected action

[0418] Output: Triggers the following action (e.g., redirects to the purchase page)

[0419] Step 10:

[0420] User: Enter feedback on the proposal. For example, enter an opinion such as "This plan does not provide enough data capacity" and a feeling of "dissatisfaction".

[0421] Input: User feedback text, facial images, and audio data

[0422] Output: Feedback data and sentiment data

[0423] Step 11:

[0424] Terminal: Acquires feedback data, analyzes it with an emotion engine, and sends it to the server as an HTTP POST request.

[0425] Input: Feedback data, sentiment data

[0426] Output: HTTP POST request (feedback data and sentiment data)

[0427] Step 12:

[0428] Server: Analyzes received feedback and sentiment data to update the generative artificial intelligence and sentiment analysis engine models. This allows for more personalized suggestions and improved user satisfaction.

[0429] Input: HTTP POST request (feedback data and sentiment data)

[0430] Output: Updated generative artificial intelligence model and sentiment analysis engine

[0431] (Application Example 2)

[0432] 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".

[0433] Traditional customer support systems have failed to adequately consider not only user requests but also their emotions. As a result, users often do not receive suggestions tailored to their feelings, leading to decreased satisfaction. Furthermore, insufficient personalization that takes user emotions into account makes it difficult to offer special offers or discounts that will stimulate purchase intent. Solving these problems is essential.

[0434] 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.

[0435] In this invention, the server includes means for inputting requests and emotions from the user, means for transmitting the requests and emotions to the server, means for performing natural language analysis and sentiment analysis on the requests and emotions on the server, means for obtaining appropriate information from a database based on the natural language analysis results and sentiment analysis results, means for generating optimal suggestions that take the user's emotions into consideration using generative artificial intelligence, means for transmitting the generated suggestions to the user's terminal, means for displaying the suggestions to the user, means for the user to select an action based on the displayed suggestions, means for transmitting the user's actions to the server, means for receiving and analyzing feedback and emotions from the user on the server, means for updating the generative artificial intelligence based on the feedback and sentiment analysis results, and includes special offers and discount suggestions. This enables personalized suggestions that take the user's emotions into consideration, improving user satisfaction and stimulating purchase intent.

[0436] A "means for inputting user requests and emotions" refers to a means that provides an interface that allows users to input their requests along with their emotions at that time.

[0437] "Means for sending the aforementioned requests and emotions to the server" refers to communication means for sending requests and emotions entered by the user to the server.

[0438] "Means for performing natural language processing and sentiment analysis on the server" refers to means for analyzing requests and sentiments on the server using natural language processing technology and sentiment recognition technology.

[0439] "Means for obtaining appropriate information from a database based on the natural language analysis results and sentiment analysis results" refers to means for obtaining relevant information from a database based on the analysis results.

[0440] "A method for generating optimal suggestions that take user emotions into consideration using generative artificial intelligence" refers to a method that uses generative artificial intelligence to generate optimal suggestions while taking into account the user's emotional state.

[0441] "Means for sending the generated proposal to the user's terminal" refers to communication means for sending the generated proposal to the user's terminal.

[0442] "Means for displaying the proposal to the user" refers to means for displaying the content of the proposal to the user on a terminal.

[0443] "Means by which the user selects an action based on the displayed suggestion" refers to means equipped with an interface for the user to select an action based on the displayed suggestion.

[0444] "Means for transmitting the user's actions to the server" refers to communication means for transmitting the actions selected by the user to the server.

[0445] "Means for receiving and analyzing user feedback and emotions on the server" refers to means for receiving user feedback and the emotions associated with it, and for analyzing them.

[0446] "Means for updating the generative artificial intelligence based on the aforementioned feedback and sentiment analysis results" refers to means for updating the model of the generative artificial intelligence based on feedback obtained from the user and the results of sentiment analysis.

[0447] "Special offers and discount proposals" refer to offers that include special discounts or promotions for users.

[0448] This invention is a customer support system that recognizes user emotions and makes emotion-based suggestions. The system consists of a user terminal, a server, a database, a generative artificial intelligence system, and an emotion engine.

[0449] System Configuration

[0450] 1. User terminal

[0451] Users access the system using smartphones or tablets.

[0452] Users utilize an interface that allows them to input requests and the emotions associated with those requests. Emotion recognition employs facial recognition and voice analysis technologies.

[0453] Specific software libraries such as OpenCV (face recognition) and TextBlob (speech analysis) are used.

[0454] 2. Server

[0455] Receives request and sentiment data sent from the user's terminal.

[0456] The data is analyzed using a natural language processing engine (e.g., Python's NLTK library) and an emotion engine (deep learning model).

[0457] Based on the analysis results, relevant information is retrieved from the database. Here, SQL queries are used to obtain the necessary information from the database.

[0458] Using generative artificial intelligence, the system generates optimal suggestions that take user emotions into consideration. These suggestions may include special discounts or cashback offers.

[0459] 3. Database

[0460] The database stores user requests and suggestions and provides information in response to queries from the server.

[0461] SQL and NoSQL databases may be used.

[0462] Explanation of the program's processing

[0463] The user uses their smartphone's camera and microphone to input their requests and associated emotions into the system. OpenCV is used for facial recognition, and TextBlob for speech analysis. This allows the system to recognize what emotions the user is experiencing (e.g., stress, frustration, satisfaction).

[0464] The server receives user requests and emotion data, analyzes the requests using a natural language processing engine, and analyzes the emotions using an emotion engine. Based on the analysis results, it retrieves appropriate information from the database and inputs it into the generative artificial intelligence. The generative AI generates optimal suggestions that take the user's emotions into account and sends these suggestions to the user's terminal in an appropriate data format such as JSON. The user reviews the suggestions on their terminal and selects an action. The server receives the user's selected action and also collects feedback information. It analyzes the feedback and the emotions associated with it, and updates the generative AI model.

[0465] Specific example

[0466] The following is an example of a prompt message used by the server to generate the best possible suggestions for the user using generative artificial intelligence:

[0467] User's emotion: Stress

[0468] Offer details: 10% special discount, 1000 yen cashback

[0469] Please create the best message to deliver to this user.

[0470] A concrete example would be a user who enters "I'm looking for the latest smartphone, but I'm worried because they're expensive," and the system determines that the user is experiencing stress. Based on the analysis, the server retrieves the latest smartphone models and price information from its database, and uses generative artificial intelligence to include special discounts and cashback offers in its suggestions. These suggestions are sent to the user's device, and the user, receiving the special offers, becomes more inclined to purchase and makes a selection. The system then collects feedback from the user and incorporates it into future suggestions.

[0471] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0472] Step 1:

[0473] The user inputs requests and emotions on their device. Using a smartphone or tablet, the user inputs their requests in text format and simultaneously inputs emotions using the camera and microphone. The input emotion data is then analyzed using OpenCV for facial recognition and TextBlob for speech analysis. This process yields request data and emotion data as input.

[0474] Step 2:

[0475] The device sends request and sentiment data to the server. The data sent from the device is in JSON format and is sent to the server as an HTTP request. The output obtained here is the request and sentiment data sent to the server.

[0476] Step 3:

[0477] The server analyzes the received request and sentiment data. First, the server uses a natural language processing engine (e.g., Python's NLTK library) to analyze the request, and then uses a sentiment engine (deep learning model) to analyze the sentiment data. The input consists of the received request and sentiment data, and the output is the analysis results.

[0478] Step 4:

[0479] The server retrieves information from the database based on the analysis results. Based on the analyzed requests and sentiments, it uses SQL queries to retrieve corresponding information from the database. For example, it executes a query such as "SELECT model, price FROM devices WHERE release_date > '2023-01-01'". The input for this step is the analysis results, and the output is the retrieved data.

[0480] Step 5:

[0481] The server uses generative artificial intelligence to generate optimal suggestions. Based on acquired data and sentiment analysis results, the generative AI generates the most suitable suggestions for the user. These suggestions may include special discounts or cashback offers. The output is the generated suggestions.

[0482] Step 6:

[0483] The server sends the generated proposal to the terminal. The generated proposal is formatted in JSON format and sent to the user's terminal as an HTTP response. The output is the proposal data sent to the terminal.

[0484] Step 7:

[0485] The terminal displays the suggestions to the user. The terminal displays the received suggestion data on the user's interface. The output is the displayed suggestion content.

[0486] Step 8:

[0487] The user selects an action based on the suggestions. The user reviews the displayed suggestions and chooses which action to take. This provides the user's selection as input.

[0488] Step 9:

[0489] The device sends user actions to the server. The user's selected action data is sent to the server in JSON format. The output is the action data sent to the server.

[0490] Step 10:

[0491] The server receives and analyzes user feedback and sentiment. It receives user feedback data and re-entered sentiment data and performs analysis. The analysis method is the same as in step 3. The input for this step is feedback and sentiment data, and the output is the analysis results.

[0492] Step 11:

[0493] The server updates the generative artificial intelligence based on feedback and sentiment analysis results. The generative AI model is updated based on the analyzed feedback and sentiment analysis results. This results in more personalized suggestions for future requests. The output is the updated generative AI model.

[0494] Step 12:

[0495] The server generates new proposals based on the updated generative artificial intelligence model. Using the updated model based on feedback, it prepares more optimal proposals for the next request. The inputs to this step are the updated model and the new request, and the output is the new proposal.

[0496] 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.

[0497] 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.

[0498] 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.

[0499] [Second Embodiment]

[0500] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0501] 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.

[0502] 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).

[0503] 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.

[0504] 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.

[0505] 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).

[0506] 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.

[0507] 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.

[0508] 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.

[0509] 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.

[0510] 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.

[0511] 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".

[0512] This invention is a customer support system that provides appropriate suggestions to meet the diverse needs of users, and is implemented as follows.

[0513] System Configuration

[0514] This system consists of a terminal for inputting user requests, a server that analyzes the requests and provides appropriate information, and a database. Furthermore, generative artificial intelligence processes the data between these components to generate optimal suggestions.

[0515] Program processing

[0516] Enter the user's request.

[0517] Terminal: The user enters their request through the terminal application. For example, the user might enter "I want the latest Android model, but I want to use it as cheaply as possible" into the terminal application.

[0518] Sending and receiving requests

[0519] Terminal: The entered request is sent from the terminal to the server. HTTP requests are used to send the request data to the server.

[0520] Server: The server receives this request data and proceeds to the next step.

[0521] Request Natural Language Processing

[0522] Server: The server analyzes the received request using a natural language processing engine to extract the user's intent. For example, it might understand that the user is "looking for the latest model at a low price."

[0523] Database matching

[0524] Server: Based on the analysis results, retrieves information on the latest Android models and pricing plans from the database. It uses SQL queries such as "SELECT model, price FROM android_devices WHERE release_date > '2023-01-01'".

[0525] AI-powered proposal generation

[0526] Server: Based on data acquired and analyzed using generative artificial intelligence, it generates optimal suggestions for the user. For example, it utilizes TensorFlow models to suggest the best model and pricing plan, taking into account price, performance, and the user's past usage history.

[0527] Format and send the proposal.

[0528] Server: Formats the generated proposals into an output format that is easy for the user to understand. Converts the formatted proposal content into JSON format and sends it to the terminal as an HTTP response.

[0529] Display of proposed content

[0530] Terminal: Displays suggestions received from the server to the user. Users can check things like "the latest Android models, their prices, and promotional information."

[0531] User selections and actions

[0532] User: Review the suggestions and choose the most suitable option. For example, the user might "select a specific Android model and proceed with the purchase."

[0533] Device: Triggers the following action based on the user's selection: redirect to a purchase link or initiate a change of price plan.

[0534] Feedback collection and analysis

[0535] User: Enter feedback on the proposal. Provide comments such as, "This plan does not provide enough data capacity."

[0536] Terminal: Sends feedback data to the server.

[0537] Server: Analyzes received feedback and uses it as data to improve generative artificial intelligence and the system. This is reflected in future suggestions, improving the user experience.

[0538] Specific example

[0539] Specifically, if a user enters "I want to transfer 50,000 images from iCloud to Android," the system will operate as follows:

[0540] 1. Natural Language Processing: The server extracts the request content, "Migrate iCloud images."

[0541] 2. Database matching: Retrieve information about data migration from iCloud to Android from the database.

[0542] 3. Proposal Generation: Using generative artificial intelligence, we propose the optimal data migration application and procedure.

[0543] 4. Display of proposed content: Display the proposed content on the user's device and provide specific instructions and links.

[0544] 5. Feedback Collection: After users complete the migration process, they will provide feedback about their experience.

[0545] This system allows users to solve problems quickly and accurately, leading to increased satisfaction.

[0546] The following describes the processing flow.

[0547] Step 1:

[0548] Terminal: The user launches the application on their terminal and enters their request. For example, they might enter, "I want the latest Android model, but I want to use it as cheaply as possible."

[0549] Step 2:

[0550] Terminal: Sends the entered request to the server in an appropriate data format such as JSON. Sends it to the server as an HTTP request.

[0551] Step 3:

[0552] Server: Passes the received request to the natural language processing engine for analysis. For example, it might use the Python NLTK library to extract the user's intent.

[0553] Step 4:

[0554] Server: Based on the results of natural language processing, it retrieves the latest Android model and pricing plan information from the database using SQL queries. For example, it executes a query such as "SELECT model, price FROM android_devices WHERE release_date > '2023-01-01'".

[0555] Step 5:

[0556] Server: Inputs acquired data into a generative artificial intelligence system to generate optimal suggestions for the user. For example, it uses a TensorFlow model to suggest models and plans considering price, performance, and the user's past usage history.

[0557] Step 6:

[0558] Server: Formats the generated proposal content and converts it into a JSON response. The generated proposal includes details such as "model name," "price," and "campaign information."

[0559] Step 7:

[0560] Server: Sends the formatted proposal content to the terminal as an HTTP response.

[0561] Step 8:

[0562] Device: Displays received suggestions to the user. For example, it displays detailed information and pricing for new Android models, as well as related promotional information.

[0563] Step 9:

[0564] User: Review the suggestions and choose the most suitable option. For example, select a specific Android model and proceed with the purchase.

[0565] Step 10:

[0566] Device: Based on the user's selection, the following actions are triggered. For example, redirecting to the purchase page for the selected Android model, or initiating the process of changing the price plan.

[0567] Step 11:

[0568] User: Enter feedback on the proposal. For example, provide comments such as, "This plan does not provide enough data capacity."

[0569] Step 12:

[0570] Terminal: Sends the input feedback data to the server.

[0571] Step 13:

[0572] Server: Analyzes received feedback and uses it to improve the generative artificial intelligence and the system. For example, it analyzes the content of the feedback and incorporates it into future suggestions.

[0573] (Example 1)

[0574] 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."

[0575] There is a need for customer support systems that can provide quick and appropriate solutions to the diverse needs of today's users. In particular, when user requests are complex, it is not easy to accurately understand those requests and provide appropriate information. Traditional systems often fail to adequately analyze user requests and can only provide insufficient solutions, leading to decreased user satisfaction.

[0576] 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.

[0577] In this invention, the server includes means for performing natural language analysis on user requests, means for obtaining appropriate information from a database based on the natural language analysis results, and means for generating optimal suggestions for the user using generative artificial intelligence. This makes it possible to accurately understand user requests and quickly provide optimal suggestions.

[0578] "Means for inputting user requests" refers to devices or software that provide an interface that allows users to input arbitrary requests or questions in text or voice.

[0579] "Means for sending the request to the server" refers to communication protocols, devices, and software for sending the input request to the server via the Internet.

[0580] "Means for performing natural language analysis on the server" refers to natural language processing technology and software used by the server to analyze the received request and understand the user's intent.

[0581] "Means for obtaining appropriate information from a database based on the results of natural language analysis" refers to a process and software for searching and obtaining appropriate data from a database based on the information obtained as a result of natural language analysis.

[0582] "Means of generating optimal suggestions for users using generative artificial intelligence" refers to the process and software that uses generative artificial intelligence (e.g., machine learning models) to generate optimal suggestions based on analyzed requests and acquired data.

[0583] "Means for sending the generated proposal to the user's terminal" refers to the communication protocol, device, and software that send the generated proposal from the server to the user's terminal.

[0584] "Means for displaying the proposal to the user" refers to an interface and software for visually displaying the received proposal content on the user's terminal.

[0585] "Means by which the user selects an action based on the displayed suggestion" refers to an input device and software for the user to make a selection or take action based on the suggested content.

[0586] "Means for transmitting the user's actions to the server" refers to communication protocols, devices, and software that transmit the user's selected actions and feedback to the server.

[0587] "Means for receiving and analyzing user feedback on the server" refers to the process and software for receiving feedback sent by the user and analyzing its contents.

[0588] "Means for updating the generative artificial intelligence based on the aforementioned feedback" refers to the process and software for improving and updating the algorithms and models of the generative artificial intelligence based on feedback obtained from users.

[0589] This invention is a customer support system that provides appropriate suggestions to meet diverse user requests, and is implemented using the following hardware and software. The system consists of a terminal for inputting user requests, a server that analyzes the requests and provides appropriate information, and a database. Furthermore, a generative artificial intelligence processes data between these components to generate optimal suggestions.

[0590] Hardware and software

[0591] Terminal: A device used by users to input requests, including smartphones, tablets, and personal computers. This terminal has applications installed that provide a user interface.

[0592] Server: A central computer that analyzes requests and provides appropriate information. The server has natural language processing engines (e.g., spaCy and NLTK), generative artificial intelligence (e.g., TensorFlow and PyTorch), and database management systems (e.g., MySQL and PostgreSQL) installed.

[0593] Database: A management system for storing information related to user requests, including the latest Android models and pricing plan information.

[0594] System operation

[0595] The user enters a request through an application on their device, such as "I want the latest Android model, but I want to use it as cheaply as possible." This input is sent to the server using an HTTP request. The server receives this request data and parses it using a natural language processing engine. Based on the analysis results, the server retrieves relevant information from its database. Using this retrieved information, a generative artificial intelligence generates the most suitable suggestion for the user.

[0596] The generated suggestions are formatted, converted to JSON format, and sent to the device as an HTTP response. The device then displays the suggestions to the user. Based on the suggested information, the user can choose an action and proceed with the next steps. For example, the user can select the most suitable Android model and proceed with the purchase.

[0597] Specific example

[0598] Specifically, if a user wants to know "the latest Android models, their prices, and promotional information," the system works as follows: The user enters "I want the latest Android model, but I want to use it as cheaply as possible" through an application on their device. This request is sent to the server, which performs natural language processing to understand the user's intent. Next, it retrieves relevant information from the database and generates optimal suggestions using generative artificial intelligence. These suggestions are sent to the device, and the user can review them.

[0599] Example of a prompt

[0600] "Please tell me how to buy the latest Android model cheaply."

[0601] "I want to transfer 50,000 images from iCloud to my Android device."

[0602] "Please suggest which device is best suited to our new pricing plan."

[0603] This system allows users to resolve problems quickly and accurately and receive highly satisfactory service.

[0604] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0605] Step 1:

[0606] User: The user enters their request through the application on their device. For example, they might enter, "I want the latest Android model, but I want to use it as cheaply as possible." The entered text is saved to the application's internal memory.

[0607] Input: User request written in natural language

[0608] Output: Request text stored in the terminal's internal memory

[0609] Example of operation: The user enters "I want the latest Android model, but I want to use it as cheaply as possible" into the application's text box and presses the submit button.

[0610] Step 2:

[0611] Terminal: Sends user-entered requests to the server. Specifically, request data is sent to the server as an HTTP POST request. The request body contains user input in JSON format.

[0612] Input: Request text

[0613] Output: HTTP request sent to the server

[0614] Example of operation: The device sends an HTTP POST request containing "request data" to the " / api / requests" endpoint.

[0615] Step 3:

[0616] Server: The server parses the received request data. First, it extracts the JSON data from the request body and then retrieves the request text from within it.

[0617] Input: HTTP Request

[0618] Output: Request text

[0619] Example of operation: The server extracts a request from the body of an HTTP request it receives, such as "I want the latest Android model, but I want to use it as cheaply as possible."

[0620] Step 4:

[0621] Server: The server analyzes the acquired text using a natural language processing engine (e.g., spaCy or NLTK) to extract the user's intent. Specifically, it analyzes keywords and phrases in the text to extract intents such as "latest model" or "buy cheaply."

[0622] Input: Request text

[0623] Output: Analyzed intent (e.g., keywords or phrases)

[0624] Specific example of operation: The server uses spaCy to analyze the text "I want the latest Android model, but I want to use it as cheaply as possible" and extracts the intents "latest model" and "buy cheaply".

[0625] Step 5:

[0626] Server: Based on the analysis results, retrieve relevant information from the database. Specifically, it generates SQL queries and sends requests to the database. For example, it executes an SQL query such as "SELECT model, price FROM android_devices WHERE release_date > '2023-01-01'".

[0627] Input: Analyzed intent

[0628] Output: Information retrieved from the database (e.g., the latest Android models and their prices)

[0629] Example of operation: The server executes the query "SELECT model, price FROM android_devices WHERE release_date > '2023-01-01'" on the MySQL database to retrieve the latest Android models and their prices.

[0630] Step 6:

[0631] Server: Generates optimal suggestions for the user based on information and analysis results obtained using generative artificial intelligence (e.g., TensorFlow or PyTorch). Suggestion generation considers both the user's requests and information obtained from the database.

[0632] Input: Information retrieved from the database, analyzed intent

[0633] Output: Generated proposals

[0634] A concrete example of operation: The server uses a TensorFlow model to generate suggestions such as, "The latest Android model X can be purchased for 30,000 yen."

[0635] Step 7:

[0636] Server: Formats the generated proposal, converts it to JSON format, and sends it to the user's terminal as an HTTP response.

[0637] Input: Generated proposal

[0638] Output: HTTP response (JSON format) sent to the terminal.

[0639] Specific example of operation: The server formats the proposed content into JSON format such as "Model X: 30,000 yen, Campaign Y" and sends it to the terminal via HTTP response.

[0640] Step 8:

[0641] Terminal: Displays the received suggestions to the user. Specifically, the application's UI analyzes the suggestions and displays them in a visually easy-to-understand format.

[0642] Input: HTTP response (JSON format)

[0643] Output: Displayed suggestions

[0644] Example of operation: The device parses the JSON data it receives and displays a message on the screen such as, "The latest Android model X is available for purchase for 30,000 yen."

[0645] Step 9:

[0646] User: Review the suggestions and choose the most suitable option. For example, the user selects a specific Android model and proceeds with the purchase.

[0647] Input: Displayed suggestion

[0648] Output: User selection (e.g., purchase decision)

[0649] Example of the operation: The user clicks the "Purchase button" and is redirected to a screen to proceed with the purchase process.

[0650] Step 10:

[0651] Device: Triggers the following action based on the user's selection: redirect to a purchase link or initiate a change of price plan.

[0652] Input: User Selection

[0653] Output: Next triggered action

[0654] Example of the operation: The user clicks the "Purchase button" and is redirected to a screen to proceed with the purchase process.

[0655] Step 11:

[0656] User: Enter feedback on the proposal. For example, provide comments such as, "This plan does not provide enough data capacity."

[0657] Input: Feedback on the proposed content

[0658] Output: Feedback entered into the terminal

[0659] Example of the operation: The user enters "This plan does not provide enough data capacity" into the feedback text box.

[0660] Step 12:

[0661] Terminal: Sends feedback data to the server. Specifically, it sends feedback data to the server as an HTTP POST request. The request body contains the feedback in JSON format.

[0662] Input: Feedback text

[0663] Output: HTTP request sent to the server

[0664] Specific example of operation: The terminal sends an HTTP POST request containing "feedback data" to " / api / feedback".

[0665] Step 13:

[0666] Server: Analyzes received feedback data and uses it to improve generative artificial intelligence and systems. This allows for the incorporation of feedback into future suggestions, thereby improving the user experience.

[0667] Input: HTTP request (feedback data)

[0668] Output: Updated generative AI models and system improvement data

[0669] Specific example of operation: The server analyzes the feedback data it receives and uses it to refine the proposed algorithm for the next time.

[0670] (Application Example 1)

[0671] 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."

[0672] In modern e-commerce sites, users often struggle to quickly find the right product from a vast selection. Furthermore, customer support is required to provide prompt and accurate responses to diverse user needs. However, conventional systems have been unable to effectively utilize natural language processing and generative artificial intelligence, making it difficult to improve the user experience. Solving this problem is the objective of this invention.

[0673] 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.

[0674] In this invention, the server includes means for analyzing intent using a natural language processing engine, means for retrieving products and FAQs from a database, and means for returning suggestions in JSON format. This makes it possible to accurately analyze user requests and provide optimal product suggestions and support information quickly and accurately.

[0675] A "user" is a person who uses a system to input requests.

[0676] "Requests" refer to all kinds of wishes and questions that users input into the system.

[0677] A "server" is a computer that receives and analyzes user requests, and generates and provides appropriate information in conjunction with a database.

[0678] "Natural language processing" is a technology that analyzes the requests received from users and understands their intentions.

[0679] "Generative artificial intelligence" refers to an algorithm or system that generates optimal suggestions based on analyzed information.

[0680] A "terminal" is an electronic device used by a user to input requests and display suggestions from a server.

[0681] A "database" is a storage device that stores information such as product information and FAQs.

[0682] "Feedback" refers to the opinions and evaluations that users provide regarding a proposal.

[0683] A "natural language processing engine" is software or an algorithm that analyzes user requests and understands their intent.

[0684] A "proposal" is information or a choice provided to the user, generated by a generative artificial intelligence system.

[0685] "Products and FAQs" refer to product information and answers to frequently asked questions stored in the database.

[0686] JSON format is a lightweight data exchange format for structuring, storing, and transmitting data.

[0687] This invention provides appropriate product suggestions and support information in response to user requests in a customer support system for an e-commerce site. The system analyzes the user's request using natural language processing, generates optimal suggestions using generative artificial intelligence, and provides them to the user. Specific embodiments are described below.

[0688] System Configuration

[0689] The system will be implemented with the following configuration.

[0690] Device: An electronic device (e.g., smartphone, tablet) where the user enters a request and receives a suggestion.

[0691] Server: A computer that analyzes user requests and generates appropriate information.

[0692] Database: A storage device that stores product information and FAQs.

[0693] Natural language processing engine: Software used to analyze user requests.

[0694] Generative artificial intelligence: An algorithm that generates optimal suggestions based on analysis results.

[0695] Program processing

[0696] Entering and submitting requests

[0697] The user enters the request using a terminal. The request is sent to the server using an HTTP request.

[0698] natural language analysis

[0699] The server analyzes the received request using a natural language processing engine. Specifically, it uses the Hugging Face Transformers library to understand the user's intent.

[0700] Retrieving information from the database

[0701] Based on the analysis results, the server retrieves product information and FAQs from the database using SQL queries. For example, it might execute a query like "SELECT name, price, description FROM products WHERE description LIKE '%Summer Cooling Mattress%'".

[0702] Proposal generation

[0703] The server uses generative artificial intelligence (e.g., TensorFlow models) to generate optimal suggestions based on the acquired data. The suggestions are formatted in JSON format and sent to the terminal as an HTTP response.

[0704] Display of proposed content

[0705] The user's device displays the received suggestions. This allows the user to view specific product information and answers to frequently asked questions.

[0706] Feedback collection and analysis

[0707] Users input feedback on the suggestions and send it from their device to the server. The server analyzes this feedback and updates the generative artificial intelligence to improve the accuracy of future suggestions.

[0708] Specific example

[0709] If a user enters "I want a cooling mattress for summer," the system will operate as follows:

[0710] 1. The server uses a natural language processing engine to analyze the request, "cooling mattress for summer."

[0711] 2. Retrieve the relevant product information from the database.

[0712] 3. Generate optimal product suggestions using generative artificial intelligence.

[0713] 4. Format the generated proposals in JSON format and send them to the terminal.

[0714] 5. The user reviews the suggestion and chooses an action.

[0715] Example of a prompt

[0716] User: I want a cooling mattress for summer.

[0717] System: We found several mattresses made with cooling materials for summer. The following products are recommended.

[0718] Product name: "Summer Sleep Mattress"

[0719] Price: 5,000 yen

[0720] Description: This mattress uses cooling material for a comfortable sleeping experience.

[0721] In this way, we can provide quick and accurate solutions to meet user requests.

[0722] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0723] Step 1:

[0724] The user enters a request into the terminal and sends it. The user specifically enters, "I want a cooling mattress for summer." The input data is sent to the server as an HTTP request.

[0725] Step 2:

[0726] The server analyzes the received request using a natural language processing engine (e.g., Hugging Face's Transformers library). It extracts key keywords and intent from the request data and recognizes the intent as "cooling mattress for summer." This analysis result is then used for subsequent database matching.

[0727] Step 3:

[0728] Based on the analysis results, the server retrieves appropriate product information from the database. Specifically, it executes an SQL query and sends a query like "SELECT name, price, description FROM products WHERE description LIKE '%Summer Cooling Mattress%'". The database returns the corresponding product information.

[0729] Step 4:

[0730] The server processes the acquired data using generative artificial intelligence (e.g., TensorFlow models) to create optimal recommendations. It selects the best product for the user, taking into account factors such as price, performance, and the user's past search history. This information will later be returned to the user as recommendations.

[0731] Step 5:

[0732] The server formats the generated proposal in JSON format and sends it to the terminal as an HTTP response. The formatted proposal includes product name, price, and description.

[0733] Step 6:

[0734] The terminal displays the received suggestions to the user. The user can check specific product information and suggestions. Information such as "Summer Sleep Mattress" and "Comfortable Mattress with Cooling Material" will be displayed.

[0735] Step 7:

[0736] The user reviews the displayed suggestions and chooses an action, such as clicking a purchase button.

[0737] Step 8:

[0738] The device sends user actions to the server. The action data arrives at the server as an HTTP request.

[0739] Step 9:

[0740] The server receives and analyzes user feedback. It updates the generative artificial intelligence based on the feedback data and incorporates it into future suggestions. For example, if feedback states, "This mattress is comfortable, but it's expensive," the system will improve its suggestions to prioritize price.

[0741] 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.

[0742] This invention is a customer support system that provides more personalized suggestions by incorporating an emotion engine that recognizes the user's emotions, and is implemented as follows.

[0743] System Configuration

[0744] This system consists of a terminal for inputting user requests and emotions, a server that analyzes requests and emotions and provides appropriate information, a database, and generative artificial intelligence. Furthermore, an emotion engine works in conjunction with these components to provide responses based on the user's emotions.

[0745] Program processing

[0746] Enter the user's requests and feelings.

[0747] Terminal: When a user launches an application on their terminal and enters a request, their emotions are also recognized. For example, if a user enters "I want the latest Android model, but I want to use it as cheaply as possible," facial recognition and voice analysis will determine that the user is feeling dissatisfied.

[0748] Sending requests and emotions

[0749] Terminal: Sends the entered request and sentiment data to the server in an appropriate data format such as JSON. Sends it to the server as an HTTP request.

[0750] Receiving and analyzing requests and emotions

[0751] Server: Passes received requests and sentiment data to the natural language processing engine and sentiment engine for analysis. For example, it uses the Python NLTK library to extract the user's intent and the sentiment engine to recognize the user's emotion as "dissatisfaction."

[0752] Database matching

[0753] Server: Based on the analysis results, retrieves the latest Android model and pricing plan information from the database using SQL queries. For example, it executes a query such as "SELECT model, price FROM android_devices WHERE release_date > '2023-01-01'".

[0754] AI-powered proposal generation

[0755] Server: Inputs acquired data and sentiment analysis results into a generative artificial intelligence system to generate optimal suggestions that take user emotions into consideration. For example, to address dissatisfaction, it might include special discounts or additional benefits in its suggestions.

[0756] Format and send the proposal.

[0757] Server: Formats the generated suggestions, adds a message tailored to the user's sentiment, and converts it into a JSON response. The generated suggestions include details such as "model name," "price," and "campaign information."

[0758] Display of proposed content

[0759] Device: Displays received suggestions to the user. For example, it displays detailed information and pricing for new Android models, related promotional information, and special offers to address user complaints.

[0760] User selections and actions

[0761] User: Review the suggestions and choose the most suitable option. For example, select a specific Android model and proceed with the purchase.

[0762] Device: Based on the user's selection, the following actions are triggered. For example, redirecting to the purchase page for the selected Android model, or initiating the process of changing the price plan.

[0763] Feedback collection and analysis

[0764] User: Enter feedback on the proposal. In addition to opinions such as "This plan does not have enough data capacity," also enter your feelings about the proposal.

[0765] Terminal: Acquires feedback data, analyzes it with an emotion engine, and sends it to the server.

[0766] Feedback analysis and AI updates

[0767] Server: Analyzes received feedback and sentiment data to update the generative artificial intelligence and sentiment engine models. This results in more personalized suggestions for the next session, improving user satisfaction.

[0768] Specific example

[0769] Specifically, this occurs when the emotion engine determines that a user is stressed after typing, "I want to transfer 50,000 images from iCloud to my Android device."

[0770] 1. Natural Language Processing and Sentiment Recognition: The server extracts the request "migrate iCloud images" and the emotion "stress".

[0771] 2. Database matching: Retrieve information about data migration from iCloud to Android from the database.

[0772] 3. Suggestion Generation: Generative artificial intelligence is used to suggest the optimal data migration app and procedure. Free support offers are also included to reduce user stress.

[0773] 4. Display of proposals: Display the proposals on the user's device and include information about special support.

[0774] 5. Feedback Collection and Analysis: After users complete the migration process, they input their emotions along with feedback about their experience. The server analyzes this data to help improve future suggestions.

[0775] This system provides more personalized support that takes user emotions into account, leading to increased satisfaction.

[0776] The following describes the processing flow.

[0777] Step 1:

[0778] Terminal: The user launches the application on their terminal and enters their request. For example, they might enter text such as, "I want the latest Android model, but I want to use it as cheaply as possible."

[0779] Step 2:

[0780] Terminal: While the user is entering a request, the terminal's camera and microphone are used to analyze the user's facial expressions and voice tone in real time using an emotion engine. For example, facial recognition technology can detect emotions such as "dissatisfaction" or "excitement."

[0781] Step 3:

[0782] Terminal: Sends the entered request data and sentiment data to the server in an appropriate data format such as JSON. Data is sent using HTTP requests.

[0783] Step 4:

[0784] Server: Analyzes received request data and sentiment data. The natural language processing engine analyzes the text request, and the sentiment engine analyzes the sentiment data. For example, it extracts the request "I'm looking for the latest model at a low price" and the sentiment "dissatisfied."

[0785] Step 5:

[0786] Server: Based on the request analysis results, retrieve appropriate information from the database. Use SQL queries to retrieve information on the latest Android models and pricing plans. For example, execute the query "SELECT model, price FROM android_devices WHERE release_date > '2023-01-01'".

[0787] Step 6:

[0788] Server: Based on the acquired information, a generative artificial intelligence generates optimal suggestions. Furthermore, it adjusts the suggestions considering the sentiment analysis results. For example, if the user has the emotion of "dissatisfaction," it will include special discounts or additional benefits in its suggestions.

[0789] Step 7:

[0790] Server: Formats the generated suggestions, adds sentiment-based messages, and converts them into a JSON response. The generated suggestions include details such as "model name," "price," and "campaign information."

[0791] Step 8:

[0792] Server: Sends the formatted proposal content to the terminal as an HTTP response.

[0793] Step 9:

[0794] Device: Displays received suggestions to the user. For example, it displays detailed information and pricing for new Android models, related promotional information, and special offers to address user complaints.

[0795] Step 10:

[0796] User: Review the suggestions and choose the most suitable option. For example, select a specific Android model and proceed with the purchase.

[0797] Step 11:

[0798] Device: Based on the user's selection, trigger the following actions. For example, redirect to the purchase page for the selected Android model, or initiate the process of changing the price plan.

[0799] Step 12:

[0800] User: Enter feedback on the proposed plan. This includes comments such as "This plan does not provide sufficient data capacity," as well as your feelings at the time of giving feedback.

[0801] Step 13:

[0802] Terminal: Acquires feedback data and sentiment data and sends it to the server.

[0803] Step 14:

[0804] Server: Analyzes received feedback and emotion data to update the generative artificial intelligence and emotion engine models. This results in more personalized suggestions for the next session, improving user satisfaction.

[0805] This system provides personalized suggestions that take user emotions into consideration, thereby improving user satisfaction.

[0806] (Example 2)

[0807] 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".

[0808] Traditional customer support systems, while providing appropriate information in response to user requests, failed to consider user emotions, resulting in lower user satisfaction. Furthermore, they lacked systems for effectively utilizing user feedback and improving suggestions. This made it difficult to provide optimal service to users.

[0809] 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.

[0810] In this invention, the server includes means for inputting user requests and sentiment data; means for transmitting the requests and sentiment data to the server; means for performing natural language processing and sentiment analysis on the requests and sentiment data on the server; means for obtaining appropriate information from a database based on the natural language processing results and sentiment analysis results; means for generating optimal suggestions for the user using generative artificial intelligence; means for transmitting the generated suggestions to the user's terminal; means for displaying the suggestions to the user; means for the user to select an action based on the displayed suggestions; means for transmitting the user's actions to the server; means for receiving and analyzing user feedback on the server; and means for updating the generative artificial intelligence and sentiment analysis engine based on the feedback. This enables the provision of personalized suggestions that take user sentiment into consideration, thereby improving user satisfaction. Furthermore, it enables continuous improvement of the system based on feedback.

[0811] "Means for inputting user requests and emotional data" refers to functions and devices that allow users to input their desired requests and their emotional state at the time using a terminal.

[0812] "Means for sending request and sentiment data to the server" refers to a communication function for sending request and sentiment data entered on the terminal to the server.

[0813] "Means for natural language processing and sentiment analysis of request and sentiment data" refers to a function that analyzes request data received on a server using machine learning models, etc., and analyzes sentiment data using sentiment recognition engines, etc.

[0814] "Means for obtaining appropriate information" refers to the function of retrieving necessary information from a database based on the results of natural language processing and sentiment analysis.

[0815] "A means of generating optimal suggestions for users using generative artificial intelligence" refers to a function that uses a generative artificial intelligence model to create optimal suggestions for users based on acquired information and sentiment data.

[0816] "Means for sending generated suggestions to the user's terminal" refers to a communication function for sending suggestions created by generative artificial intelligence to the user's terminal.

[0817] "Means of displaying suggestions to the user" refers to a function that displays the content of suggestions received on the terminal to the user visually or audibly.

[0818] "Means for users to choose their actions" refers to interface functions that allow users to make choices in response to displayed suggestions.

[0819] "Means for sending user actions to the server" refers to a communication function for sending user-selected action data to the server.

[0820] "Means for receiving and analyzing feedback" refers to a function that receives user feedback data on a server and analyzes that data.

[0821] "Means for updating generative artificial intelligence and sentiment analysis engines" refers to functions for continuously improving generative artificial intelligence models and sentiment analysis engines based on analyzed feedback data.

[0822] This invention relates to a customer support system that provides more personalized suggestions by incorporating an emotion engine that recognizes user emotions. The system consists of a terminal for inputting user requests and emotions, a server that analyzes requests and emotions and provides appropriate information, a database, and generative artificial intelligence.

[0823] Hardware and software configuration

[0824] terminal

[0825] A terminal is a device used by users to input requests and emotional data, and includes smartphones, tablets, and PCs. The terminal's application uses the camera and microphone to acquire user input and perform facial recognition and voice analysis. A common facial recognition API (e.g., Google Cloud Vision API) is used for facial recognition, and a common voice analysis API (e.g., Amazon Transcribe) is used for voice analysis.

[0826] server

[0827] The server analyzes received requests and sentiment data and processes them to generate appropriate suggestions. The server uses Python's Natural Language Toolkit (NLTK) and spaCy for its natural language processing engine, and machine learning models for its sentiment analysis engine. A database (e.g., an SQL database) is also integrated to retrieve relevant information based on the analysis results and sentiment data. For generative artificial intelligence, advanced generative models such as GPT-3 are used.

[0828] Specific processing of the program

[0829] The program of this system performs the following specific processing.

[0830] Request and emotional input

[0831] The user launches an application on their device and enters a request, while simultaneously capturing emotional data through the camera and microphone. For example, if the user enters "I want the latest Android model, but I want to use it as cheaply as possible," the device's camera will capture a dissatisfied expression.

[0832] Sending requests and emotions

[0833] The device converts the acquired request and sentiment data into JSON format and sends it to the server as an HTTP POST request.

[0834] Receiving and analyzing requests and emotions

[0835] The server analyzes received requests and sentiment data using a natural language processing engine and a sentiment analysis engine. The natural language processing engine extracts the user's requests, and the sentiment analysis engine recognizes the user's emotions.

[0836] Database matching

[0837] Based on the analysis results, the server uses SQL queries to retrieve appropriate information from the database. For example, it might execute a query like "SELECT model, price FROM android_devices WHERE release_date > '2023-01-01'" to retrieve the latest Android model information.

[0838] AI-powered proposal generation

[0839] The server inputs prompts into a generative artificial intelligence model based on the acquired information and sentiment analysis results, generating the most suitable suggestions for the user. For example, it uses the following prompts:

[0840] "This user wants the latest Android device, but is dissatisfied with the lack of a suitable solution to their needs, as they want to use it as cheaply as possible. Please provide them with the best possible offer."

[0841] Submitting and displaying proposals

[0842] The generated suggestions are reformatted and sent to the user's device as a JSON response. The user can then visually or audibly review the suggestions on their device. For example, a suggestion such as "Latest Android model: Android XYZ, Price: 50,000 JPY, Special discount 10%" might be displayed.

[0843] Feedback collection and analysis

[0844] The user enters feedback on the suggestion into their device and sends it back to the server. The server analyzes the feedback and sentiment data, and uses the results to update the generative artificial intelligence model and sentiment analysis engine, making the next suggestion more accurate.

[0845] Specific example

[0846] For example, if a user enters "I want to transfer 50,000 images from iCloud to Android" and the sentiment analysis engine determines that the user is feeling stressed:

[0847] 1. The server extracts the request "migrate iCloud images" and the emotion "stress".

[0848] 2. Retrieve information about data migration from iCloud to Android from the database.

[0849] 3. Use generative artificial intelligence to suggest the optimal data migration app and procedure, and include a free support offer.

[0850] 4. Display the proposed content on the user's device and include information about special support.

[0851] 5. After the user completes the migration process, they input their emotions along with feedback about their experience. The server analyzes this data to help improve future suggestions.

[0852] This system provides more personalized support that takes user emotions into account, leading to increased satisfaction.

[0853] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0854] Step 1:

[0855] User: Launch the application on the device and enter a request in the text field. For example, enter a request such as, "I want the latest Android model, but I want to use it as cheaply as possible." Generate input data (request text).

[0856] Terminal: Acquires input data and generates emotion data using face recognition APIs and voice analysis APIs with the camera and microphone. For example, it can use the Google Cloud Vision API to recognize an unhappy expression and obtain emotion data.

[0857] Input: User's requested text, user's facial image, or voice data

[0858] Output: Request text, dissatisfaction data

[0859] Step 2:

[0860] Terminal: Converts the acquired request and sentiment data into JSON format and sends it to the server using an HTTP POST request.

[0861] Input: Request text, dissatisfaction data

[0862] Output: HTTP POST request (JSON format)

[0863] Step 3:

[0864] Server: Receives HTTP requests and extracts requests and sentiments from JSON data. The extracted data is then passed to a natural language processing engine (e.g., NLTK or spaCy) and a sentiment analysis engine for analysis.

[0865] Input: HTTP POST request (JSON format)

[0866] Output: Request data as analysis results (e.g., "latest Android model"), sentiment data ("dissatisfied")

[0867] Step 4:

[0868] Server: Based on the analysis results, it executes an SQL query to retrieve the necessary information from the database. For example, to retrieve "information about the latest Android models," it executes "SELECT model, price FROM android_devices WHERE release_date > '2023-01-01'".

[0869] Input: Requested data for analysis results ("latest Android model")

[0870] Output: Database query results (e.g., model name and price of a specific device)

[0871] Step 5:

[0872] Server: Based on the acquired information, it inputs prompts into a generative artificial intelligence (e.g., GPT-3) to generate the best possible suggestions for the user. For example, it uses the following prompts:

[0873] "This user wants the latest Android device, but is dissatisfied with the lack of a suitable solution to their needs, as they want to use it as cheaply as possible. Please provide them with the best possible offer."

[0874] Input: Database query results, analyzed sentiment data

[0875] Output: Generated proposal details (e.g., model name, price, special discount information)

[0876] Step 6:

[0877] Server: Converts the generated proposal into a JSON response and sends it to the terminal as an HTTP response.

[0878] Input: Generated proposal content

[0879] Output: HTTP response (JSON format)

[0880] Step 7:

[0881] Terminal: Displays suggestions received from the server to the user. For example, it displays detailed information and pricing for the latest Android models, as well as related promotional information and special offers.

[0882] Input: HTTP response (JSON format)

[0883] Output: Suggestions displayed to the user

[0884] Step 8:

[0885] User: Review the suggestions and choose the best option. For example, select a recommended Android model and proceed with the purchase.

[0886] Input: Suggestions displayed to the user

[0887] Output: User's selected action

[0888] Step 9:

[0889] Device: Based on the user's selection, the following actions are triggered. For example, redirecting to the purchase page for the selected Android model, or initiating the process of changing the price plan.

[0890] Input: User's selected action

[0891] Output: Triggers the following action (e.g., redirects to the purchase page)

[0892] Step 10:

[0893] User: Enter feedback on the proposal. For example, enter an opinion such as "This plan does not provide enough data capacity" and a feeling of "dissatisfaction".

[0894] Input: User feedback text, facial images, and audio data

[0895] Output: Feedback data and sentiment data

[0896] Step 11:

[0897] Terminal: Acquires feedback data, analyzes it with an emotion engine, and sends it to the server as an HTTP POST request.

[0898] Input: Feedback data, sentiment data

[0899] Output: HTTP POST request (feedback data and sentiment data)

[0900] Step 12:

[0901] Server: Analyzes received feedback and sentiment data to update the generative artificial intelligence and sentiment analysis engine models. This allows for more personalized suggestions and improved user satisfaction.

[0902] Input: HTTP POST request (feedback data and sentiment data)

[0903] Output: Updated generative artificial intelligence model and sentiment analysis engine

[0904] (Application Example 2)

[0905] 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."

[0906] Traditional customer support systems have failed to adequately consider not only user requests but also their emotions. As a result, users often do not receive suggestions tailored to their feelings, leading to decreased satisfaction. Furthermore, insufficient personalization that takes user emotions into account makes it difficult to offer special offers or discounts that will stimulate purchase intent. Solving these problems is essential.

[0907] 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.

[0908] In this invention, the server includes means for inputting requests and emotions from the user, means for transmitting the requests and emotions to the server, means for performing natural language analysis and sentiment analysis on the requests and emotions on the server, means for obtaining appropriate information from a database based on the natural language analysis results and sentiment analysis results, means for generating optimal suggestions that take the user's emotions into consideration using generative artificial intelligence, means for transmitting the generated suggestions to the user's terminal, means for displaying the suggestions to the user, means for the user to select an action based on the displayed suggestions, means for transmitting the user's actions to the server, means for receiving and analyzing feedback and emotions from the user on the server, means for updating the generative artificial intelligence based on the feedback and sentiment analysis results, and includes special offers and discount suggestions. This enables personalized suggestions that take the user's emotions into consideration, improving user satisfaction and stimulating purchase intent.

[0909] A "means for inputting user requests and emotions" refers to a means that provides an interface that allows users to input their requests along with their emotions at that time.

[0910] "Means for sending the aforementioned requests and emotions to the server" refers to communication means for sending requests and emotions entered by the user to the server.

[0911] "Means for performing natural language processing and sentiment analysis on the server" refers to means for analyzing requests and sentiments on the server using natural language processing technology and sentiment recognition technology.

[0912] "Means for obtaining appropriate information from a database based on the natural language analysis results and sentiment analysis results" refers to means for obtaining relevant information from a database based on the analysis results.

[0913] "A method for generating optimal suggestions that take user emotions into consideration using generative artificial intelligence" refers to a method that uses generative artificial intelligence to generate optimal suggestions while taking into account the user's emotional state.

[0914] "Means for sending the generated proposal to the user's terminal" refers to communication means for sending the generated proposal to the user's terminal.

[0915] "Means for displaying the proposal to the user" refers to means for displaying the content of the proposal to the user on a terminal.

[0916] "Means by which the user selects an action based on the displayed suggestion" refers to means equipped with an interface for the user to select an action based on the displayed suggestion.

[0917] "Means for transmitting the user's actions to the server" refers to communication means for transmitting the actions selected by the user to the server.

[0918] "Means for receiving and analyzing user feedback and emotions on the server" refers to means for receiving user feedback and the emotions associated with it, and for analyzing them.

[0919] "Means for updating the generative artificial intelligence based on the aforementioned feedback and sentiment analysis results" refers to means for updating the model of the generative artificial intelligence based on feedback obtained from the user and the results of sentiment analysis.

[0920] "Special offers and discount proposals" refer to offers that include special discounts or promotions for users.

[0921] This invention is a customer support system that recognizes user emotions and makes emotion-based suggestions. The system consists of a user terminal, a server, a database, a generative artificial intelligence system, and an emotion engine.

[0922] System Configuration

[0923] 1. User terminal

[0924] Users access the system using smartphones or tablets.

[0925] Users utilize an interface that allows them to input requests and the emotions associated with those requests. Emotion recognition employs facial recognition and voice analysis technologies.

[0926] Specific software libraries such as OpenCV (face recognition) and TextBlob (speech analysis) are used.

[0927] 2. Server

[0928] Receives request and sentiment data sent from the user's terminal.

[0929] The data is analyzed using a natural language processing engine (e.g., Python's NLTK library) and an emotion engine (deep learning model).

[0930] Based on the analysis results, relevant information is retrieved from the database. Here, SQL queries are used to obtain the necessary information from the database.

[0931] Using generative artificial intelligence, the system generates optimal suggestions that take user emotions into consideration. These suggestions may include special discounts or cashback offers.

[0932] 3. Database

[0933] The database stores user requests and suggestions and provides information in response to queries from the server.

[0934] SQL and NoSQL databases may be used.

[0935] Explanation of the program's processing

[0936] The user uses their smartphone's camera and microphone to input their requests and associated emotions into the system. OpenCV is used for facial recognition, and TextBlob for speech analysis. This allows the system to recognize what emotions the user is experiencing (e.g., stress, frustration, satisfaction).

[0937] The server receives user requests and emotion data, analyzes the requests using a natural language processing engine, and analyzes the emotions using an emotion engine. Based on the analysis results, it retrieves appropriate information from the database and inputs it into the generative artificial intelligence. The generative AI generates optimal suggestions that take the user's emotions into account and sends these suggestions to the user's terminal in an appropriate data format such as JSON. The user reviews the suggestions on their terminal and selects an action. The server receives the user's selected action and also collects feedback information. It analyzes the feedback and the emotions associated with it, and updates the generative AI model.

[0938] Specific example

[0939] The following is an example of a prompt message used by the server to generate the best possible suggestions for the user using generative artificial intelligence:

[0940] User's emotion: Stress

[0941] Offer details: 10% special discount, 1000 yen cashback

[0942] Please create the best message to deliver to this user.

[0943] A concrete example would be a user who enters "I'm looking for the latest smartphone, but I'm worried because they're expensive," and the system determines that the user is experiencing stress. Based on the analysis, the server retrieves the latest smartphone models and price information from its database, and uses generative artificial intelligence to include special discounts and cashback offers in its suggestions. These suggestions are sent to the user's device, and the user, receiving the special offers, becomes more inclined to purchase and makes a selection. The system then collects feedback from the user and incorporates it into future suggestions.

[0944] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0945] Step 1:

[0946] The user inputs requests and emotions on their device. Using a smartphone or tablet, the user inputs their requests in text format and simultaneously inputs emotions using the camera and microphone. The input emotion data is then analyzed using OpenCV for facial recognition and TextBlob for speech analysis. This process yields request data and emotion data as input.

[0947] Step 2:

[0948] The device sends request and sentiment data to the server. The data sent from the device is in JSON format and is sent to the server as an HTTP request. The output obtained here is the request and sentiment data sent to the server.

[0949] Step 3:

[0950] The server analyzes the received request and sentiment data. First, the server uses a natural language processing engine (e.g., Python's NLTK library) to analyze the request, and then uses a sentiment engine (deep learning model) to analyze the sentiment data. The input consists of the received request and sentiment data, and the output is the analysis results.

[0951] Step 4:

[0952] The server retrieves information from the database based on the analysis results. Based on the analyzed requests and sentiments, it uses SQL queries to retrieve corresponding information from the database. For example, it executes a query such as "SELECT model, price FROM devices WHERE release_date > '2023-01-01'". The input for this step is the analysis results, and the output is the retrieved data.

[0953] Step 5:

[0954] The server uses generative artificial intelligence to generate optimal suggestions. Based on acquired data and sentiment analysis results, the generative AI generates the most suitable suggestions for the user. These suggestions may include special discounts or cashback offers. The output is the generated suggestions.

[0955] Step 6:

[0956] The server sends the generated proposal to the terminal. The generated proposal is formatted in JSON format and sent to the user's terminal as an HTTP response. The output is the proposal data sent to the terminal.

[0957] Step 7:

[0958] The terminal displays the suggestions to the user. The terminal displays the received suggestion data on the user's interface. The output is the displayed suggestion content.

[0959] Step 8:

[0960] The user selects an action based on the suggestions. The user reviews the displayed suggestions and chooses which action to take. This provides the user's selection as input.

[0961] Step 9:

[0962] The device sends user actions to the server. The user's selected action data is sent to the server in JSON format. The output is the action data sent to the server.

[0963] Step 10:

[0964] The server receives and analyzes user feedback and sentiment. It receives user feedback data and re-entered sentiment data and performs analysis. The analysis method is the same as in step 3. The input for this step is feedback and sentiment data, and the output is the analysis results.

[0965] Step 11:

[0966] The server updates the generative artificial intelligence based on feedback and sentiment analysis results. The generative AI model is updated based on the analyzed feedback and sentiment analysis results. This results in more personalized suggestions for future requests. The output is the updated generative AI model.

[0967] Step 12:

[0968] The server generates new proposals based on the updated generative artificial intelligence model. Using the updated model based on feedback, it prepares more optimal proposals for the next request. The inputs to this step are the updated model and the new request, and the output is the new proposal.

[0969] 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.

[0970] 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.

[0971] 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.

[0972] [Third Embodiment]

[0973] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0974] 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.

[0975] 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).

[0976] 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.

[0977] 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.

[0978] 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).

[0979] 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.

[0980] 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.

[0981] 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.

[0982] 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.

[0983] 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.

[0984] 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".

[0985] This invention is a customer support system that provides appropriate suggestions to meet the diverse needs of users, and is implemented as follows.

[0986] System Configuration

[0987] This system consists of a terminal for inputting user requests, a server that analyzes the requests and provides appropriate information, and a database. Furthermore, generative artificial intelligence processes the data between these components to generate optimal suggestions.

[0988] Program processing

[0989] Enter the user's request.

[0990] Terminal: The user enters their request through the terminal application. For example, the user might enter "I want the latest Android model, but I want to use it as cheaply as possible" into the terminal application.

[0991] Sending and receiving requests

[0992] Terminal: The entered request is sent from the terminal to the server. HTTP requests are used to send the request data to the server.

[0993] Server: The server receives this request data and proceeds to the next step.

[0994] Request Natural Language Processing

[0995] Server: The server analyzes the received request using a natural language processing engine to extract the user's intent. For example, it might understand that the user is "looking for the latest model at a low price."

[0996] Database matching

[0997] Server: Based on the analysis results, retrieves information on the latest Android models and pricing plans from the database. It uses SQL queries such as "SELECT model, price FROM android_devices WHERE release_date > '2023-01-01'".

[0998] AI-powered proposal generation

[0999] Server: Based on data acquired and analyzed using generative artificial intelligence, it generates optimal suggestions for the user. For example, it utilizes TensorFlow models to suggest the best model and pricing plan, taking into account price, performance, and the user's past usage history.

[1000] Format and send the proposal.

[1001] Server: Formats the generated proposals into an output format that is easy for the user to understand. Converts the formatted proposal content into JSON format and sends it to the terminal as an HTTP response.

[1002] Display of proposed content

[1003] Terminal: Displays suggestions received from the server to the user. Users can check things like "the latest Android models, their prices, and promotional information."

[1004] User selections and actions

[1005] User: Review the suggestions and choose the most suitable option. For example, the user might "select a specific Android model and proceed with the purchase."

[1006] Device: Triggers the following action based on the user's selection: redirect to a purchase link or initiate a change of price plan.

[1007] Feedback collection and analysis

[1008] User: Enter feedback on the proposal. Provide comments such as, "This plan does not provide enough data capacity."

[1009] Terminal: Sends feedback data to the server.

[1010] Server: Analyzes received feedback and uses it as data to improve generative artificial intelligence and the system. This is reflected in future suggestions, improving the user experience.

[1011] Specific example

[1012] Specifically, if a user enters "I want to transfer 50,000 images from iCloud to Android," the system will operate as follows:

[1013] 1. Natural Language Processing: The server extracts the request content, "Migrate iCloud images."

[1014] 2. Database matching: Retrieve information about data migration from iCloud to Android from the database.

[1015] 3. Proposal Generation: Using generative artificial intelligence, we propose the optimal data migration application and procedure.

[1016] 4. Display of proposed content: Display the proposed content on the user's device and provide specific instructions and links.

[1017] 5. Feedback Collection: After users complete the migration process, they will provide feedback about their experience.

[1018] This system allows users to solve problems quickly and accurately, leading to increased satisfaction.

[1019] The following describes the processing flow.

[1020] Step 1:

[1021] Terminal: The user launches the application on their terminal and enters their request. For example, they might enter, "I want the latest Android model, but I want to use it as cheaply as possible."

[1022] Step 2:

[1023] Terminal: Sends the entered request to the server in an appropriate data format such as JSON. Sends it to the server as an HTTP request.

[1024] Step 3:

[1025] Server: Passes the received request to the natural language processing engine for analysis. For example, it might use the Python NLTK library to extract the user's intent.

[1026] Step 4:

[1027] Server: Based on the results of natural language processing, it retrieves the latest Android model and pricing plan information from the database using SQL queries. For example, it executes a query such as "SELECT model, price FROM android_devices WHERE release_date > '2023-01-01'".

[1028] Step 5:

[1029] Server: Inputs acquired data into a generative artificial intelligence system to generate optimal suggestions for the user. For example, it uses a TensorFlow model to suggest models and plans considering price, performance, and the user's past usage history.

[1030] Step 6:

[1031] Server: Formats the generated proposal content and converts it into a JSON response. The generated proposal includes details such as "model name," "price," and "campaign information."

[1032] Step 7:

[1033] Server: Sends the formatted proposal content to the terminal as an HTTP response.

[1034] Step 8:

[1035] Device: Displays received suggestions to the user. For example, it displays detailed information and pricing for new Android models, as well as related promotional information.

[1036] Step 9:

[1037] User: Review the suggestions and choose the most suitable option. For example, select a specific Android model and proceed with the purchase.

[1038] Step 10:

[1039] Device: Based on the user's selection, the following actions are triggered. For example, redirecting to the purchase page for the selected Android model, or initiating the process of changing the price plan.

[1040] Step 11:

[1041] User: Enter feedback on the proposal. For example, provide comments such as, "This plan does not provide enough data capacity."

[1042] Step 12:

[1043] Terminal: Sends the input feedback data to the server.

[1044] Step 13:

[1045] Server: Analyzes received feedback and uses it to improve the generative artificial intelligence and the system. For example, it analyzes the content of the feedback and incorporates it into future suggestions.

[1046] (Example 1)

[1047] 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."

[1048] There is a need for customer support systems that can provide quick and appropriate solutions to the diverse needs of today's users. In particular, when user requests are complex, it is not easy to accurately understand those requests and provide appropriate information. Traditional systems often fail to adequately analyze user requests and can only provide insufficient solutions, leading to decreased user satisfaction.

[1049] 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.

[1050] In this invention, the server includes means for performing natural language analysis on user requests, means for obtaining appropriate information from a database based on the natural language analysis results, and means for generating optimal suggestions for the user using generative artificial intelligence. This makes it possible to accurately understand user requests and quickly provide optimal suggestions.

[1051] "Means for inputting user requests" refers to devices or software that provide an interface that allows users to input arbitrary requests or questions in text or voice.

[1052] "Means for sending the request to the server" refers to communication protocols, devices, and software for sending the input request to the server via the Internet.

[1053] "Means for performing natural language analysis on the server" refers to natural language processing technology and software used by the server to analyze the received request and understand the user's intent.

[1054] "Means for obtaining appropriate information from a database based on the results of natural language analysis" refers to a process and software for searching and obtaining appropriate data from a database based on the information obtained as a result of natural language analysis.

[1055] "Means of generating optimal suggestions for users using generative artificial intelligence" refers to the process and software that uses generative artificial intelligence (e.g., machine learning models) to generate optimal suggestions based on analyzed requests and acquired data.

[1056] "Means for sending the generated proposal to the user's terminal" refers to the communication protocol, device, and software that send the generated proposal from the server to the user's terminal.

[1057] "Means for displaying the proposal to the user" refers to an interface and software for visually displaying the received proposal content on the user's terminal.

[1058] "Means by which the user selects an action based on the displayed suggestion" refers to an input device and software for the user to make a selection or take action based on the suggested content.

[1059] "Means for transmitting the user's actions to the server" refers to communication protocols, devices, and software that transmit the user's selected actions and feedback to the server.

[1060] "Means for receiving and analyzing user feedback on the server" refers to the process and software for receiving feedback sent by the user and analyzing its contents.

[1061] "Means for updating the generative artificial intelligence based on the aforementioned feedback" refers to the process and software for improving and updating the algorithms and models of the generative artificial intelligence based on feedback obtained from users.

[1062] This invention is a customer support system that provides appropriate suggestions to meet diverse user requests, and is implemented using the following hardware and software. The system consists of a terminal for inputting user requests, a server that analyzes the requests and provides appropriate information, and a database. Furthermore, a generative artificial intelligence processes data between these components to generate optimal suggestions.

[1063] Hardware and software

[1064] Terminal: A device used by users to input requests, including smartphones, tablets, and personal computers. This terminal has applications installed that provide a user interface.

[1065] Server: A central computer that analyzes requests and provides appropriate information. The server has natural language processing engines (e.g., spaCy and NLTK), generative artificial intelligence (e.g., TensorFlow and PyTorch), and database management systems (e.g., MySQL and PostgreSQL) installed.

[1066] Database: A management system for storing information related to user requests, including the latest Android models and pricing plan information.

[1067] System operation

[1068] The user enters a request through an application on their device, such as "I want the latest Android model, but I want to use it as cheaply as possible." This input is sent to the server using an HTTP request. The server receives this request data and parses it using a natural language processing engine. Based on the analysis results, the server retrieves relevant information from its database. Using this retrieved information, a generative artificial intelligence generates the most suitable suggestion for the user.

[1069] The generated suggestions are formatted, converted to JSON format, and sent to the device as an HTTP response. The device then displays the suggestions to the user. Based on the suggested information, the user can choose an action and proceed with the next steps. For example, the user can select the most suitable Android model and proceed with the purchase.

[1070] Specific example

[1071] Specifically, if a user wants to know "the latest Android models, their prices, and promotional information," the system works as follows: The user enters "I want the latest Android model, but I want to use it as cheaply as possible" through an application on their device. This request is sent to the server, which performs natural language processing to understand the user's intent. Next, it retrieves relevant information from the database and generates optimal suggestions using generative artificial intelligence. These suggestions are sent to the device, and the user can review them.

[1072] Example of a prompt

[1073] "Please tell me how to buy the latest Android model cheaply."

[1074] "I want to transfer 50,000 images from iCloud to my Android device."

[1075] "Please suggest which device is best suited to our new pricing plan."

[1076] This system allows users to resolve problems quickly and accurately and receive highly satisfactory service.

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

[1078] Step 1:

[1079] User: The user enters their request through the application on their device. For example, they might enter, "I want the latest Android model, but I want to use it as cheaply as possible." The entered text is saved to the application's internal memory.

[1080] Input: User request written in natural language

[1081] Output: Request text stored in the terminal's internal memory

[1082] Example of operation: The user enters "I want the latest Android model, but I want to use it as cheaply as possible" into the application's text box and presses the submit button.

[1083] Step 2:

[1084] Terminal: Sends user-entered requests to the server. Specifically, request data is sent to the server as an HTTP POST request. The request body contains user input in JSON format.

[1085] Input: Request text

[1086] Output: HTTP request sent to the server

[1087] Example of operation: The device sends an HTTP POST request containing "request data" to the " / api / requests" endpoint.

[1088] Step 3:

[1089] Server: The server parses the received request data. First, it extracts the JSON data from the request body and then retrieves the request text from within it.

[1090] Input: HTTP Request

[1091] Output: Request text

[1092] Example of operation: The server extracts a request from the body of an HTTP request it receives, such as "I want the latest Android model, but I want to use it as cheaply as possible."

[1093] Step 4:

[1094] Server: The server analyzes the acquired text using a natural language processing engine (e.g., spaCy or NLTK) to extract the user's intent. Specifically, it analyzes keywords and phrases in the text to extract intents such as "latest model" or "buy cheaply."

[1095] Input: Request text

[1096] Output: Analyzed intent (e.g., keywords or phrases)

[1097] Specific example of operation: The server uses spaCy to analyze the text "I want the latest Android model, but I want to use it as cheaply as possible" and extracts the intents "latest model" and "buy cheaply".

[1098] Step 5:

[1099] Server: Based on the analysis results, retrieve relevant information from the database. Specifically, it generates SQL queries and sends requests to the database. For example, it executes an SQL query such as "SELECT model, price FROM android_devices WHERE release_date > '2023-01-01'".

[1100] Input: Analyzed intent

[1101] Output: Information retrieved from the database (e.g., the latest Android models and their prices)

[1102] Example of operation: The server executes the query "SELECT model, price FROM android_devices WHERE release_date > '2023-01-01'" on the MySQL database to retrieve the latest Android models and their prices.

[1103] Step 6:

[1104] Server: Generates optimal suggestions for the user based on information and analysis results obtained using generative artificial intelligence (e.g., TensorFlow or PyTorch). Suggestion generation considers both the user's requests and information obtained from the database.

[1105] Input: Information retrieved from the database, analyzed intent

[1106] Output: Generated proposals

[1107] A concrete example of operation: The server uses a TensorFlow model to generate suggestions such as, "The latest Android model X can be purchased for 30,000 yen."

[1108] Step 7:

[1109] Server: Formats the generated proposal, converts it to JSON format, and sends it to the user's terminal as an HTTP response.

[1110] Input: Generated proposal

[1111] Output: HTTP response (JSON format) sent to the terminal.

[1112] Specific example of operation: The server formats the proposed content into JSON format such as "Model X: 30,000 yen, Campaign Y" and sends it to the terminal via HTTP response.

[1113] Step 8:

[1114] Terminal: Displays the received suggestions to the user. Specifically, the application's UI analyzes the suggestions and displays them in a visually easy-to-understand format.

[1115] Input: HTTP response (JSON format)

[1116] Output: Displayed suggestions

[1117] Example of operation: The device parses the JSON data it receives and displays a message on the screen such as, "The latest Android model X is available for purchase for 30,000 yen."

[1118] Step 9:

[1119] User: Review the suggestions and choose the most suitable option. For example, the user selects a specific Android model and proceeds with the purchase.

[1120] Input: Displayed suggestion

[1121] Output: User selection (e.g., purchase decision)

[1122] Example of the operation: The user clicks the "Purchase button" and is redirected to a screen to proceed with the purchase process.

[1123] Step 10:

[1124] Device: Triggers the following action based on the user's selection: redirect to a purchase link or initiate a change of price plan.

[1125] Input: User Selection

[1126] Output: Next triggered action

[1127] Example of the operation: The user clicks the "Purchase button" and is redirected to a screen to proceed with the purchase process.

[1128] Step 11:

[1129] User: Enter feedback on the proposal. For example, provide comments such as, "This plan does not provide enough data capacity."

[1130] Input: Feedback on the proposed content

[1131] Output: Feedback entered into the terminal

[1132] Example of the operation: The user enters "This plan does not provide enough data capacity" into the feedback text box.

[1133] Step 12:

[1134] Terminal: Sends feedback data to the server. Specifically, it sends feedback data to the server as an HTTP POST request. The request body contains the feedback in JSON format.

[1135] Input: Feedback text

[1136] Output: HTTP request sent to the server

[1137] Specific example of operation: The terminal sends an HTTP POST request containing "feedback data" to " / api / feedback".

[1138] Step 13:

[1139] Server: Analyzes received feedback data and uses it to improve generative artificial intelligence and systems. This allows for the incorporation of feedback into future suggestions, thereby improving the user experience.

[1140] Input: HTTP request (feedback data)

[1141] Output: Updated generative AI models and system improvement data

[1142] Specific example of operation: The server analyzes the feedback data it receives and uses it to refine the proposed algorithm for the next time.

[1143] (Application Example 1)

[1144] 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."

[1145] In modern e-commerce sites, users often struggle to quickly find the right product from a vast selection. Furthermore, customer support is required to provide prompt and accurate responses to diverse user needs. However, conventional systems have been unable to effectively utilize natural language processing and generative artificial intelligence, making it difficult to improve the user experience. Solving this problem is the objective of this invention.

[1146] 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.

[1147] In this invention, the server includes means for analyzing intent using a natural language processing engine, means for retrieving products and FAQs from a database, and means for returning suggestions in JSON format. This makes it possible to accurately analyze user requests and provide optimal product suggestions and support information quickly and accurately.

[1148] A "user" is a person who uses a system to input requests.

[1149] "Requests" refer to all kinds of wishes and questions that users input into the system.

[1150] A "server" is a computer that receives and analyzes user requests, and generates and provides appropriate information in conjunction with a database.

[1151] "Natural language processing" is a technology that analyzes the requests received from users and understands their intentions.

[1152] "Generative artificial intelligence" refers to an algorithm or system that generates optimal suggestions based on analyzed information.

[1153] A "terminal" is an electronic device used by a user to input requests and display suggestions from a server.

[1154] A "database" is a storage device that stores information such as product information and FAQs.

[1155] "Feedback" refers to the opinions and evaluations that users provide regarding a proposal.

[1156] A "natural language processing engine" is software or an algorithm that analyzes user requests and understands their intent.

[1157] A "proposal" is information or a choice provided to the user, generated by a generative artificial intelligence system.

[1158] "Products and FAQs" refer to product information and answers to frequently asked questions stored in the database.

[1159] JSON format is a lightweight data exchange format for structuring, storing, and transmitting data.

[1160] This invention provides appropriate product suggestions and support information in response to user requests in a customer support system for an e-commerce site. The system analyzes the user's request using natural language processing, generates optimal suggestions using generative artificial intelligence, and provides them to the user. Specific embodiments are described below.

[1161] System Configuration

[1162] The system will be implemented with the following configuration.

[1163] Device: An electronic device (e.g., smartphone, tablet) where the user enters a request and receives a suggestion.

[1164] Server: A computer that analyzes user requests and generates appropriate information.

[1165] Database: A storage device that stores product information and FAQs.

[1166] Natural language processing engine: Software used to analyze user requests.

[1167] Generative artificial intelligence: An algorithm that generates optimal suggestions based on analysis results.

[1168] Program processing

[1169] Entering and submitting requests

[1170] The user enters the request using a terminal. The request is sent to the server using an HTTP request.

[1171] natural language analysis

[1172] The server analyzes the received request using a natural language processing engine. Specifically, it uses the Hugging Face Transformers library to understand the user's intent.

[1173] Retrieving information from the database

[1174] Based on the analysis results, the server retrieves product information and FAQs from the database using SQL queries. For example, it might execute a query like "SELECT name, price, description FROM products WHERE description LIKE '%Summer Cooling Mattress%'".

[1175] Proposal generation

[1176] The server uses generative artificial intelligence (e.g., TensorFlow models) to generate optimal suggestions based on the acquired data. The suggestions are formatted in JSON format and sent to the terminal as an HTTP response.

[1177] Display of proposed content

[1178] The user's device displays the received suggestions. This allows the user to view specific product information and answers to frequently asked questions.

[1179] Feedback collection and analysis

[1180] Users input feedback on the suggestions and send it from their device to the server. The server analyzes this feedback and updates the generative artificial intelligence to improve the accuracy of future suggestions.

[1181] Specific example

[1182] If a user enters "I want a cooling mattress for summer," the system will operate as follows:

[1183] 1. The server uses a natural language processing engine to analyze the request, "cooling mattress for summer."

[1184] 2. Retrieve the relevant product information from the database.

[1185] 3. Generate optimal product suggestions using generative artificial intelligence.

[1186] 4. Format the generated proposals in JSON format and send them to the terminal.

[1187] 5. The user reviews the suggestion and chooses an action.

[1188] Example of a prompt

[1189] User: I want a cooling mattress for summer.

[1190] System: We found several mattresses made with cooling materials for summer. The following products are recommended.

[1191] Product name: "Summer Sleep Mattress"

[1192] Price: 5,000 yen

[1193] Description: This mattress uses cooling material for a comfortable sleeping experience.

[1194] In this way, we can provide quick and accurate solutions to meet user requests.

[1195] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1196] Step 1:

[1197] The user enters a request into the terminal and sends it. The user specifically enters, "I want a cooling mattress for summer." The input data is sent to the server as an HTTP request.

[1198] Step 2:

[1199] The server analyzes the received request using a natural language processing engine (e.g., Hugging Face's Transformers library). It extracts key keywords and intent from the request data and recognizes the intent as "cooling mattress for summer." This analysis result is then used for subsequent database matching.

[1200] Step 3:

[1201] Based on the analysis results, the server retrieves appropriate product information from the database. Specifically, it executes an SQL query and sends a query like "SELECT name, price, description FROM products WHERE description LIKE '%Summer Cooling Mattress%'". The database returns the corresponding product information.

[1202] Step 4:

[1203] The server processes the acquired data using generative artificial intelligence (e.g., TensorFlow models) to create optimal recommendations. It selects the best product for the user, taking into account factors such as price, performance, and the user's past search history. This information will later be returned to the user as recommendations.

[1204] Step 5:

[1205] The server formats the generated proposal in JSON format and sends it to the terminal as an HTTP response. The formatted proposal includes product name, price, and description.

[1206] Step 6:

[1207] The terminal displays the received suggestions to the user. The user can check specific product information and suggestions. Information such as "Summer Sleep Mattress" and "Comfortable Mattress with Cooling Material" will be displayed.

[1208] Step 7:

[1209] The user reviews the displayed suggestions and chooses an action, such as clicking a purchase button.

[1210] Step 8:

[1211] The device sends user actions to the server. The action data arrives at the server as an HTTP request.

[1212] Step 9:

[1213] The server receives and analyzes user feedback. It updates the generative artificial intelligence based on the feedback data and incorporates it into future suggestions. For example, if feedback states, "This mattress is comfortable, but it's expensive," the system will improve its suggestions to prioritize price.

[1214] 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.

[1215] This invention is a customer support system that provides more personalized suggestions by incorporating an emotion engine that recognizes the user's emotions, and is implemented as follows.

[1216] System Configuration

[1217] This system consists of a terminal for inputting user requests and emotions, a server that analyzes requests and emotions and provides appropriate information, a database, and generative artificial intelligence. Furthermore, an emotion engine works in conjunction with these components to provide responses based on the user's emotions.

[1218] Program processing

[1219] Enter the user's requests and feelings.

[1220] Terminal: When a user launches an application on their terminal and enters a request, their emotions are also recognized. For example, if a user enters "I want the latest Android model, but I want to use it as cheaply as possible," facial recognition and voice analysis will determine that the user is feeling dissatisfied.

[1221] Sending requests and emotions

[1222] Terminal: Sends the entered request and sentiment data to the server in an appropriate data format such as JSON. Sends it to the server as an HTTP request.

[1223] Receiving and analyzing requests and emotions

[1224] Server: Passes received requests and sentiment data to the natural language processing engine and sentiment engine for analysis. For example, it uses the Python NLTK library to extract the user's intent and the sentiment engine to recognize the user's emotion as "dissatisfaction."

[1225] Database matching

[1226] Server: Based on the analysis results, retrieves the latest Android model and pricing plan information from the database using SQL queries. For example, it executes a query such as "SELECT model, price FROM android_devices WHERE release_date > '2023-01-01'".

[1227] AI-powered proposal generation

[1228] Server: Inputs acquired data and sentiment analysis results into a generative artificial intelligence system to generate optimal suggestions that take user emotions into consideration. For example, to address dissatisfaction, it might include special discounts or additional benefits in its suggestions.

[1229] Format and send the proposal.

[1230] Server: Formats the generated suggestions, adds a message tailored to the user's sentiment, and converts it into a JSON response. The generated suggestions include details such as "model name," "price," and "campaign information."

[1231] Display of proposed content

[1232] Device: Displays received suggestions to the user. For example, it displays detailed information and pricing for new Android models, related promotional information, and special offers to address user complaints.

[1233] User selections and actions

[1234] User: Review the suggestions and choose the most suitable option. For example, select a specific Android model and proceed with the purchase.

[1235] Device: Based on the user's selection, the following actions are triggered. For example, redirecting to the purchase page for the selected Android model, or initiating the process of changing the price plan.

[1236] Feedback collection and analysis

[1237] User: Enter feedback on the proposal. In addition to opinions such as "This plan does not have enough data capacity," also enter your feelings about the proposal.

[1238] Terminal: Acquires feedback data, analyzes it with an emotion engine, and sends it to the server.

[1239] Feedback analysis and AI updates

[1240] Server: Analyzes received feedback and sentiment data to update the generative artificial intelligence and sentiment engine models. This results in more personalized suggestions for the next session, improving user satisfaction.

[1241] Specific example

[1242] Specifically, this occurs when the emotion engine determines that a user is stressed after typing, "I want to transfer 50,000 images from iCloud to my Android device."

[1243] 1. Natural Language Processing and Sentiment Recognition: The server extracts the request "migrate iCloud images" and the emotion "stress".

[1244] 2. Database matching: Retrieve information about data migration from iCloud to Android from the database.

[1245] 3. Suggestion Generation: Generative artificial intelligence is used to suggest the optimal data migration app and procedure. Free support offers are also included to reduce user stress.

[1246] 4. Display of proposals: Display the proposals on the user's device and include information about special support.

[1247] 5. Feedback Collection and Analysis: After users complete the migration process, they input their emotions along with feedback about their experience. The server analyzes this data to help improve future suggestions.

[1248] This system provides more personalized support that takes user emotions into account, leading to increased satisfaction.

[1249] The following describes the processing flow.

[1250] Step 1:

[1251] Terminal: The user launches the application on their terminal and enters their request. For example, they might enter text such as, "I want the latest Android model, but I want to use it as cheaply as possible."

[1252] Step 2:

[1253] Terminal: While the user is entering a request, the terminal's camera and microphone are used to analyze the user's facial expressions and voice tone in real time using an emotion engine. For example, facial recognition technology can detect emotions such as "dissatisfaction" or "excitement."

[1254] Step 3:

[1255] Terminal: Sends the entered request data and sentiment data to the server in an appropriate data format such as JSON. Data is sent using HTTP requests.

[1256] Step 4:

[1257] Server: Analyzes received request data and sentiment data. The natural language processing engine analyzes the text request, and the sentiment engine analyzes the sentiment data. For example, it extracts the request "I'm looking for the latest model at a low price" and the sentiment "dissatisfied."

[1258] Step 5:

[1259] Server: Based on the request analysis results, retrieve appropriate information from the database. Use SQL queries to retrieve information on the latest Android models and pricing plans. For example, execute the query "SELECT model, price FROM android_devices WHERE release_date > '2023-01-01'".

[1260] Step 6:

[1261] Server: Based on the acquired information, a generative artificial intelligence generates optimal suggestions. Furthermore, it adjusts the suggestions considering the sentiment analysis results. For example, if the user has the emotion of "dissatisfaction," it will include special discounts or additional benefits in its suggestions.

[1262] Step 7:

[1263] Server: Formats the generated suggestions, adds sentiment-based messages, and converts them into a JSON response. The generated suggestions include details such as "model name," "price," and "campaign information."

[1264] Step 8:

[1265] Server: Sends the formatted proposal content to the terminal as an HTTP response.

[1266] Step 9:

[1267] Device: Displays received suggestions to the user. For example, it displays detailed information and pricing for new Android models, related promotional information, and special offers to address user complaints.

[1268] Step 10:

[1269] User: Review the suggestions and choose the most suitable option. For example, select a specific Android model and proceed with the purchase.

[1270] Step 11:

[1271] Device: Based on the user's selection, trigger the following actions. For example, redirect to the purchase page for the selected Android model, or initiate the process of changing the price plan.

[1272] Step 12:

[1273] User: Enter feedback on the proposed plan. This includes comments such as "This plan does not provide sufficient data capacity," as well as your feelings at the time of giving feedback.

[1274] Step 13:

[1275] Terminal: Acquires feedback data and sentiment data and sends it to the server.

[1276] Step 14:

[1277] Server: Analyzes received feedback and emotion data to update the generative artificial intelligence and emotion engine models. This results in more personalized suggestions for the next session, improving user satisfaction.

[1278] This system provides personalized suggestions that take user emotions into consideration, thereby improving user satisfaction.

[1279] (Example 2)

[1280] 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."

[1281] Traditional customer support systems, while providing appropriate information in response to user requests, failed to consider user emotions, resulting in lower user satisfaction. Furthermore, they lacked systems for effectively utilizing user feedback and improving suggestions. This made it difficult to provide optimal service to users.

[1282] 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.

[1283] In this invention, the server includes means for inputting user requests and sentiment data; means for transmitting the requests and sentiment data to the server; means for performing natural language processing and sentiment analysis on the requests and sentiment data on the server; means for obtaining appropriate information from a database based on the natural language processing results and sentiment analysis results; means for generating optimal suggestions for the user using generative artificial intelligence; means for transmitting the generated suggestions to the user's terminal; means for displaying the suggestions to the user; means for the user to select an action based on the displayed suggestions; means for transmitting the user's actions to the server; means for receiving and analyzing user feedback on the server; and means for updating the generative artificial intelligence and sentiment analysis engine based on the feedback. This enables the provision of personalized suggestions that take user sentiment into consideration, thereby improving user satisfaction. Furthermore, it enables continuous improvement of the system based on feedback.

[1284] "Means for inputting user requests and emotional data" refers to functions and devices that allow users to input their desired requests and their emotional state at the time using a terminal.

[1285] "Means for sending request and sentiment data to the server" refers to a communication function for sending request and sentiment data entered on the terminal to the server.

[1286] "Means for natural language processing and sentiment analysis of request and sentiment data" refers to a function that analyzes request data received on a server using machine learning models, etc., and analyzes sentiment data using sentiment recognition engines, etc.

[1287] "Means for obtaining appropriate information" refers to the function of retrieving necessary information from a database based on the results of natural language processing and sentiment analysis.

[1288] "A means of generating optimal suggestions for users using generative artificial intelligence" refers to a function that uses a generative artificial intelligence model to create optimal suggestions for users based on acquired information and sentiment data.

[1289] "Means for sending generated suggestions to the user's terminal" refers to a communication function for sending suggestions created by generative artificial intelligence to the user's terminal.

[1290] "Means of displaying suggestions to the user" refers to a function that displays the content of suggestions received on the terminal to the user visually or audibly.

[1291] "Means for users to choose their actions" refers to interface functions that allow users to make choices in response to displayed suggestions.

[1292] "Means for sending user actions to the server" refers to a communication function for sending user-selected action data to the server.

[1293] "Means for receiving and analyzing feedback" refers to a function that receives user feedback data on a server and analyzes that data.

[1294] "Means for updating generative artificial intelligence and sentiment analysis engines" refers to functions for continuously improving generative artificial intelligence models and sentiment analysis engines based on analyzed feedback data.

[1295] This invention relates to a customer support system that provides more personalized suggestions by incorporating an emotion engine that recognizes user emotions. The system consists of a terminal for inputting user requests and emotions, a server that analyzes requests and emotions and provides appropriate information, a database, and generative artificial intelligence.

[1296] Hardware and software configuration

[1297] terminal

[1298] A terminal is a device used by users to input requests and emotional data, and includes smartphones, tablets, and PCs. The terminal's application uses the camera and microphone to acquire user input and perform facial recognition and voice analysis. A common facial recognition API (e.g., Google Cloud Vision API) is used for facial recognition, and a common voice analysis API (e.g., Amazon Transcribe) is used for voice analysis.

[1299] server

[1300] The server analyzes received requests and sentiment data and processes them to generate appropriate suggestions. The server uses Python's Natural Language Toolkit (NLTK) and spaCy for its natural language processing engine, and machine learning models for its sentiment analysis engine. A database (e.g., an SQL database) is also integrated to retrieve relevant information based on the analysis results and sentiment data. For generative artificial intelligence, advanced generative models such as GPT-3 are used.

[1301] Specific processing of the program

[1302] The program of this system performs the following specific processing.

[1303] Request and emotional input

[1304] The user launches an application on their device and enters a request, while simultaneously capturing emotional data through the camera and microphone. For example, if the user enters "I want the latest Android model, but I want to use it as cheaply as possible," the device's camera will capture a dissatisfied expression.

[1305] Sending requests and emotions

[1306] The device converts the acquired request and sentiment data into JSON format and sends it to the server as an HTTP POST request.

[1307] Receiving and analyzing requests and emotions

[1308] The server analyzes received requests and sentiment data using a natural language processing engine and a sentiment analysis engine. The natural language processing engine extracts the user's requests, and the sentiment analysis engine recognizes the user's emotions.

[1309] Database matching

[1310] Based on the analysis results, the server uses SQL queries to retrieve appropriate information from the database. For example, it might execute a query like "SELECT model, price FROM android_devices WHERE release_date > '2023-01-01'" to retrieve the latest Android model information.

[1311] AI-powered proposal generation

[1312] The server inputs prompts into a generative artificial intelligence model based on the acquired information and sentiment analysis results, generating the most suitable suggestions for the user. For example, it uses the following prompts:

[1313] "This user wants the latest Android device, but is dissatisfied with the lack of a suitable solution to their needs, as they want to use it as cheaply as possible. Please provide them with the best possible offer."

[1314] Submitting and displaying proposals

[1315] The generated suggestions are reformatted and sent to the user's device as a JSON response. The user can then visually or audibly review the suggestions on their device. For example, a suggestion such as "Latest Android model: Android XYZ, Price: 50,000 JPY, Special discount 10%" might be displayed.

[1316] Feedback collection and analysis

[1317] The user enters feedback on the suggestion into their device and sends it back to the server. The server analyzes the feedback and sentiment data, and uses the results to update the generative artificial intelligence model and sentiment analysis engine, making the next suggestion more accurate.

[1318] Specific example

[1319] For example, if a user enters "I want to transfer 50,000 images from iCloud to Android" and the sentiment analysis engine determines that the user is feeling stressed:

[1320] 1. The server extracts the request "migrate iCloud images" and the emotion "stress".

[1321] 2. Retrieve information about data migration from iCloud to Android from the database.

[1322] 3. Use generative artificial intelligence to suggest the optimal data migration app and procedure, and include a free support offer.

[1323] 4. Display the proposed content on the user's device and include information about special support.

[1324] 5. After the user completes the migration process, they input their emotions along with feedback about their experience. The server analyzes this data to help improve future suggestions.

[1325] This system provides more personalized support that takes user emotions into account, leading to increased satisfaction.

[1326] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1327] Step 1:

[1328] User: Launch the application on the device and enter a request in the text field. For example, enter a request such as, "I want the latest Android model, but I want to use it as cheaply as possible." Generate input data (request text).

[1329] Terminal: Acquires input data and generates emotion data using face recognition APIs and voice analysis APIs with the camera and microphone. For example, it can use the Google Cloud Vision API to recognize an unhappy expression and obtain emotion data.

[1330] Input: User's requested text, user's facial image, or voice data

[1331] Output: Request text, dissatisfaction data

[1332] Step 2:

[1333] Terminal: Converts the acquired request and sentiment data into JSON format and sends it to the server using an HTTP POST request.

[1334] Input: Request text, dissatisfaction data

[1335] Output: HTTP POST request (JSON format)

[1336] Step 3:

[1337] Server: Receives HTTP requests and extracts requests and sentiments from JSON data. The extracted data is then passed to a natural language processing engine (e.g., NLTK or spaCy) and a sentiment analysis engine for analysis.

[1338] Input: HTTP POST request (JSON format)

[1339] Output: Request data as analysis results (e.g., "latest Android model"), sentiment data ("dissatisfied")

[1340] Step 4:

[1341] Server: Based on the analysis results, it executes an SQL query to retrieve the necessary information from the database. For example, to retrieve "information about the latest Android models," it executes "SELECT model, price FROM android_devices WHERE release_date > '2023-01-01'".

[1342] Input: Requested data for analysis results ("latest Android model")

[1343] Output: Database query results (e.g., model name and price of a specific device)

[1344] Step 5:

[1345] Server: Based on the acquired information, it inputs prompts into a generative artificial intelligence (e.g., GPT-3) to generate the best possible suggestions for the user. For example, it uses the following prompts:

[1346] "This user wants the latest Android device, but is dissatisfied with the lack of a suitable solution to their needs, as they want to use it as cheaply as possible. Please provide them with the best possible offer."

[1347] Input: Database query results, analyzed sentiment data

[1348] Output: Generated proposal details (e.g., model name, price, special discount information)

[1349] Step 6:

[1350] Server: Converts the generated proposal into a JSON response and sends it to the terminal as an HTTP response.

[1351] Input: Generated proposal content

[1352] Output: HTTP response (JSON format)

[1353] Step 7:

[1354] Terminal: Displays suggestions received from the server to the user. For example, it displays detailed information and pricing for the latest Android models, as well as related promotional information and special offers.

[1355] Input: HTTP response (JSON format)

[1356] Output: Suggestions displayed to the user

[1357] Step 8:

[1358] User: Review the suggestions and choose the best option. For example, select a recommended Android model and proceed with the purchase.

[1359] Input: Suggestions displayed to the user

[1360] Output: User's selected action

[1361] Step 9:

[1362] Device: Based on the user's selection, the following actions are triggered. For example, redirecting to the purchase page for the selected Android model, or initiating the process of changing the price plan.

[1363] Input: User's selected action

[1364] Output: Triggers the following action (e.g., redirects to the purchase page)

[1365] Step 10:

[1366] User: Enter feedback on the proposal. For example, enter an opinion such as "This plan does not provide enough data capacity" and a feeling of "dissatisfaction".

[1367] Input: User feedback text, facial images, and audio data

[1368] Output: Feedback data and sentiment data

[1369] Step 11:

[1370] Terminal: Acquires feedback data, analyzes it with an emotion engine, and sends it to the server as an HTTP POST request.

[1371] Input: Feedback data, sentiment data

[1372] Output: HTTP POST request (feedback data and sentiment data)

[1373] Step 12:

[1374] Server: Analyzes received feedback and sentiment data to update the generative artificial intelligence and sentiment analysis engine models. This allows for more personalized suggestions and improved user satisfaction.

[1375] Input: HTTP POST request (feedback data and sentiment data)

[1376] Output: Updated generative artificial intelligence model and sentiment analysis engine

[1377] (Application Example 2)

[1378] 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."

[1379] Traditional customer support systems have failed to adequately consider not only user requests but also their emotions. As a result, users often do not receive suggestions tailored to their feelings, leading to decreased satisfaction. Furthermore, insufficient personalization that takes user emotions into account makes it difficult to offer special offers or discounts that will stimulate purchase intent. Solving these problems is essential.

[1380] 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.

[1381] In this invention, the server includes means for inputting requests and emotions from the user, means for transmitting the requests and emotions to the server, means for performing natural language analysis and sentiment analysis on the requests and emotions on the server, means for obtaining appropriate information from a database based on the natural language analysis results and sentiment analysis results, means for generating optimal suggestions that take the user's emotions into consideration using generative artificial intelligence, means for transmitting the generated suggestions to the user's terminal, means for displaying the suggestions to the user, means for the user to select an action based on the displayed suggestions, means for transmitting the user's actions to the server, means for receiving and analyzing feedback and emotions from the user on the server, means for updating the generative artificial intelligence based on the feedback and sentiment analysis results, and includes special offers and discount suggestions. This enables personalized suggestions that take the user's emotions into consideration, improving user satisfaction and stimulating purchase intent.

[1382] A "means for inputting user requests and emotions" refers to a means that provides an interface that allows users to input their requests along with their emotions at that time.

[1383] "Means for sending the aforementioned requests and emotions to the server" refers to communication means for sending requests and emotions entered by the user to the server.

[1384] "Means for performing natural language processing and sentiment analysis on the server" refers to means for analyzing requests and sentiments on the server using natural language processing technology and sentiment recognition technology.

[1385] "Means for obtaining appropriate information from a database based on the natural language analysis results and sentiment analysis results" refers to means for obtaining relevant information from a database based on the analysis results.

[1386] "A method for generating optimal suggestions that take user emotions into consideration using generative artificial intelligence" refers to a method that uses generative artificial intelligence to generate optimal suggestions while taking into account the user's emotional state.

[1387] "Means for sending the generated proposal to the user's terminal" refers to communication means for sending the generated proposal to the user's terminal.

[1388] "Means for displaying the proposal to the user" refers to means for displaying the content of the proposal to the user on a terminal.

[1389] "Means by which the user selects an action based on the displayed suggestion" refers to means equipped with an interface for the user to select an action based on the displayed suggestion.

[1390] "Means for transmitting the user's actions to the server" refers to communication means for transmitting the actions selected by the user to the server.

[1391] "Means for receiving and analyzing user feedback and emotions on the server" refers to means for receiving user feedback and the emotions associated with it, and for analyzing them.

[1392] "Means for updating the generative artificial intelligence based on the aforementioned feedback and sentiment analysis results" refers to means for updating the model of the generative artificial intelligence based on feedback obtained from the user and the results of sentiment analysis.

[1393] "Special offers and discount proposals" refer to offers that include special discounts or promotions for users.

[1394] This invention is a customer support system that recognizes user emotions and makes emotion-based suggestions. The system consists of a user terminal, a server, a database, a generative artificial intelligence system, and an emotion engine.

[1395] System Configuration

[1396] 1. User terminal

[1397] Users access the system using smartphones or tablets.

[1398] Users utilize an interface that allows them to input requests and the emotions associated with those requests. Emotion recognition employs facial recognition and voice analysis technologies.

[1399] Specific software libraries such as OpenCV (face recognition) and TextBlob (speech analysis) are used.

[1400] 2. Server

[1401] Receives request and sentiment data sent from the user's terminal.

[1402] The data is analyzed using a natural language processing engine (e.g., Python's NLTK library) and an emotion engine (deep learning model).

[1403] Based on the analysis results, relevant information is retrieved from the database. Here, SQL queries are used to obtain the necessary information from the database.

[1404] Using generative artificial intelligence, the system generates optimal suggestions that take user emotions into consideration. These suggestions may include special discounts or cashback offers.

[1405] 3. Database

[1406] The database stores user requests and suggestions and provides information in response to queries from the server.

[1407] SQL and NoSQL databases may be used.

[1408] Explanation of the program's processing

[1409] The user uses their smartphone's camera and microphone to input their requests and associated emotions into the system. OpenCV is used for facial recognition, and TextBlob for speech analysis. This allows the system to recognize what emotions the user is experiencing (e.g., stress, frustration, satisfaction).

[1410] The server receives user requests and emotion data, analyzes the requests using a natural language processing engine, and analyzes the emotions using an emotion engine. Based on the analysis results, it retrieves appropriate information from the database and inputs it into the generative artificial intelligence. The generative AI generates optimal suggestions that take the user's emotions into account and sends these suggestions to the user's terminal in an appropriate data format such as JSON. The user reviews the suggestions on their terminal and selects an action. The server receives the user's selected action and also collects feedback information. It analyzes the feedback and the emotions associated with it, and updates the generative AI model.

[1411] Specific example

[1412] The following is an example of a prompt message used by the server to generate the best possible suggestions for the user using generative artificial intelligence:

[1413] User's emotion: Stress

[1414] Offer details: 10% special discount, 1000 yen cashback

[1415] Please create the best message to deliver to this user.

[1416] A concrete example would be a user who enters "I'm looking for the latest smartphone, but I'm worried because they're expensive," and the system determines that the user is experiencing stress. Based on the analysis, the server retrieves the latest smartphone models and price information from its database, and uses generative artificial intelligence to include special discounts and cashback offers in its suggestions. These suggestions are sent to the user's device, and the user, receiving the special offers, becomes more inclined to purchase and makes a selection. The system then collects feedback from the user and incorporates it into future suggestions.

[1417] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1418] Step 1:

[1419] The user inputs requests and emotions on their device. Using a smartphone or tablet, the user inputs their requests in text format and simultaneously inputs emotions using the camera and microphone. The input emotion data is then analyzed using OpenCV for facial recognition and TextBlob for speech analysis. This process yields request data and emotion data as input.

[1420] Step 2:

[1421] The device sends request and sentiment data to the server. The data sent from the device is in JSON format and is sent to the server as an HTTP request. The output obtained here is the request and sentiment data sent to the server.

[1422] Step 3:

[1423] The server analyzes the received request and sentiment data. First, the server uses a natural language processing engine (e.g., Python's NLTK library) to analyze the request, and then uses a sentiment engine (deep learning model) to analyze the sentiment data. The input consists of the received request and sentiment data, and the output is the analysis results.

[1424] Step 4:

[1425] The server retrieves information from the database based on the analysis results. Based on the analyzed requests and sentiments, it uses SQL queries to retrieve corresponding information from the database. For example, it executes a query such as "SELECT model, price FROM devices WHERE release_date > '2023-01-01'". The input for this step is the analysis results, and the output is the retrieved data.

[1426] Step 5:

[1427] The server uses generative artificial intelligence to generate optimal suggestions. Based on acquired data and sentiment analysis results, the generative AI generates the most suitable suggestions for the user. These suggestions may include special discounts or cashback offers. The output is the generated suggestions.

[1428] Step 6:

[1429] The server sends the generated proposal to the terminal. The generated proposal is formatted in JSON format and sent to the user's terminal as an HTTP response. The output is the proposal data sent to the terminal.

[1430] Step 7:

[1431] The terminal displays the suggestions to the user. The terminal displays the received suggestion data on the user's interface. The output is the displayed suggestion content.

[1432] Step 8:

[1433] The user selects an action based on the suggestions. The user reviews the displayed suggestions and chooses which action to take. This provides the user's selection as input.

[1434] Step 9:

[1435] The device sends user actions to the server. The user's selected action data is sent to the server in JSON format. The output is the action data sent to the server.

[1436] Step 10:

[1437] The server receives and analyzes user feedback and sentiment. It receives user feedback data and re-entered sentiment data and performs analysis. The analysis method is the same as in step 3. The input for this step is feedback and sentiment data, and the output is the analysis results.

[1438] Step 11:

[1439] The server updates the generative artificial intelligence based on feedback and sentiment analysis results. The generative AI model is updated based on the analyzed feedback and sentiment analysis results. This results in more personalized suggestions for future requests. The output is the updated generative AI model.

[1440] Step 12:

[1441] The server generates new proposals based on the updated generative artificial intelligence model. Using the updated model based on feedback, it prepares more optimal proposals for the next request. The inputs to this step are the updated model and the new request, and the output is the new proposal.

[1442] 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.

[1443] 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.

[1444] 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.

[1445] [Fourth Embodiment]

[1446] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1447] 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.

[1448] 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).

[1449] 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.

[1450] 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.

[1451] 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).

[1452] 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.

[1453] 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.

[1454] 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.

[1455] 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.

[1456] 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.

[1457] 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.

[1458] 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".

[1459] This invention is a customer support system that provides appropriate suggestions to meet the diverse needs of users, and is implemented as follows.

[1460] System Configuration

[1461] This system consists of a terminal for inputting user requests, a server that analyzes the requests and provides appropriate information, and a database. Furthermore, generative artificial intelligence processes the data between these components to generate optimal suggestions.

[1462] Program processing

[1463] Enter the user's request.

[1464] Terminal: The user enters their request through the terminal application. For example, the user might enter "I want the latest Android model, but I want to use it as cheaply as possible" into the terminal application.

[1465] Sending and receiving requests

[1466] Terminal: The entered request is sent from the terminal to the server. HTTP requests are used to send the request data to the server.

[1467] Server: The server receives this request data and proceeds to the next step.

[1468] Request Natural Language Processing

[1469] Server: The server analyzes the received request using a natural language processing engine to extract the user's intent. For example, it might understand that the user is "looking for the latest model at a low price."

[1470] Database matching

[1471] Server: Based on the analysis results, retrieves information on the latest Android models and pricing plans from the database. It uses SQL queries such as "SELECT model, price FROM android_devices WHERE release_date > '2023-01-01'".

[1472] AI-powered proposal generation

[1473] Server: Based on data acquired and analyzed using generative artificial intelligence, it generates optimal suggestions for the user. For example, it utilizes TensorFlow models to suggest the best model and pricing plan, taking into account price, performance, and the user's past usage history.

[1474] Format and send the proposal.

[1475] Server: Formats the generated proposals into an output format that is easy for the user to understand. Converts the formatted proposal content into JSON format and sends it to the terminal as an HTTP response.

[1476] Display of proposed content

[1477] Terminal: Displays suggestions received from the server to the user. Users can check things like "the latest Android models, their prices, and promotional information."

[1478] User selections and actions

[1479] User: Review the suggestions and choose the most suitable option. For example, the user might "select a specific Android model and proceed with the purchase."

[1480] Device: Triggers the following action based on the user's selection: redirect to a purchase link or initiate a change of price plan.

[1481] Feedback collection and analysis

[1482] User: Enter feedback on the proposal. Provide comments such as, "This plan does not provide enough data capacity."

[1483] Terminal: Sends feedback data to the server.

[1484] Server: Analyzes received feedback and uses it as data to improve generative artificial intelligence and the system. This is reflected in future suggestions, improving the user experience.

[1485] Specific example

[1486] Specifically, if a user enters "I want to transfer 50,000 images from iCloud to Android," the system will operate as follows:

[1487] 1. Natural Language Processing: The server extracts the request content, "Migrate iCloud images."

[1488] 2. Database matching: Retrieve information about data migration from iCloud to Android from the database.

[1489] 3. Proposal Generation: Using generative artificial intelligence, we propose the optimal data migration application and procedure.

[1490] 4. Display of proposed content: Display the proposed content on the user's device and provide specific instructions and links.

[1491] 5. Feedback Collection: After users complete the migration process, they will provide feedback about their experience.

[1492] This system allows users to solve problems quickly and accurately, leading to increased satisfaction.

[1493] The following describes the processing flow.

[1494] Step 1:

[1495] Terminal: The user launches the application on their terminal and enters their request. For example, they might enter, "I want the latest Android model, but I want to use it as cheaply as possible."

[1496] Step 2:

[1497] Terminal: Sends the entered request to the server in an appropriate data format such as JSON. Sends it to the server as an HTTP request.

[1498] Step 3:

[1499] Server: Passes the received request to the natural language processing engine for analysis. For example, it might use the Python NLTK library to extract the user's intent.

[1500] Step 4:

[1501] Server: Based on the results of natural language processing, it retrieves the latest Android model and pricing plan information from the database using SQL queries. For example, it executes a query such as "SELECT model, price FROM android_devices WHERE release_date > '2023-01-01'".

[1502] Step 5:

[1503] Server: Inputs acquired data into a generative artificial intelligence system to generate optimal suggestions for the user. For example, it uses a TensorFlow model to suggest models and plans considering price, performance, and the user's past usage history.

[1504] Step 6:

[1505] Server: Formats the generated proposal content and converts it into a JSON response. The generated proposal includes details such as "model name," "price," and "campaign information."

[1506] Step 7:

[1507] Server: Sends the formatted proposal content to the terminal as an HTTP response.

[1508] Step 8:

[1509] Device: Displays received suggestions to the user. For example, it displays detailed information and pricing for new Android models, as well as related promotional information.

[1510] Step 9:

[1511] User: Review the suggestions and choose the most suitable option. For example, select a specific Android model and proceed with the purchase.

[1512] Step 10:

[1513] Device: Based on the user's selection, the following actions are triggered. For example, redirecting to the purchase page for the selected Android model, or initiating the process of changing the price plan.

[1514] Step 11:

[1515] User: Enter feedback on the proposal. For example, provide comments such as, "This plan does not provide enough data capacity."

[1516] Step 12:

[1517] Terminal: Sends the input feedback data to the server.

[1518] Step 13:

[1519] Server: Analyzes received feedback and uses it to improve the generative artificial intelligence and the system. For example, it analyzes the content of the feedback and incorporates it into future suggestions.

[1520] (Example 1)

[1521] 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".

[1522] There is a need for customer support systems that can provide quick and appropriate solutions to the diverse needs of today's users. In particular, when user requests are complex, it is not easy to accurately understand those requests and provide appropriate information. Traditional systems often fail to adequately analyze user requests and can only provide insufficient solutions, leading to decreased user satisfaction.

[1523] 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.

[1524] In this invention, the server includes means for performing natural language analysis on user requests, means for obtaining appropriate information from a database based on the natural language analysis results, and means for generating optimal suggestions for the user using generative artificial intelligence. This makes it possible to accurately understand user requests and quickly provide optimal suggestions.

[1525] "Means for inputting user requests" refers to devices or software that provide an interface that allows users to input arbitrary requests or questions in text or voice.

[1526] "Means for sending the request to the server" refers to communication protocols, devices, and software for sending the input request to the server via the Internet.

[1527] "Means for performing natural language analysis on the server" refers to natural language processing technology and software used by the server to analyze the received request and understand the user's intent.

[1528] "Means for obtaining appropriate information from a database based on the results of natural language analysis" refers to a process and software for searching and obtaining appropriate data from a database based on the information obtained as a result of natural language analysis.

[1529] "Means of generating optimal suggestions for users using generative artificial intelligence" refers to the process and software that uses generative artificial intelligence (e.g., machine learning models) to generate optimal suggestions based on analyzed requests and acquired data.

[1530] "Means for sending the generated proposal to the user's terminal" refers to the communication protocol, device, and software that send the generated proposal from the server to the user's terminal.

[1531] "Means for displaying the proposal to the user" refers to an interface and software for visually displaying the received proposal content on the user's terminal.

[1532] "Means by which the user selects an action based on the displayed suggestion" refers to an input device and software for the user to make a selection or take action based on the suggested content.

[1533] "Means for transmitting the user's actions to the server" refers to communication protocols, devices, and software that transmit the user's selected actions and feedback to the server.

[1534] "Means for receiving and analyzing user feedback on the server" refers to the process and software for receiving feedback sent by the user and analyzing its contents.

[1535] "Means for updating the generative artificial intelligence based on the aforementioned feedback" refers to the process and software for improving and updating the algorithms and models of the generative artificial intelligence based on feedback obtained from users.

[1536] This invention is a customer support system that provides appropriate suggestions to meet diverse user requests, and is implemented using the following hardware and software. The system consists of a terminal for inputting user requests, a server that analyzes the requests and provides appropriate information, and a database. Furthermore, a generative artificial intelligence processes data between these components to generate optimal suggestions.

[1537] Hardware and software

[1538] Terminal: A device used by users to input requests, including smartphones, tablets, and personal computers. This terminal has applications installed that provide a user interface.

[1539] Server: A central computer that analyzes requests and provides appropriate information. The server has natural language processing engines (e.g., spaCy and NLTK), generative artificial intelligence (e.g., TensorFlow and PyTorch), and database management systems (e.g., MySQL and PostgreSQL) installed.

[1540] Database: A management system for storing information related to user requests, including the latest Android models and pricing plan information.

[1541] System operation

[1542] The user enters a request through an application on their device, such as "I want the latest Android model, but I want to use it as cheaply as possible." This input is sent to the server using an HTTP request. The server receives this request data and parses it using a natural language processing engine. Based on the analysis results, the server retrieves relevant information from its database. Using this retrieved information, a generative artificial intelligence generates the most suitable suggestion for the user.

[1543] The generated suggestions are formatted, converted to JSON format, and sent to the device as an HTTP response. The device then displays the suggestions to the user. Based on the suggested information, the user can choose an action and proceed with the next steps. For example, the user can select the most suitable Android model and proceed with the purchase.

[1544] Specific example

[1545] Specifically, if a user wants to know "the latest Android models, their prices, and promotional information," the system works as follows: The user enters "I want the latest Android model, but I want to use it as cheaply as possible" through an application on their device. This request is sent to the server, which performs natural language processing to understand the user's intent. Next, it retrieves relevant information from the database and generates optimal suggestions using generative artificial intelligence. These suggestions are sent to the device, and the user can review them.

[1546] Example of a prompt

[1547] "Please tell me how to buy the latest Android model cheaply."

[1548] "I want to transfer 50,000 images from iCloud to my Android device."

[1549] "Please suggest which device is best suited to our new pricing plan."

[1550] This system allows users to resolve problems quickly and accurately and receive highly satisfactory service.

[1551] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1552] Step 1:

[1553] User: The user enters their request through the application on their device. For example, they might enter, "I want the latest Android model, but I want to use it as cheaply as possible." The entered text is saved to the application's internal memory.

[1554] Input: User request written in natural language

[1555] Output: Request text stored in the terminal's internal memory

[1556] Example of operation: The user enters "I want the latest Android model, but I want to use it as cheaply as possible" into the application's text box and presses the submit button.

[1557] Step 2:

[1558] Terminal: Sends user-entered requests to the server. Specifically, request data is sent to the server as an HTTP POST request. The request body contains user input in JSON format.

[1559] Input: Request text

[1560] Output: HTTP request sent to the server

[1561] Example of operation: The device sends an HTTP POST request containing "request data" to the " / api / requests" endpoint.

[1562] Step 3:

[1563] Server: The server parses the received request data. First, it extracts the JSON data from the request body and then retrieves the request text from within it.

[1564] Input: HTTP Request

[1565] Output: Request text

[1566] Example of operation: The server extracts a request from the body of an HTTP request it receives, such as "I want the latest Android model, but I want to use it as cheaply as possible."

[1567] Step 4:

[1568] Server: The server analyzes the acquired text using a natural language processing engine (e.g., spaCy or NLTK) to extract the user's intent. Specifically, it analyzes keywords and phrases in the text to extract intents such as "latest model" or "buy cheaply."

[1569] Input: Request text

[1570] Output: Analyzed intent (e.g., keywords or phrases)

[1571] Specific example of operation: The server uses spaCy to analyze the text "I want the latest Android model, but I want to use it as cheaply as possible" and extracts the intents "latest model" and "buy cheaply".

[1572] Step 5:

[1573] Server: Based on the analysis results, retrieve relevant information from the database. Specifically, it generates SQL queries and sends requests to the database. For example, it executes an SQL query such as "SELECT model, price FROM android_devices WHERE release_date > '2023-01-01'".

[1574] Input: Analyzed intent

[1575] Output: Information retrieved from the database (e.g., the latest Android models and their prices)

[1576] Example of operation: The server executes the query "SELECT model, price FROM android_devices WHERE release_date > '2023-01-01'" on the MySQL database to retrieve the latest Android models and their prices.

[1577] Step 6:

[1578] Server: Generates optimal suggestions for the user based on information and analysis results obtained using generative artificial intelligence (e.g., TensorFlow or PyTorch). Suggestion generation considers both the user's requests and information obtained from the database.

[1579] Input: Information retrieved from the database, analyzed intent

[1580] Output: Generated proposals

[1581] A concrete example of operation: The server uses a TensorFlow model to generate suggestions such as, "The latest Android model X can be purchased for 30,000 yen."

[1582] Step 7:

[1583] Server: Formats the generated proposal, converts it to JSON format, and sends it to the user's terminal as an HTTP response.

[1584] Input: Generated proposal

[1585] Output: HTTP response (JSON format) sent to the terminal.

[1586] Specific example of operation: The server formats the proposed content into JSON format such as "Model X: 30,000 yen, Campaign Y" and sends it to the terminal via HTTP response.

[1587] Step 8:

[1588] Terminal: Displays the received suggestions to the user. Specifically, the application's UI analyzes the suggestions and displays them in a visually easy-to-understand format.

[1589] Input: HTTP response (JSON format)

[1590] Output: Displayed suggestions

[1591] Example of operation: The device parses the JSON data it receives and displays a message on the screen such as, "The latest Android model X is available for purchase for 30,000 yen."

[1592] Step 9:

[1593] User: Review the suggestions and choose the most suitable option. For example, the user selects a specific Android model and proceeds with the purchase.

[1594] Input: Displayed suggestion

[1595] Output: User selection (e.g., purchase decision)

[1596] Example of the operation: The user clicks the "Purchase button" and is redirected to a screen to proceed with the purchase process.

[1597] Step 10:

[1598] Device: Triggers the following action based on the user's selection: redirect to a purchase link or initiate a change of price plan.

[1599] Input: User Selection

[1600] Output: Next triggered action

[1601] Example of the operation: The user clicks the "Purchase button" and is redirected to a screen to proceed with the purchase process.

[1602] Step 11:

[1603] User: Enter feedback on the proposal. For example, provide comments such as, "This plan does not provide enough data capacity."

[1604] Input: Feedback on the proposed content

[1605] Output: Feedback entered into the terminal

[1606] Example of the operation: The user enters "This plan does not provide enough data capacity" into the feedback text box.

[1607] Step 12:

[1608] Terminal: Sends feedback data to the server. Specifically, it sends feedback data to the server as an HTTP POST request. The request body contains the feedback in JSON format.

[1609] Input: Feedback text

[1610] Output: HTTP request sent to the server

[1611] Specific example of operation: The terminal sends an HTTP POST request containing "feedback data" to " / api / feedback".

[1612] Step 13:

[1613] Server: Analyzes received feedback data and uses it to improve generative artificial intelligence and systems. This allows for the incorporation of feedback into future suggestions, thereby improving the user experience.

[1614] Input: HTTP request (feedback data)

[1615] Output: Updated generative AI models and system improvement data

[1616] Specific example of operation: The server analyzes the feedback data it receives and uses it to refine the proposed algorithm for the next time.

[1617] (Application Example 1)

[1618] 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".

[1619] In modern e-commerce sites, users often struggle to quickly find the right product from a vast selection. Furthermore, customer support is required to provide prompt and accurate responses to diverse user needs. However, conventional systems have been unable to effectively utilize natural language processing and generative artificial intelligence, making it difficult to improve the user experience. Solving this problem is the objective of this invention.

[1620] 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.

[1621] In this invention, the server includes means for analyzing intent using a natural language processing engine, means for retrieving products and FAQs from a database, and means for returning suggestions in JSON format. This makes it possible to accurately analyze user requests and provide optimal product suggestions and support information quickly and accurately.

[1622] A "user" is a person who uses a system to input requests.

[1623] "Requests" refer to all kinds of wishes and questions that users input into the system.

[1624] A "server" is a computer that receives and analyzes user requests, and generates and provides appropriate information in conjunction with a database.

[1625] "Natural language processing" is a technology that analyzes the requests received from users and understands their intentions.

[1626] "Generative artificial intelligence" refers to an algorithm or system that generates optimal suggestions based on analyzed information.

[1627] A "terminal" is an electronic device used by a user to input requests and display suggestions from a server.

[1628] A "database" is a storage device that stores information such as product information and FAQs.

[1629] "Feedback" refers to the opinions and evaluations that users provide regarding a proposal.

[1630] A "natural language processing engine" is software or an algorithm that analyzes user requests and understands their intent.

[1631] A "proposal" is information or a choice provided to the user, generated by a generative artificial intelligence system.

[1632] "Products and FAQs" refer to product information and answers to frequently asked questions stored in the database.

[1633] JSON format is a lightweight data exchange format for structuring, storing, and transmitting data.

[1634] This invention provides appropriate product suggestions and support information in response to user requests in a customer support system for an e-commerce site. The system analyzes the user's request using natural language processing, generates optimal suggestions using generative artificial intelligence, and provides them to the user. Specific embodiments are described below.

[1635] System Configuration

[1636] The system will be implemented with the following configuration.

[1637] Device: An electronic device (e.g., smartphone, tablet) where the user enters a request and receives a suggestion.

[1638] Server: A computer that analyzes user requests and generates appropriate information.

[1639] Database: A storage device that stores product information and FAQs.

[1640] Natural language processing engine: Software used to analyze user requests.

[1641] Generative artificial intelligence: An algorithm that generates optimal suggestions based on analysis results.

[1642] Program processing

[1643] Entering and submitting requests

[1644] The user enters the request using a terminal. The request is sent to the server using an HTTP request.

[1645] natural language analysis

[1646] The server analyzes the received request using a natural language processing engine. Specifically, it uses the Hugging Face Transformers library to understand the user's intent.

[1647] Retrieving information from the database

[1648] Based on the analysis results, the server retrieves product information and FAQs from the database using SQL queries. For example, it might execute a query like "SELECT name, price, description FROM products WHERE description LIKE '%Summer Cooling Mattress%'".

[1649] Proposal generation

[1650] The server uses generative artificial intelligence (e.g., TensorFlow models) to generate optimal suggestions based on the acquired data. The suggestions are formatted in JSON format and sent to the terminal as an HTTP response.

[1651] Display of proposed content

[1652] The user's device displays the received suggestions. This allows the user to view specific product information and answers to frequently asked questions.

[1653] Feedback collection and analysis

[1654] Users input feedback on the suggestions and send it from their device to the server. The server analyzes this feedback and updates the generative artificial intelligence to improve the accuracy of future suggestions.

[1655] Specific example

[1656] If a user enters "I want a cooling mattress for summer," the system will operate as follows:

[1657] 1. The server uses a natural language processing engine to analyze the request, "cooling mattress for summer."

[1658] 2. Retrieve the relevant product information from the database.

[1659] 3. Generate optimal product suggestions using generative artificial intelligence.

[1660] 4. Format the generated proposals in JSON format and send them to the terminal.

[1661] 5. The user reviews the suggestion and chooses an action.

[1662] Example of a prompt

[1663] User: I want a cooling mattress for summer.

[1664] System: We found several mattresses made with cooling materials for summer. The following products are recommended.

[1665] Product name: "Summer Sleep Mattress"

[1666] Price: 5,000 yen

[1667] Description: This mattress uses cooling material for a comfortable sleeping experience.

[1668] In this way, we can provide quick and accurate solutions to meet user requests.

[1669] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1670] Step 1:

[1671] The user enters a request into the terminal and sends it. The user specifically enters, "I want a cooling mattress for summer." The input data is sent to the server as an HTTP request.

[1672] Step 2:

[1673] The server analyzes the received request using a natural language processing engine (e.g., Hugging Face's Transformers library). It extracts key keywords and intent from the request data and recognizes the intent as "cooling mattress for summer." This analysis result is then used for subsequent database matching.

[1674] Step 3:

[1675] Based on the analysis results, the server retrieves appropriate product information from the database. Specifically, it executes an SQL query and sends a query like "SELECT name, price, description FROM products WHERE description LIKE '%Summer Cooling Mattress%'". The database returns the corresponding product information.

[1676] Step 4:

[1677] The server processes the acquired data using generative artificial intelligence (e.g., TensorFlow models) to create optimal recommendations. It selects the best product for the user, taking into account factors such as price, performance, and the user's past search history. This information will later be returned to the user as recommendations.

[1678] Step 5:

[1679] The server formats the generated proposal in JSON format and sends it to the terminal as an HTTP response. The formatted proposal includes product name, price, and description.

[1680] Step 6:

[1681] The terminal displays the received suggestions to the user. The user can check specific product information and suggestions. Information such as "Summer Sleep Mattress" and "Comfortable Mattress with Cooling Material" will be displayed.

[1682] Step 7:

[1683] The user reviews the displayed suggestions and chooses an action, such as clicking a purchase button.

[1684] Step 8:

[1685] The device sends user actions to the server. The action data arrives at the server as an HTTP request.

[1686] Step 9:

[1687] The server receives and analyzes user feedback. It updates the generative artificial intelligence based on the feedback data and incorporates it into future suggestions. For example, if feedback states, "This mattress is comfortable, but it's expensive," the system will improve its suggestions to prioritize price.

[1688] 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.

[1689] This invention is a customer support system that provides more personalized suggestions by incorporating an emotion engine that recognizes the user's emotions, and is implemented as follows.

[1690] System Configuration

[1691] This system consists of a terminal for inputting user requests and emotions, a server that analyzes requests and emotions and provides appropriate information, a database, and generative artificial intelligence. Furthermore, an emotion engine works in conjunction with these components to provide responses based on the user's emotions.

[1692] Program processing

[1693] Enter the user's requests and feelings.

[1694] Terminal: When a user launches an application on their terminal and enters a request, their emotions are also recognized. For example, if a user enters "I want the latest Android model, but I want to use it as cheaply as possible," facial recognition and voice analysis will determine that the user is feeling dissatisfied.

[1695] Sending requests and emotions

[1696] Terminal: Sends the entered request and sentiment data to the server in an appropriate data format such as JSON. Sends it to the server as an HTTP request.

[1697] Receiving and analyzing requests and emotions

[1698] Server: Passes received requests and sentiment data to the natural language processing engine and sentiment engine for analysis. For example, it uses the Python NLTK library to extract the user's intent and the sentiment engine to recognize the user's emotion as "dissatisfaction."

[1699] Database matching

[1700] Server: Based on the analysis results, retrieves the latest Android model and pricing plan information from the database using SQL queries. For example, it executes a query such as "SELECT model, price FROM android_devices WHERE release_date > '2023-01-01'".

[1701] AI-powered proposal generation

[1702] Server: Inputs acquired data and sentiment analysis results into a generative artificial intelligence system to generate optimal suggestions that take user emotions into consideration. For example, to address dissatisfaction, it might include special discounts or additional benefits in its suggestions.

[1703] Format and send the proposal.

[1704] Server: Formats the generated suggestions, adds a message tailored to the user's sentiment, and converts it into a JSON response. The generated suggestions include details such as "model name," "price," and "campaign information."

[1705] Display of proposed content

[1706] Device: Displays received suggestions to the user. For example, it displays detailed information and pricing for new Android models, related promotional information, and special offers to address user complaints.

[1707] User selections and actions

[1708] User: Review the suggestions and choose the most suitable option. For example, select a specific Android model and proceed with the purchase.

[1709] Device: Based on the user's selection, the following actions are triggered. For example, redirecting to the purchase page for the selected Android model, or initiating the process of changing the price plan.

[1710] Feedback collection and analysis

[1711] User: Enter feedback on the proposal. In addition to opinions such as "This plan does not have enough data capacity," also enter your feelings about the proposal.

[1712] Terminal: Acquires feedback data, analyzes it with an emotion engine, and sends it to the server.

[1713] Feedback analysis and AI updates

[1714] Server: Analyzes received feedback and sentiment data to update the generative artificial intelligence and sentiment engine models. This results in more personalized suggestions for the next session, improving user satisfaction.

[1715] Specific example

[1716] Specifically, this occurs when the emotion engine determines that a user is stressed after typing, "I want to transfer 50,000 images from iCloud to my Android device."

[1717] 1. Natural Language Processing and Sentiment Recognition: The server extracts the request "migrate iCloud images" and the emotion "stress".

[1718] 2. Database matching: Retrieve information about data migration from iCloud to Android from the database.

[1719] 3. Suggestion Generation: Generative artificial intelligence is used to suggest the optimal data migration app and procedure. Free support offers are also included to reduce user stress.

[1720] 4. Display of proposals: Display the proposals on the user's device and include information about special support.

[1721] 5. Feedback Collection and Analysis: After users complete the migration process, they input their emotions along with feedback about their experience. The server analyzes this data to help improve future suggestions.

[1722] This system provides more personalized support that takes user emotions into account, leading to increased satisfaction.

[1723] The following describes the processing flow.

[1724] Step 1:

[1725] Terminal: The user launches the application on their terminal and enters their request. For example, they might enter text such as, "I want the latest Android model, but I want to use it as cheaply as possible."

[1726] Step 2:

[1727] Terminal: While the user is entering a request, the terminal's camera and microphone are used to analyze the user's facial expressions and voice tone in real time using an emotion engine. For example, facial recognition technology can detect emotions such as "dissatisfaction" or "excitement."

[1728] Step 3:

[1729] Terminal: Sends the entered request data and sentiment data to the server in an appropriate data format such as JSON. Data is sent using HTTP requests.

[1730] Step 4:

[1731] Server: Analyzes received request data and sentiment data. The natural language processing engine analyzes the text request, and the sentiment engine analyzes the sentiment data. For example, it extracts the request "I'm looking for the latest model at a low price" and the sentiment "dissatisfied."

[1732] Step 5:

[1733] Server: Based on the request analysis results, retrieve appropriate information from the database. Use SQL queries to retrieve information on the latest Android models and pricing plans. For example, execute the query "SELECT model, price FROM android_devices WHERE release_date > '2023-01-01'".

[1734] Step 6:

[1735] Server: Based on the acquired information, a generative artificial intelligence generates optimal suggestions. Furthermore, it adjusts the suggestions considering the sentiment analysis results. For example, if the user has the emotion of "dissatisfaction," it will include special discounts or additional benefits in its suggestions.

[1736] Step 7:

[1737] Server: Formats the generated suggestions, adds sentiment-based messages, and converts them into a JSON response. The generated suggestions include details such as "model name," "price," and "campaign information."

[1738] Step 8:

[1739] Server: Sends the formatted proposal content to the terminal as an HTTP response.

[1740] Step 9:

[1741] Device: Displays received suggestions to the user. For example, it displays detailed information and pricing for new Android models, related promotional information, and special offers to address user complaints.

[1742] Step 10:

[1743] User: Review the suggestions and choose the most suitable option. For example, select a specific Android model and proceed with the purchase.

[1744] Step 11:

[1745] Device: Based on the user's selection, trigger the following actions. For example, redirect to the purchase page for the selected Android model, or initiate the process of changing the price plan.

[1746] Step 12:

[1747] User: Enter feedback on the proposed plan. This includes comments such as "This plan does not provide sufficient data capacity," as well as your feelings at the time of giving feedback.

[1748] Step 13:

[1749] Terminal: Acquires feedback data and sentiment data and sends it to the server.

[1750] Step 14:

[1751] Server: Analyzes received feedback and emotion data to update the generative artificial intelligence and emotion engine models. This results in more personalized suggestions for the next session, improving user satisfaction.

[1752] This system provides personalized suggestions that take user emotions into consideration, thereby improving user satisfaction.

[1753] (Example 2)

[1754] 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".

[1755] Traditional customer support systems, while providing appropriate information in response to user requests, failed to consider user emotions, resulting in lower user satisfaction. Furthermore, they lacked systems for effectively utilizing user feedback and improving suggestions. This made it difficult to provide optimal service to users.

[1756] 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.

[1757] In this invention, the server includes means for inputting user requests and sentiment data; means for transmitting the requests and sentiment data to the server; means for performing natural language processing and sentiment analysis on the requests and sentiment data on the server; means for obtaining appropriate information from a database based on the natural language processing results and sentiment analysis results; means for generating optimal suggestions for the user using generative artificial intelligence; means for transmitting the generated suggestions to the user's terminal; means for displaying the suggestions to the user; means for the user to select an action based on the displayed suggestions; means for transmitting the user's actions to the server; means for receiving and analyzing user feedback on the server; and means for updating the generative artificial intelligence and sentiment analysis engine based on the feedback. This enables the provision of personalized suggestions that take user sentiment into consideration, thereby improving user satisfaction. Furthermore, it enables continuous improvement of the system based on feedback.

[1758] "Means for inputting user requests and emotional data" refers to functions and devices that allow users to input their desired requests and their emotional state at the time using a terminal.

[1759] "Means for sending request and sentiment data to the server" refers to a communication function for sending request and sentiment data entered on the terminal to the server.

[1760] "Means for natural language processing and sentiment analysis of request and sentiment data" refers to a function that analyzes request data received on a server using machine learning models, etc., and analyzes sentiment data using sentiment recognition engines, etc.

[1761] "Means for obtaining appropriate information" refers to the function of retrieving necessary information from a database based on the results of natural language processing and sentiment analysis.

[1762] "A means of generating optimal suggestions for users using generative artificial intelligence" refers to a function that uses a generative artificial intelligence model to create optimal suggestions for users based on acquired information and sentiment data.

[1763] "Means for sending generated suggestions to the user's terminal" refers to a communication function for sending suggestions created by generative artificial intelligence to the user's terminal.

[1764] "Means of displaying suggestions to the user" refers to a function that displays the content of suggestions received on the terminal to the user visually or audibly.

[1765] "Means for users to choose their actions" refers to interface functions that allow users to make choices in response to displayed suggestions.

[1766] "Means for sending user actions to the server" refers to a communication function for sending user-selected action data to the server.

[1767] "Means for receiving and analyzing feedback" refers to a function that receives user feedback data on a server and analyzes that data.

[1768] "Means for updating generative artificial intelligence and sentiment analysis engines" refers to functions for continuously improving generative artificial intelligence models and sentiment analysis engines based on analyzed feedback data.

[1769] This invention relates to a customer support system that provides more personalized suggestions by incorporating an emotion engine that recognizes user emotions. The system consists of a terminal for inputting user requests and emotions, a server that analyzes requests and emotions and provides appropriate information, a database, and generative artificial intelligence.

[1770] Hardware and software configuration

[1771] terminal

[1772] A terminal is a device used by users to input requests and emotional data, and includes smartphones, tablets, and PCs. The terminal's application uses the camera and microphone to acquire user input and perform facial recognition and voice analysis. A common facial recognition API (e.g., Google Cloud Vision API) is used for facial recognition, and a common voice analysis API (e.g., Amazon Transcribe) is used for voice analysis.

[1773] server

[1774] The server analyzes received requests and sentiment data and processes them to generate appropriate suggestions. The server uses Python's Natural Language Toolkit (NLTK) and spaCy for its natural language processing engine, and machine learning models for its sentiment analysis engine. A database (e.g., an SQL database) is also integrated to retrieve relevant information based on the analysis results and sentiment data. For generative artificial intelligence, advanced generative models such as GPT-3 are used.

[1775] Specific processing of the program

[1776] The program of this system performs the following specific processing.

[1777] Request and emotional input

[1778] The user launches an application on their device and enters a request, while simultaneously capturing emotional data through the camera and microphone. For example, if the user enters "I want the latest Android model, but I want to use it as cheaply as possible," the device's camera will capture a dissatisfied expression.

[1779] Sending requests and emotions

[1780] The device converts the acquired request and sentiment data into JSON format and sends it to the server as an HTTP POST request.

[1781] Receiving and analyzing requests and emotions

[1782] The server analyzes received requests and sentiment data using a natural language processing engine and a sentiment analysis engine. The natural language processing engine extracts the user's requests, and the sentiment analysis engine recognizes the user's emotions.

[1783] Database matching

[1784] Based on the analysis results, the server uses SQL queries to retrieve appropriate information from the database. For example, it might execute a query like "SELECT model, price FROM android_devices WHERE release_date > '2023-01-01'" to retrieve the latest Android model information.

[1785] AI-powered proposal generation

[1786] The server inputs prompts into a generative artificial intelligence model based on the acquired information and sentiment analysis results, generating the most suitable suggestions for the user. For example, it uses the following prompts:

[1787] "This user wants the latest Android device, but is dissatisfied with the lack of a suitable solution to their needs, as they want to use it as cheaply as possible. Please provide them with the best possible offer."

[1788] Submitting and displaying proposals

[1789] The generated suggestions are reformatted and sent to the user's device as a JSON response. The user can then visually or audibly review the suggestions on their device. For example, a suggestion such as "Latest Android model: Android XYZ, Price: 50,000 JPY, Special discount 10%" might be displayed.

[1790] Feedback collection and analysis

[1791] The user enters feedback on the suggestion into their device and sends it back to the server. The server analyzes the feedback and sentiment data, and uses the results to update the generative artificial intelligence model and sentiment analysis engine, making the next suggestion more accurate.

[1792] Specific example

[1793] For example, if a user enters "I want to transfer 50,000 images from iCloud to Android" and the sentiment analysis engine determines that the user is feeling stressed:

[1794] 1. The server extracts the request "migrate iCloud images" and the emotion "stress".

[1795] 2. Retrieve information about data migration from iCloud to Android from the database.

[1796] 3. Use generative artificial intelligence to suggest the optimal data migration app and procedure, and include a free support offer.

[1797] 4. Display the proposed content on the user's device and include information about special support.

[1798] 5. After the user completes the migration process, they input their emotions along with feedback about their experience. The server analyzes this data to help improve future suggestions.

[1799] This system provides more personalized support that takes user emotions into account, leading to increased satisfaction.

[1800] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1801] Step 1:

[1802] User: Launch the application on the device and enter a request in the text field. For example, enter a request such as, "I want the latest Android model, but I want to use it as cheaply as possible." Generate input data (request text).

[1803] Terminal: Acquires input data and generates emotion data using face recognition APIs and voice analysis APIs with the camera and microphone. For example, it can use the Google Cloud Vision API to recognize an unhappy expression and obtain emotion data.

[1804] Input: User's requested text, user's facial image, or voice data

[1805] Output: Request text, dissatisfaction data

[1806] Step 2:

[1807] Terminal: Converts the acquired request and sentiment data into JSON format and sends it to the server using an HTTP POST request.

[1808] Input: Request text, dissatisfaction data

[1809] Output: HTTP POST request (JSON format)

[1810] Step 3:

[1811] Server: Receives HTTP requests and extracts requests and sentiments from JSON data. The extracted data is then passed to a natural language processing engine (e.g., NLTK or spaCy) and a sentiment analysis engine for analysis.

[1812] Input: HTTP POST request (JSON format)

[1813] Output: Request data as analysis results (e.g., "latest Android model"), sentiment data ("dissatisfied")

[1814] Step 4:

[1815] Server: Based on the analysis results, it executes an SQL query to retrieve the necessary information from the database. For example, to retrieve "information about the latest Android models," it executes "SELECT model, price FROM android_devices WHERE release_date > '2023-01-01'".

[1816] Input: Requested data for analysis results ("latest Android model")

[1817] Output: Database query results (e.g., model name and price of a specific device)

[1818] Step 5:

[1819] Server: Based on the acquired information, it inputs prompts into a generative artificial intelligence (e.g., GPT-3) to generate the best possible suggestions for the user. For example, it uses the following prompts:

[1820] "This user wants the latest Android device, but is dissatisfied with the lack of a suitable solution to their needs, as they want to use it as cheaply as possible. Please provide them with the best possible offer."

[1821] Input: Database query results, analyzed sentiment data

[1822] Output: Generated proposal details (e.g., model name, price, special discount information)

[1823] Step 6:

[1824] Server: Converts the generated proposal into a JSON response and sends it to the terminal as an HTTP response.

[1825] Input: Generated proposal content

[1826] Output: HTTP response (JSON format)

[1827] Step 7:

[1828] Terminal: Displays suggestions received from the server to the user. For example, it displays detailed information and pricing for the latest Android models, as well as related promotional information and special offers.

[1829] Input: HTTP response (JSON format)

[1830] Output: Suggestions displayed to the user

[1831] Step 8:

[1832] User: Review the suggestions and choose the best option. For example, select a recommended Android model and proceed with the purchase.

[1833] Input: Suggestions displayed to the user

[1834] Output: User's selected action

[1835] Step 9:

[1836] Device: Based on the user's selection, the following actions are triggered. For example, redirecting to the purchase page for the selected Android model, or initiating the process of changing the price plan.

[1837] Input: User's selected action

[1838] Output: Triggers the following action (e.g., redirects to the purchase page)

[1839] Step 10:

[1840] User: Enter feedback on the proposal. For example, enter an opinion such as "This plan does not provide enough data capacity" and a feeling of "dissatisfaction".

[1841] Input: User feedback text, facial images, and audio data

[1842] Output: Feedback data and sentiment data

[1843] Step 11:

[1844] Terminal: Acquires feedback data, analyzes it with an emotion engine, and sends it to the server as an HTTP POST request.

[1845] Input: Feedback data, sentiment data

[1846] Output: HTTP POST request (feedback data and sentiment data)

[1847] Step 12:

[1848] Server: Analyzes received feedback and sentiment data to update the generative artificial intelligence and sentiment analysis engine models. This allows for more personalized suggestions and improved user satisfaction.

[1849] Input: HTTP POST request (feedback data and sentiment data)

[1850] Output: Updated generative artificial intelligence model and sentiment analysis engine

[1851] (Application Example 2)

[1852] 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".

[1853] Traditional customer support systems have failed to adequately consider not only user requests but also their emotions. As a result, users often do not receive suggestions tailored to their feelings, leading to decreased satisfaction. Furthermore, insufficient personalization that takes user emotions into account makes it difficult to offer special offers or discounts that will stimulate purchase intent. Solving these problems is essential.

[1854] 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.

[1855] In this invention, the server includes means for inputting requests and emotions from the user, means for transmitting the requests and emotions to the server, means for performing natural language analysis and sentiment analysis on the requests and emotions on the server, means for obtaining appropriate information from a database based on the natural language analysis results and sentiment analysis results, means for generating optimal suggestions that take the user's emotions into consideration using generative artificial intelligence, means for transmitting the generated suggestions to the user's terminal, means for displaying the suggestions to the user, means for the user to select an action based on the displayed suggestions, means for transmitting the user's actions to the server, means for receiving and analyzing feedback and emotions from the user on the server, means for updating the generative artificial intelligence based on the feedback and sentiment analysis results, and includes special offers and discount suggestions. This enables personalized suggestions that take the user's emotions into consideration, improving user satisfaction and stimulating purchase intent.

[1856] A "means for inputting user requests and emotions" refers to a means that provides an interface that allows users to input their requests along with their emotions at that time.

[1857] "Means for sending the aforementioned requests and emotions to the server" refers to communication means for sending requests and emotions entered by the user to the server.

[1858] "Means for performing natural language processing and sentiment analysis on the server" refers to means for analyzing requests and sentiments on the server using natural language processing technology and sentiment recognition technology.

[1859] "Means for obtaining appropriate information from a database based on the natural language analysis results and sentiment analysis results" refers to means for obtaining relevant information from a database based on the analysis results.

[1860] "A method for generating optimal suggestions that take user emotions into consideration using generative artificial intelligence" refers to a method that uses generative artificial intelligence to generate optimal suggestions while taking into account the user's emotional state.

[1861] "Means for sending the generated proposal to the user's terminal" refers to communication means for sending the generated proposal to the user's terminal.

[1862] "Means for displaying the proposal to the user" refers to means for displaying the content of the proposal to the user on a terminal.

[1863] "Means by which the user selects an action based on the displayed suggestion" refers to means equipped with an interface for the user to select an action based on the displayed suggestion.

[1864] "Means for transmitting the user's actions to the server" refers to communication means for transmitting the actions selected by the user to the server.

[1865] "Means for receiving and analyzing user feedback and emotions on the server" refers to means for receiving user feedback and the emotions associated with it, and for analyzing them.

[1866] "Means for updating the generative artificial intelligence based on the aforementioned feedback and sentiment analysis results" refers to means for updating the model of the generative artificial intelligence based on feedback obtained from the user and the results of sentiment analysis.

[1867] "Special offers and discount proposals" refer to offers that include special discounts or promotions for users.

[1868] This invention is a customer support system that recognizes user emotions and makes emotion-based suggestions. The system consists of a user terminal, a server, a database, a generative artificial intelligence system, and an emotion engine.

[1869] System Configuration

[1870] 1. User terminal

[1871] Users access the system using smartphones or tablets.

[1872] Users utilize an interface that allows them to input requests and the emotions associated with those requests. Emotion recognition employs facial recognition and voice analysis technologies.

[1873] Specific software libraries such as OpenCV (face recognition) and TextBlob (speech analysis) are used.

[1874] 2. Server

[1875] Receives request and sentiment data sent from the user's terminal.

[1876] The data is analyzed using a natural language processing engine (e.g., Python's NLTK library) and an emotion engine (deep learning model).

[1877] Based on the analysis results, relevant information is retrieved from the database. Here, SQL queries are used to obtain the necessary information from the database.

[1878] Using generative artificial intelligence, the system generates optimal suggestions that take user emotions into consideration. These suggestions may include special discounts or cashback offers.

[1879] 3. Database

[1880] The database stores user requests and suggestions and provides information in response to queries from the server.

[1881] SQL and NoSQL databases may be used.

[1882] Explanation of the program's processing

[1883] The user uses their smartphone's camera and microphone to input their requests and associated emotions into the system. OpenCV is used for facial recognition, and TextBlob for speech analysis. This allows the system to recognize what emotions the user is experiencing (e.g., stress, frustration, satisfaction).

[1884] The server receives user requests and emotion data, analyzes the requests using a natural language processing engine, and analyzes the emotions using an emotion engine. Based on the analysis results, it retrieves appropriate information from the database and inputs it into the generative artificial intelligence. The generative AI generates optimal suggestions that take the user's emotions into account and sends these suggestions to the user's terminal in an appropriate data format such as JSON. The user reviews the suggestions on their terminal and selects an action. The server receives the user's selected action and also collects feedback information. It analyzes the feedback and the emotions associated with it, and updates the generative AI model.

[1885] Specific example

[1886] The following is an example of a prompt message used by the server to generate the best possible suggestions for the user using generative artificial intelligence:

[1887] User's emotion: Stress

[1888] Offer details: 10% special discount, 1000 yen cashback

[1889] Please create the best message to deliver to this user.

[1890] A concrete example would be a user who enters "I'm looking for the latest smartphone, but I'm worried because they're expensive," and the system determines that the user is experiencing stress. Based on the analysis, the server retrieves the latest smartphone models and price information from its database, and uses generative artificial intelligence to include special discounts and cashback offers in its suggestions. These suggestions are sent to the user's device, and the user, receiving the special offers, becomes more inclined to purchase and makes a selection. The system then collects feedback from the user and incorporates it into future suggestions.

[1891] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1892] Step 1:

[1893] The user inputs requests and emotions on their device. Using a smartphone or tablet, the user inputs their requests in text format and simultaneously inputs emotions using the camera and microphone. The input emotion data is then analyzed using OpenCV for facial recognition and TextBlob for speech analysis. This process yields request data and emotion data as input.

[1894] Step 2:

[1895] The device sends request and sentiment data to the server. The data sent from the device is in JSON format and is sent to the server as an HTTP request. The output obtained here is the request and sentiment data sent to the server.

[1896] Step 3:

[1897] The server analyzes the received request and sentiment data. First, the server uses a natural language processing engine (e.g., Python's NLTK library) to analyze the request, and then uses a sentiment engine (deep learning model) to analyze the sentiment data. The input consists of the received request and sentiment data, and the output is the analysis results.

[1898] Step 4:

[1899] The server retrieves information from the database based on the analysis results. Based on the analyzed requests and sentiments, it uses SQL queries to retrieve corresponding information from the database. For example, it executes a query such as "SELECT model, price FROM devices WHERE release_date > '2023-01-01'". The input for this step is the analysis results, and the output is the retrieved data.

[1900] Step 5:

[1901] The server uses generative artificial intelligence to generate optimal suggestions. Based on acquired data and sentiment analysis results, the generative AI generates the most suitable suggestions for the user. These suggestions may include special discounts or cashback offers. The output is the generated suggestions.

[1902] Step 6:

[1903] The server sends the generated proposal to the terminal. The generated proposal is formatted in JSON format and sent to the user's terminal as an HTTP response. The output is the proposal data sent to the terminal.

[1904] Step 7:

[1905] The terminal displays the suggestions to the user. The terminal displays the received suggestion data on the user's interface. The output is the displayed suggestion content.

[1906] Step 8:

[1907] The user selects an action based on the suggestions. The user reviews the displayed suggestions and chooses which action to take. This provides the user's selection as input.

[1908] Step 9:

[1909] The device sends user actions to the server. The user's selected action data is sent to the server in JSON format. The output is the action data sent to the server.

[1910] Step 10:

[1911] The server receives and analyzes user feedback and sentiment. It receives user feedback data and re-entered sentiment data and performs analysis. The analysis method is the same as in step 3. The input for this step is feedback and sentiment data, and the output is the analysis results.

[1912] Step 11:

[1913] The server updates the generative artificial intelligence based on feedback and sentiment analysis results. The generative AI model is updated based on the analyzed feedback and sentiment analysis results. This results in more personalized suggestions for future requests. The output is the updated generative AI model.

[1914] Step 12:

[1915] The server generates new proposals based on the updated generative artificial intelligence model. Using the updated model based on feedback, it prepares more optimal proposals for the next request. The inputs to this step are the updated model and the new request, and the output is the new proposal.

[1916] 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.

[1917] 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.

[1918] 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.

[1919] 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.

[1920] 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.

[1921] 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.

[1922] 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.

[1923] 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.

[1924] 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."

[1925] 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.

[1926] 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.

[1927] 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.

[1928] 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.

[1929] 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.

[1930] 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.

[1931] 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 h...

Claims

1. A means of inputting user requests, Means for sending the aforementioned request to the server, The server includes means for performing natural language analysis on the request, A means for obtaining appropriate information from a database based on the natural language processing results, A means of generating optimal suggestions for the user using generative artificial intelligence, Means for sending the generated proposal to the user's terminal, A means for displaying the aforementioned proposal to the user, A means by which the user selects an action based on the displayed suggestion, Means for transmitting the user's actions to the server, The server has means for receiving and analyzing user feedback, A means for updating the generative artificial intelligence based on the aforementioned feedback, A system that includes this.

2. The system according to claim 1, further comprising means for a user to select the optimal option for a pricing plan.

3. The system according to claim 1, further comprising means for transferring user image data between different operating systems.

Images