System

Abstract

A system is provided.SOLUTION: A system comprising: means for receiving a query from a user; means for analyzing the query using a natural language processing technique and extracting important keywords; means for searching for past response information, solution cases, and general technical information based on the extracted keywords; means for inferring a necessary action using a generated AI from the searched information; means for processing the inferred action and presenting the processed action to a respondent; and means for transmitting a response from the respondent to the user using the presented action.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the 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] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] In conventional service desk operations, the response to user inquiries is largely dependent on the skill of the person answering the inquiry, resulting in indefinite response and resolution times. Furthermore, if past response information and case studies cannot be properly utilized, similar problems tend to recur, making efficient responses difficult. The present invention aims to provide a system that can respond to user inquiries quickly and accurately, thereby reducing service desk operating hours and quickly resolving issues. [Means for solving the problem]

[0005] The present invention provides a system that includes means for receiving inquiries from users, analyzing the inquiries using natural language processing technology, extracting important keywords, means for searching past response information, solution cases, and general technical information based on the extracted keywords, means for inferring necessary actions from the search results using a generation AI, means for processing the inferred actions and presenting them to a respondent, and means for the respondent to send a response to the user using the suggested actions.This allows inquiries to be handled efficiently regardless of the respondent's skill level, thereby making it possible to improve the efficiency and quality of service desk operations.

[0006] Below are definitions of important words.

[0007] "Inquiry" means a question or problem report submitted by a User to the Service Desk.

[0008] "Natural language processing technology" refers to all technologies that use computers to interpret, analyze, and generate human language.

[0009] "Keywords" refer to important words or phrases that summarize the content of the inquiry.

[0010] "Past response information" refers to records of similar inquiries that the service desk has handled in the past.

[0011] "Solution cases" refer to specific examples of problems that have occurred in the past and how they have been solved.

[0012] "Technical information" refers to the collective term for technical knowledge and information related to a particular issue.

[0013] "Search" refers to the act of retrieving relevant information from a database or other source based on specific keywords or phrases.

[0014] "Generative AI" refers to a system that uses artificial intelligence technology to generate new information and predictions based on past data and patterns.

[0015] "Action" refers to the specific steps that need to be taken to resolve the problem.

[0016] The term "respondent" refers to a person in charge at the service desk who responds to user inquiries.

[0017] "User" means an individual or organization that contacts the Service Desk. [Brief explanation of the drawings]

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

[0019] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

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

[0021] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

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

[0023] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0024] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0025] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0026] [First embodiment]

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

[0028] 1, a 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 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0030] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0031] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the 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 of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0033] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

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

[0035] 2, in the data processing device 12, a specific process 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" according to the technology of the present 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 process 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.

[0036] The storage 32 stores a data generation model 58 and an 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 process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the 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 process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0038] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0039] Understood. The following is a description of the "Mode for carrying out the invention."

[0040] The present invention relates to a service desk system for quickly and accurately responding to inquiries from users, and more specifically, to a system including means for analyzing received inquiries and predicting appropriate actions.

[0041] Specifically, the system operates as follows: When a user makes an inquiry to the service desk, the inquiry is received by a terminal. Here, it is assumed that the inquiry is input as character string data.

[0042] The device forwards the received inquiry to the server, which then analyzes the inquiry using natural language processing technology. The purpose of the analysis is to extract important keywords contained in the inquiry and the intent of the problem. For example, if a user enters an inquiry such as "My PC suddenly stopped starting up. Even when I turn it on, the screen remains black," the server will extract keywords such as "My PC won't start up" and "The screen is black."

[0043] The server then searches for past support information, solution cases, and general technical information based on the extracted keywords. This uses a knowledge base containing a wealth of technical information and past troubleshooting records. The server quickly searches these databases to gather relevant information.

[0044] The server uses generative AI technology based on the collected information to predict the necessary actions and generate detailed instructions. For example, in the case of the above inquiry, the server would predict specific actions such as "check the power cable connection," "remove and reinsert the battery," and "reset the BIOS."

[0045] The next predicted action is presented to the user by the device. At this time, the device processes the predicted action and displays it in an intuitive and easy-to-understand UI. This allows the respondent to easily check the presented steps and instruct the user on the appropriate action. Specifically, instructions such as "Make sure the power cable is connected correctly. Next, remove the battery, wait a few seconds, and then reinsert it" are sent to the user.

[0046] In this way, the present invention enables quick and consistent responses to user inquiries, improving the efficiency and quality of service desk operations. Appropriate actions can be proposed with high accuracy, regardless of the skills or experience of the respondent, reducing operating time and enabling early problem resolution.

[0047] The processing flow will be explained below.

[0048] Step 1:

[0049] A user enters a query into the service desk interface. For example, the user enters, "My PC suddenly stopped starting up. When I turn it on, the screen remains black."

[0050] Step 2:

[0051] The terminal receives a query from a user. The terminal checks the input data and performs format conversion. Specifically, it converts the string data of the query into structured data.

[0052] Step 3:

[0053] The device sends the query data to the server, often via an API.

[0054] Step 4:

[0055] The server analyzes the inquiry data it receives. Natural language processing technology is used to extract important keywords and phrases from the inquiry. For example, important information such as "My PC won't start" or "The screen is black" is extracted here.

[0056] Step 5:

[0057] The server searches past correspondence information, solution cases, and technical information based on the extracted keywords. The server issues search queries to knowledge bases and related databases to extract the most relevant information.

[0058] Step 6:

[0059] The server uses generative AI based on the search results to predict the necessary actions. The generated actions are organized into specific steps. For example, actions such as "check the power cable connection," "remove and reinsert the battery," and "reset the BIOS" are predicted.

[0060] Step 7:

[0061] The device receives the inferred action from the server, processes it in a format that is easy for the user to view, and presents it to the respondent. It is displayed in an intuitive and easy-to-understand format using a UI.

[0062] Step 8:

[0063] The respondent checks the suggested actions displayed on the device and makes any necessary adjustments. The final answer is then sent to the user. For example, specific instructions such as "Make sure the power cable is connected properly. Then remove the battery, wait a few seconds, and then reinsert it."

[0064] Step 9:

[0065] The user performs specific actions as instructed by the respondent, checks whether the results resolve the issue, and if further inquiries are required, returns to step 1.

[0066] Example 1

[0067] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0068] With conventional service desk systems, it was difficult to respond to user inquiries appropriately and quickly, resulting in inconsistent response quality. Furthermore, because responses depended heavily on the skills and experience of the responders, it was difficult to provide efficient and consistent support.

[0069] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0070] In this invention, the server includes means for receiving inquiries from users, means for analyzing the inquiries using natural language processing technology and extracting important keywords, means for searching past response information, solution cases, and general technical information based on the extracted keywords, means for predicting necessary actions using a generative AI model, means for processing the predicted actions and displaying them on a terminal, and means for the terminal to present an answer to the user using the displayed actions. This enables quick and consistent responses, thereby improving the efficiency and quality of service desk operations.

[0071] An "inquiry" is a question or a trouble report that a user sends to the service desk.

[0072] "Natural language processing technology" is a technology that enables computers to understand, analyze, and process human natural language.

[0073] "Keywords" are important words or phrases extracted from a user's query.

[0074] "Past response information" refers to data on problems that the service desk has previously handled and how they were resolved.

[0075] A "solution case" is a specific method or procedure that solves a specific problem.

[0076] "General technical information" refers to information about technologies and methods that are widely and commonly used.

[0077] A "generative AI model" is an artificial intelligence model used to generate appropriate actions or responses to inquiries.

[0078] A "terminal" is a device (such as a computer or smartphone) that a user uses to access the service desk, send an inquiry, and receive a response.

[0079] "Analysis" refers to the process of understanding a user's query and extracting important information.

[0080] "Inference" means that the generative AI model predicts the appropriate action or response to a query.

[0081] A "knowledge base search engine" is a system for searching for necessary information from a database containing past correspondence information and technical information.

[0082] The present invention relates to a service desk system for quickly and accurately responding to inquiries from users, and more specifically, to a system including means for analyzing received inquiries and predicting appropriate actions.

[0083] The program for this system works as follows: When a user makes an inquiry to the service desk, the inquiry is received by a terminal. The terminal obtains the text data sent by the user in the form of a string. The hardware used here includes terminals such as PCs and smartphones, as well as high-performance database servers.

[0084] Next, the received query string data is transferred from the terminal to the server. The communication protocol used here is HTTP or HTTPS. The server analyzes the received query data using natural language processing (NLP). For this process, an NLP library such as "SpaCy" or "NLTK" is used.

[0085] The purpose of the analysis is to extract important keywords and context. For example, if a user enters a query such as "My PC suddenly stopped starting up. Even when I turn it on, the screen remains black," the server will extract keywords such as "My PC won't start up" and "The screen is black." Furthermore, it performs context analysis to understand the overall intent of the query.

[0086] Next, the server searches for past correspondence information and technical information in the knowledge base based on the extracted keywords. The databases used here include MySQL (registered trademark) and PostgreSQL. The server uses a high-speed search algorithm to quickly gather the necessary information.

[0087] The server then uses a generative AI model (e.g., GPT-3®) to infer appropriate actions for the query and generate detailed instructions. During this generation process, keywords and contextual information are provided as input. For example, the server might provide the generative AI model with inputs such as "PC won't start" and "black screen," and the model would generate instructions such as "check the power cable connection," "remove and reinsert the battery," and "reset the BIOS."

[0088] Finally, the device presents the inferred action received from the server to the user. Front-end frameworks used here include "React" and "Vue.js," for example. The presented content is processed into a format that is intuitively easy for the user to understand. Specifically, the device displays instructions to the user such as, "Please make sure the power cable is connected properly. Then, remove the battery, wait a few seconds, and then reinsert it."

[0089] As a specific example, assume that a user makes the following inquiry to the service desk:

[0090] "My PC won't start. When I turn it on, the power light comes on, but the screen is black."

[0091] In response to this query, the server generates the following response:

[0092] "Check the power cable connection"

[0093] "Remove and reinsert the battery"

[0094] Please reset the BIOS settings.

[0095] This allows the system to provide quick and consistent responses to user inquiries, improving the efficiency and quality of service desk operations.

[0096] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0097] Step 1:

[0098] When a user makes an inquiry to the service desk, the inquiry is received by a terminal. Here, the inquiry entered by the user is sent to the terminal as text data. For example, suppose a user enters "My PC suddenly stopped starting up. Even when I turn it on, the screen remains black" and presses the send button. This input data is received by the terminal in the form of a string.

[0099] Step 2:

[0100] The terminal transfers the received inquiry data to the server. This process uses the HTTP or HTTPS protocol. Specifically, the terminal sends the user's input to the server as an API request. At this time, metadata such as the user ID and inquiry content are also transferred. The input data is the inquiry content in string format, and the output is the API request received on the server side.

[0101] Step 3:

[0102] The server analyzes the received query data using natural language processing (NLP) technology. NLP libraries such as "SpaCy" and "NLTK" are used for this process. The purpose of the analysis is to extract important keywords and context contained in the query. The input data is the text data of the query received as an API request, and the output is the extracted keywords and analyzed data. A specific example of how this works is when the server extracts keywords such as "My PC won't start" and "The screen is completely black."

[0103] Step 4:

[0104] Next, the server searches for past support information and technical information in the knowledge base based on the extracted keywords. This process uses databases such as "MySQL" and "PostgreSQL." The input data are the extracted keywords, and the output is related technical information and past troubleshooting cases. Specifically, the server searches for technical documents and solutions in the database that contain the keyword "PC won't start."

[0105] Step 5:

[0106] The server uses a generative AI model (for example, GPT-3) to predict the appropriate action for the inquiry and generate detailed instructions. The input data is searched technical information, past solution cases, and extracted keywords, and the output is the predicted action and response procedure. As a specific example of how it works, the server inputs keywords such as "PC won't start" and "black screen" into the generative AI model, and the model generates instructions such as "check the power cable connection," "remove and reinsert the battery," and "reset the BIOS."

[0107] Step 6:

[0108] Finally, the device presents the inferred action received from the server to the user. Front-end frameworks such as "React" and "Vue.js" are used here. The input data are the inferred action and response procedure, and the output is the action instruction displayed to the user. Specifically, the device displays the following instruction to the user: "Please make sure the power cable is connected properly. Then, remove the battery, wait a few seconds, and then reinsert it." In this way, the user can receive the appropriate response procedure in an intuitive and easy-to-understand format.

[0109] (Application example 1)

[0110] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0111] Modern brick-and-mortar stores are required to respond to customer inquiries quickly and accurately. However, because responses depend on the technical knowledge and experience of store staff, there is a problem of inconsistency in the quality and speed of responses. In addition, it is difficult for store staff to immediately come up with appropriate actions, which can lead to a decrease in customer satisfaction.

[0112] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0113] In this invention, the server includes means for receiving an inquiry from a user, means for analyzing the inquiry using natural language processing technology and extracting important keywords, means for searching past response information, solution cases, and general technical information based on the extracted keywords, means for inferring a necessary action from the searched information using a generative AI, means for processing the inferred action and presenting it to a respondent using an intuitive user interface, and means for the respondent to send an answer to the user using the presented action. This enables store staff to quickly determine the appropriate action and respond to customers with consistent quality.

[0114] The "means for receiving inquiries from users" refers to a system or device that receives inquiries when customers make inquiries to the service desk.

[0115] The "means of analyzing using natural language processing technology and extracting important keywords" is a technology for analyzing text-based inquiry data and identifying keywords that indicate the core of the problem.

[0116] "Means for searching past correspondence information, solution cases, and general technical information based on extracted keywords" is a method for efficiently searching for related information from knowledge bases and databases using important keywords.

[0117] "Means for inferring necessary actions using generative AI" refers to a system or algorithm that uses generative AI technology based on collected information to infer specific actions to resolve a user's problem.

[0118] The "means for processing the inferred actions and presenting them to the respondent using an intuitive user interface" refers to a method for converting the inferred actions into a visually easy-to-understand format and presenting them in an easy-to-understand manner to store staff and respondents.

[0119] "Means for responders to send answers to users using suggested actions" refers to a means that uses an intuitive interface to enable staff to quickly take appropriate actions and provide solutions to users.

[0120] The present invention relates to a service desk system for a brick-and-mortar store that responds to inquiries from customers quickly and accurately. Specifically, when a user makes an inquiry to a store service desk, the inquiry is received by a terminal via a smartphone.

[0121] The device forwards the received query to the server, which then analyzes the query using natural language processing technology (e.g., the NLTK library) and extracts important keywords related to the user's problem. For example, if a user enters the query "The cash register isn't working," the server extracts the keywords "cash register" and "not working."

[0122] Next, the server searches knowledge bases (e.g., using Tf-idf vectorization) that store past correspondence information and general technical information based on the extracted keywords. The server searches these databases at high speed to collect related information.

[0123] The server uses generative AI technology (e.g., Hugging Face's Transformers library) based on the collected information to predict the necessary action. For example, in response to a query such as "The cash register isn't working," the server predicts a specific action such as "Please turn the cash register off and then on again."

[0124] The terminal presents the predicted action to the user in an intuitively easy-to-understand user interface, allowing store staff to easily check the presented action and quickly instruct the user on the appropriate action.

[0125] For example, if a user types "The cash register won't work," the server extracts the keywords "cash register" and "not working," searches the knowledge base, and the AI ​​generates detailed instructions based on the information found. The terminal then displays an action to the user, such as "Turn the cash register off and then on again."

[0126] An example of an input prompt for a generative AI model is as follows:

[0127] "The customer has reported that their cash register is not working. Generate a detailed troubleshooting step."

[0128] As a result, the service desk system of the present invention enables quick and consistent responses to customer inquiries and can propose appropriate actions with high accuracy, regardless of the skills or experience of store staff.

[0129] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0130] Step 1:

[0131] A user inputs a query into a terminal. The input query is received by the terminal as character string data. The input at this time is the content of the user's query, and the output is the query received by the terminal.

[0132] Step 2:

[0133] The terminal transfers the received query to the server. The input here is the query data received by the terminal, and the output is the query data transferred to the server. In specific operations, the terminal transmits the query data to the server via network communication.

[0134] Step 3:

[0135] The server receives the query data and analyzes it using natural language processing techniques (e.g., the NLTK library). The input here is the query data transferred to the server, and the output is the important keywords extracted by the analysis. Specific operations include text tokenization and morphological analysis.

[0136] Step 4:

[0137] The server uses the extracted keywords to search a knowledge base that contains past correspondence information, solution cases, and general technical information. The input here is the extracted keywords, and the output is related information obtained from the knowledge base. Specific operations include Tf-idf vectorization and a search algorithm.

[0138] Step 5:

[0139] The server uses generative AI technology (e.g., Hugging Face's Transformers library) to infer the necessary actions based on the collected information. The input here is relevant information collected from a knowledge base, and the output is the specific inferred action. Specific actions include generating prompts for the generative AI model and inputting them into the model.

[0140] Step 6:

[0141] The server sends the predicted action to the terminal, which then presents it to the store staff in an intuitive user interface. The input here is the action sent from the server, and the output is the action displayed on the terminal's display. The specific operation is displayed in a visually easy-to-understand format using UI elements.

[0142] Step 7:

[0143] The store staff instructs the user on the appropriate action based on the content displayed on the terminal. The input here is the action displayed on the terminal display, and the output is a specific solution provided to the user. As a specific action, the staff performs the displayed procedure and conveys the instructions to the user.

[0144] This enables the system to respond quickly and accurately to customer inquiries.

[0145] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0146] Understood. The following is a description of the "Mode for carrying out the invention."

[0147] The present invention relates to a service desk system for quickly and accurately responding to inquiries from users. Specifically, the system combines a means for analyzing received inquiries and predicting appropriate actions with an emotion engine that recognizes user emotions.

[0148] Specifically, the system operates as follows: When a user makes an inquiry to the service desk, the inquiry is received by a terminal. Here, it is assumed that the inquiry is entered as character string data. At the same time, an emotion engine that identifies the user's emotions also operates, recognizing the user's emotional state.

[0149] The device forwards the received query and emotional information to the server. The server uses natural language processing technology to analyze the query and extract important keywords and phrases, while also analyzing the user's emotional state obtained from the emotion engine. For example, if a user enters a query such as "My PC suddenly stopped starting up. Even when I turn it on, the screen remains black," and the emotion engine recognizes that "the user is anxious," the server will extract keywords such as "My PC won't start up" and "The screen is black," and also record the user's emotional state.

[0150] The server then searches past correspondence information, solution cases, and general technical information based on the extracted keywords and emotional state. This uses a knowledge base of many technical information and past troubleshooting records. The server quickly searches these databases to gather the most relevant information.

[0151] The server uses generative AI technology based on the collected information and the user's emotional state to predict the necessary actions and generate detailed instructions. For example, in the case of the above inquiry, the server would predict specific actions such as "check the power cable connection," "remove and reinsert the battery," and "reset the BIOS," and for users who are feeling anxious, would include additional emotionally sensitive instructions such as "please stay calm and follow the steps."

[0152] The next predicted action is presented to the user by the device. At this time, the device processes the predicted action and displays it in an intuitive and easy-to-understand UI. This allows the respondent to easily confirm the presented steps and instruct the user on the appropriate action. Specifically, instructions such as "Make sure the power cable is connected correctly. Next, remove the battery, wait a few seconds, and then reinsert it. Please stay calm and proceed without rushing" are sent to the user.

[0153] In this way, the present invention enables quick and consistent responses to user inquiries, improving the efficiency and quality of service desk operations. Appropriate actions can be suggested with high accuracy, regardless of the skill or experience of the respondent, reducing operating time and quickly resolving problems. Furthermore, the use of an emotion engine enables responses that take into account the user's emotional state, contributing to an improved user experience.

[0154] The processing flow will be explained below.

[0155] Understood. The process flow will be explained in detail below.

[0156] Step 1:

[0157] A user inputs a query into the service desk interface. Data for extracting emotions from the user's voice and text is also input at the same time. For example, the user may input, "My PC suddenly stopped starting up. Even when I turn it on, the screen remains black." The voice tone indicates anxiety.

[0158] Step 2:

[0159] The device receives the user's query and emotion data. The device checks the input data and converts the format. Specifically, it converts the query string data into structured data and extracts emotion data using the emotion engine.

[0160] Step 3:

[0161] The device sends query data and emotion data to the server via an API. The data sent includes the query content and the user's emotional state.

[0162] Step 4:

[0163] The server analyzes the inquiry data it receives. Natural language processing technology is used to extract important keywords and phrases from the inquiry. At this time, an emotion engine is used to analyze the user's emotional state. For example, keywords such as "PC won't start" and "black screen" and the emotional state of "anxiety" are extracted.

[0164] Step 5:

[0165] The server searches for past support information, solution cases, and technical information based on the extracted keywords and the user's emotional state. The server issues a search query to the knowledge base to collect related information. For example, it searches the database for solutions to "PC startup problems" and "screen not displaying."

[0166] Step 6:

[0167] Based on the information collected by the server, a generative AI is used to predict the necessary actions. At this time, the user's emotional state is also taken into consideration, and actions including detailed explanations are generated for users who are feeling particularly anxious. Specifically, in addition to actions such as "check the power cable connection," "remove and reinsert the battery," and "reset the BIOS," the system also includes considerations such as "please remain calm while working."

[0168] Step 7:

[0169] The device receives the inferred action from the server. The device processes this in a format that is easy for the user to see and presents it to the respondent. The UI displays it in an intuitive and easy-to-understand format. For example, instructions such as "1. Make sure the power cable is properly connected. 2. Remove the battery, wait a few seconds, and then reinsert it. 3. Please remain calm as you work." are displayed.

[0170] Step 8:

[0171] The respondent can review the suggested actions displayed on their device, make any necessary adjustments, and then send the final answer to the user, allowing them to work on solving the problem with peace of mind.

[0172] Step 9:

[0173] The user takes specific action as instructed by the respondent. The user completes the steps and checks whether the issue is resolved. If the issue is not resolved, the user makes a new inquiry and starts again with a new step 1.

[0174] Example 2

[0175] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0176] Currently, service desks often provide answers without considering the user's emotional state, and appropriate solutions are needed. Furthermore, because answers depend on the skill and experience of the person answering the questions, it is difficult to provide consistently high-quality service. Therefore, there is a need for a system that can quickly and accurately propose appropriate solutions while taking the user's emotional state into consideration.

[0177] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving an inquiry from a user, means for analyzing the inquiry using natural language processing technology and extracting important keywords, means for transferring the inquiry and the emotional state to the server, means for the server to search past response information, solution cases, and general technical information based on the extracted keywords and the emotional state, means for predicting and generating a required action using a generative AI model based on the searched information, means for processing the predicted and generated action and intuitively presenting it to the user, and means for the user to execute the presented action. This enables consistent, high-quality responses while taking the user's emotional state into consideration.

[0178] "User" refers to any individual or corporation that makes an inquiry using this system.

[0179] An "inquiry" refers to text data such as a question, a trouble report, or an information request that a user sends to the service desk.

[0180] "Terminal" refers to an input device such as a computer, smartphone, or tablet that a user uses to make an inquiry.

[0181] "Server" refers to a central processing unit that receives queries and emotional states sent from terminals, analyzes, searches, and generates results.

[0182] "Natural language processing technology" refers to algorithms and methods for understanding, analyzing, and processing language data.

[0183] "Keywords" refer to words or phrases extracted from inquiries that express the essence of the problem.

[0184] "Emotional state" refers to the psychological state that a user exhibits at the time of inquiry, and includes emotions such as anxiety, anger, and joy.

[0185] "Transfer means" refers to the protocol or method by which the query and emotional state are communicated from the terminal to the server.

[0186] "Knowledge base" refers to a database that stores past response information, solution cases, and general technical information.

[0187] A "generative AI model" refers to a system that uses machine learning and artificial intelligence techniques to generate new information and procedures.

[0188] An "action" refers to a solution or approach to a specific problem.

[0189] "Intuitive presentation means" refers to interfaces and methods for visually displaying information in a way that is easy for users to understand and act on.

[0190] The "means of execution" refers to the procedure or method by which the user actually executes the countermeasure in accordance with the presented action.

[0191] The present invention relates to a system for quickly and accurately responding to user inquiries. Specifically, the system combines a means for analyzing received inquiries and predicting appropriate actions with an emotion engine that recognizes the user's emotions.

[0192] When a user makes an inquiry to the service desk, the inquiry is received by the terminal as string data. The terminal analyzes the content of the inquiry using natural language processing technology (e.g., Natural Language Toolkit (NLTK) or SpaCy) and extracts important keywords. At the same time, it uses an emotion engine (e.g., a general emotion analysis API) to analyze the user's emotional state and identify emotions such as "anxiety," "anger," and "joy."

[0193] The extracted keywords and emotional state are transferred from the terminal to the server. Based on the received data, the server uses a knowledge base search engine (e.g., ElasticSearch® or MICROSOFT® SQL Server) to search past correspondence information, solution cases, and general technical information to collect related information. At this time, a search query is generated based on the extracted keywords and emotional state.

[0194] Furthermore, the server uses a generative AI model (e.g., GPT-4 (registered trademark)) to infer the necessary actions from the collected information and generate specific instructions for the user in natural language. For example, it generates specific actions such as "check the power cable connection," "remove and reinsert the battery," and "reset the BIOS." It also adds explanations that take the user's emotions into consideration. For example, it includes advice such as "stay calm and check the steps step by step."

[0195] The generated results are then processed by the device into an intuitive and easy-to-understand UI and presented to the user, allowing the user to easily understand and carry out the presented procedures.

[0196] As a concrete example, consider a case where a user makes an inquiry saying, "My PC suddenly won't start up. Even when I turn it on, the screen is black." and the emotion engine recognizes that the user is anxious. The server extracts keywords such as "PC won't start up" and "black screen" and identifies the emotional state as "anxiety." Based on this, it searches for related information from the knowledge base and uses a generative AI model to generate the following specific actions:

[0197] Step 1: Check if the power cable is properly connected.

[0198] Step 2: Remove the battery, wait a few seconds, then reinsert it.

[0199] Step 3: Try resetting your BIOS.

[0200] Emotionally sensitive side note: Stay calm and check things out step by step.

[0201] Example prompt sentence:

[0202] "The user says, 'My PC suddenly stopped starting up. Even when I turn it on, the screen is black.' The emotion engine recognizes that the user is anxious. In response to this inquiry, please generate an action that the user should take and an explanation that takes into account their emotions."

[0203] This system improves the efficiency of service desk operations and the user experience by quickly providing appropriate actions while taking into consideration the user's feelings.

[0204] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0205] Step 1:

[0206] A user makes an inquiry to the service desk. The input is text data that the user enters into a browser or a dedicated inquiry form. For example, the user might enter, "My PC suddenly stopped starting up. Even when I turn it on, the screen remains black." This string of data is received by the terminal. Specifically, the user types the inquiry content into the text box and clicks the "Send" button.

[0207] Step 2:

[0208] The terminal analyzes the received inquiry using natural language processing technology. The input here is the text data received in step 1. The terminal uses libraries such as the Natural Language Toolkit (NLTK) and SpaCy to extract important keywords. For example, keywords such as "PC won't start" and "black screen" are extracted. The output is a list of extracted keywords. Specifically, the terminal runs a Python script to perform text analysis.

[0209] Step 3:

[0210] The device uses an emotion engine to analyze the user's emotional state. The input is the text data received in step 1. The emotion engine (a general emotion analysis API) is used to identify the user's emotion from the text. For example, emotion recognition may result in "the user is anxious." The output is the analyzed emotional state. Specifically, the device calls the emotion analysis API and receives the results.

[0211] Step 4:

[0212] The device transfers the extracted keywords and emotional information to the server. The input is the keyword list extracted in step 2 and the emotional state analyzed in step 3. These data are sent to the server. The output is the data transferred to the server. Specifically, the device sends the data to the server using an HTTP POST request.

[0213] Step 5:

[0214] The server searches the knowledge base based on the received data. The input here is the keyword list and emotional state transferred in step 4. The server uses a search engine such as ElasticSearch or Microsoft SQL Server to search past correspondence information, solution cases, and general technical information. The output is a list of related information. Specifically, the server generates a search query and executes the search against the database.

[0215] Step 6:

[0216] The server uses a generative AI model to infer and generate the required actions. The input is the list of relevant information obtained in step 5. A generative AI model (e.g., GPT-4) is used to generate specific steps based on the relevant information. For example, actions such as "check the power cable connection," "remove and reinsert the battery," and "reset the BIOS" are generated. An emotion-sensitive explanation is also generated. The output is the generated steps and the emotion-sensitive explanation. Specifically, the server sends the prompt text to the generative AI model and receives the generated text.

[0217] Step 7:

[0218] The server sends the generated results to the terminal. The input is the procedure and explanation generated in step 6. It sends these data to the terminal. The output is the data sent to the terminal. In concrete terms, the server sends the data to the terminal using an HTTP POST request.

[0219] Step 8:

[0220] The terminal displays the received procedures and explanations in an intuitive and easy-to-understand UI. The input is the data received in step 7. The terminal generates a user interface (UI) and presents the procedures to the user in an easy-to-understand format. The output is the procedures and explanations displayed to the user. Specifically, the terminal uses HTML and JavaScript (registered trademark) to update the user interface.

[0221] In this way, through each processing step, it is possible to respond quickly and accurately to user inquiries.

[0222] (Application example 2)

[0223] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0224] Conventional service desk systems have difficulty in providing prompt and appropriate responses to user inquiries, and have had problems with poor user experience because they do not take the user's emotions into consideration. Therefore, there is a demand for a service desk system that can recognize the user's emotions and provide prompt and appropriate responses.

[0225] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0226] In this invention, the server includes means for receiving an inquiry from a user, means for analyzing the inquiry using natural language processing technology and extracting important keywords, means for searching past response information, solved cases, and general technical information based on the extracted keywords and user emotion information, means for predicting a required action using a generative AI from the searched information and user emotion information, means for processing the predicted action and presenting it to a respondent, means for the respondent to send a response to the user using the presented action, and emotion analysis means for recognizing emotions from a user image. This makes it possible to quickly provide an appropriate response based on the content of the inquiry and the user emotion.

[0227] The "means for receiving inquiries from users" refers to a device or program for receiving information such as questions or requests made by users to the system.

[0228] "Natural language processing technology" is a technology that enables computers to understand, analyze, and generate human language.

[0229] The "means for extracting important keywords" is a device or program for identifying and extracting key words and phrases from the content of a user's inquiry.

[0230] "User emotion information" is data indicating the emotional state of the user when making an inquiry.

[0231] "Past response information" is a record of responses and responses to previous inquiries.

[0232] "Solution cases" are specific examples of problems that have occurred in the past and how they have been solved.

[0233] "General technical information" is information about widely used technologies and methods.

[0234] "Generative AI" is a technology that uses artificial intelligence to generate new information and answers.

[0235] The "means for predicting the necessary action" is a device or program for identifying the appropriate response procedure or process based on the extracted keywords and the user's emotions.

[0236] The "means for presenting to the respondent" is a device or program for visually or audibly presenting the inferred action.

[0237] The "means for transmitting a response to the user" is a device or program for transmitting the generated response procedure or information to the user.

[0238] "Emotion analysis means" refers to a device or program for analyzing and recognizing emotions from a user's facial expressions, voice, text, etc.

[0239] This invention is a service desk system that provides fast and accurate responses to user inquiries. This system combines natural language processing and sentiment analysis technology and uses generative AI models to improve the user experience and streamline problem resolution.

[0240] The system first receives a query from the user. The query is in text format and is sent to the system via a device such as a smartphone or PC. At this stage, image data from the user can also be received at the same time, and this image data will be used for sentiment analysis.

[0241] The received inquiry is analyzed by a natural language processing engine (for example, Google's Dialogflow or IBM's Watson). This analysis extracts important keywords and phrases from the inquiry. For example, from an inquiry such as "The product I ordered hasn't arrived," keywords such as "order," "product," and "didn't arrive" are extracted.

[0242] At the same time, the image data is subjected to emotion analysis using Microsoft Azure's Emotion API. This analysis identifies whether the user is in an emotional state such as anxiety, anger, or joy. For example, the received image may identify the user as "anxious."

[0243] The analyzed keywords and emotion information are transferred to a cloud server, which uses this information to search a knowledge base and obtain past response information, solution examples, general technical information, and other information. This can be achieved by using cloud services such as Amazon's AWS (registered trademark).

[0244] Next, generative AI technology (e.g., OpenAI's GPT-4) uses this information to infer the necessary action. Combining keywords and emotional information, the most appropriate response steps are generated for the user. For example, the following prompt sentences can be used to generate the necessary response steps:

[0245] Inquiry: I haven't received my order, what should I do?

[0246] User Emotion: Anxiety

[0247] Generate appropriate response procedures.

[0248] The generated response procedures are presented to the user in an easy-to-understand manner. This is done through the device's UI. For example, the system may say, "Please check the delivery status. Also, please enter your order number and contact us again. Please remain calm." Additional explanations that take emotions into consideration are also included.

[0249] Ultimately, this system allows users to receive prompt and accurate responses, improving service quality. This system, which uses various hardware and software such as servers, terminals, natural language processing engines, sentiment analysis engines, and generative AI models, will greatly contribute to improving the user experience.

[0250] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0251] Step 1:

[0252] A user enters an inquiry using a device such as a smartphone or PC, and if there is image data, attaches it and sends it to the system. The input at this time is the inquiry content in text format and image data. The output is the raw inquiry data sent to the server.

[0253] Step 2:

[0254] The server passes the received inquiry to a natural language processing engine for analysis. This process extracts important keywords and phrases from the input text. For example, keywords such as "order," "product," and "not delivered" are extracted from the text. The output is a list of keywords as the analysis results.

[0255] Step 3:

[0256] The server passes the received image data to an emotion analysis engine to recognize the user's emotional state. The emotion analysis engine (for example, Microsoft Azure's Emotion API) analyzes the image data and identifies emotions such as "anxiety," "anger," and "joy." This emotional data is returned to the server as output.

[0257] Step 4:

[0258] The server searches a cloud knowledge base based on the extracted keywords and analyzed sentiment data. The knowledge base (e.g., Amazon AWS) stores past response information, solution cases, and general technical information. Highly relevant information is output as the search results.

[0259] Step 5:

[0260] The server combines the knowledge base search results with the user's emotional data and generates a response procedure using a generative AI model (e.g., OpenAI's GPT-4). Specifically, the server inputs the following prompt sentence into the generative AI:

[0261] Inquiry: I haven't received my order, what should I do?

[0262] User Emotion: Anxiety

[0263] Generate appropriate response procedures.

[0264] The generated response procedure is obtained as output.

[0265] Step 6:

[0266] The server transfers the generated response procedures to the terminal, which displays them in an easy-to-understand format for the user. For example, specific instructions such as "Please check the delivery status. Also, please enter your order number and contact us again. Please stay calm." are displayed. The output is the response procedures presented to the user.

[0267] Step 7:

[0268] The user follows the procedure displayed on the terminal and takes action to resolve the problem. The output at this stage is the execution of the action by the user.

[0269] 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 a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the 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.

[0270] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0271] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0272] [Second embodiment]

[0273] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0274] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0275] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0276] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0277] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0278] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0279] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0280] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0281] The specific processing program 56 is an example of a "program" according to the technology of the present 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.

[0282] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0283] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. 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 process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0284] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. 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."

[0285] Understood. The following is a description of the "Mode for carrying out the invention."

[0286] The present invention relates to a service desk system for quickly and accurately responding to inquiries from users, and more specifically, to a system including means for analyzing received inquiries and predicting appropriate actions.

[0287] Specifically, the system operates as follows: When a user makes an inquiry to the service desk, the inquiry is received by a terminal. Here, it is assumed that the inquiry is input as character string data.

[0288] The device forwards the received inquiry to the server, which then analyzes the inquiry using natural language processing technology. The purpose of the analysis is to extract important keywords contained in the inquiry and the intent of the problem. For example, if a user enters an inquiry such as "My PC suddenly stopped starting up. Even when I turn it on, the screen remains black," the server will extract keywords such as "My PC won't start up" and "The screen is black."

[0289] The server then searches for past support information, solution cases, and general technical information based on the extracted keywords. This uses a knowledge base containing a wealth of technical information and past troubleshooting records. The server quickly searches these databases to gather relevant information.

[0290] The server uses generative AI technology based on the collected information to predict the necessary actions and generate detailed instructions. For example, in the case of the above inquiry, the server would predict specific actions such as "check the power cable connection," "remove and reinsert the battery," and "reset the BIOS."

[0291] The next predicted action is presented to the user by the device. At this time, the device processes the predicted action and displays it in an intuitive and easy-to-understand UI. This allows the respondent to easily check the presented steps and instruct the user on the appropriate action. Specifically, instructions such as "Make sure the power cable is connected correctly. Next, remove the battery, wait a few seconds, and then reinsert it" are sent to the user.

[0292] In this way, the present invention enables quick and consistent responses to user inquiries, improving the efficiency and quality of service desk operations. Appropriate actions can be proposed with high accuracy, regardless of the skills or experience of the respondent, reducing operating time and enabling early problem resolution.

[0293] The processing flow will be explained below.

[0294] Step 1:

[0295] A user enters a query into the service desk interface. For example, the user enters, "My PC suddenly stopped starting up. When I turn it on, the screen remains black."

[0296] Step 2:

[0297] The terminal receives a query from a user. The terminal checks the input data and performs format conversion. Specifically, it converts the string data of the query into structured data.

[0298] Step 3:

[0299] The device sends the query data to the server, often via an API.

[0300] Step 4:

[0301] The server analyzes the inquiry data it receives. Natural language processing technology is used to extract important keywords and phrases from the inquiry. For example, important information such as "My PC won't start" or "The screen is black" is extracted here.

[0302] Step 5:

[0303] The server searches past correspondence information, solution cases, and technical information based on the extracted keywords. The server issues search queries to knowledge bases and related databases to extract the most relevant information.

[0304] Step 6:

[0305] The server uses generative AI based on the search results to predict the necessary actions. The generated actions are organized into specific steps. For example, actions such as "check the power cable connection," "remove and reinsert the battery," and "reset the BIOS" are predicted.

[0306] Step 7:

[0307] The device receives the inferred action from the server, processes it in a format that is easy for the user to view, and presents it to the respondent. It is displayed in an intuitive and easy-to-understand format using a UI.

[0308] Step 8:

[0309] The respondent checks the suggested actions displayed on the device and makes any necessary adjustments. The final answer is then sent to the user. For example, specific instructions such as "Make sure the power cable is connected properly. Then remove the battery, wait a few seconds, and then reinsert it."

[0310] Step 9:

[0311] The user performs specific actions as instructed by the respondent, checks whether the results resolve the issue, and if further inquiries are required, returns to step 1.

[0312] Example 1

[0313] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0314] With conventional service desk systems, it was difficult to respond to user inquiries appropriately and quickly, resulting in inconsistent response quality. Furthermore, because responses depended heavily on the skills and experience of the responders, it was difficult to provide efficient and consistent support.

[0315] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0316] In this invention, the server includes means for receiving inquiries from users, means for analyzing the inquiries using natural language processing technology and extracting important keywords, means for searching past response information, solution cases, and general technical information based on the extracted keywords, means for predicting necessary actions using a generative AI model, means for processing the predicted actions and displaying them on a terminal, and means for the terminal to present an answer to the user using the displayed actions. This enables quick and consistent responses, thereby improving the efficiency and quality of service desk operations.

[0317] An "inquiry" is a question or a trouble report that a user sends to the service desk.

[0318] "Natural language processing technology" is a technology that enables computers to understand, analyze, and process human natural language.

[0319] "Keywords" are important words or phrases extracted from a user's query.

[0320] "Past response information" refers to data on problems that the service desk has previously handled and how they were resolved.

[0321] A "solution case" is a specific method or procedure that solves a specific problem.

[0322] "General technical information" refers to information about technologies and methods that are widely and commonly used.

[0323] A "generative AI model" is an artificial intelligence model used to generate appropriate actions or responses to inquiries.

[0324] A "terminal" is a device (such as a computer or smartphone) that a user uses to access the service desk, send an inquiry, and receive a response.

[0325] "Analysis" refers to the process of understanding a user's query and extracting important information.

[0326] "Inference" means that the generative AI model predicts the appropriate action or response to a query.

[0327] A "knowledge base search engine" is a system for searching for necessary information from a database containing past correspondence information and technical information.

[0328] The present invention relates to a service desk system for quickly and accurately responding to inquiries from users, and more specifically, to a system including means for analyzing received inquiries and predicting appropriate actions.

[0329] The program for this system works as follows: When a user makes an inquiry to the service desk, the inquiry is received by a terminal. The terminal obtains the text data sent by the user in the form of a string. The hardware used here includes terminals such as PCs and smartphones, as well as high-performance database servers.

[0330] Next, the received query string data is transferred from the terminal to the server. The communication protocol used here is HTTP or HTTPS. The server analyzes the received query data using natural language processing (NLP). For this process, an NLP library such as "SpaCy" or "NLTK" is used.

[0331] The purpose of the analysis is to extract important keywords and context. For example, if a user enters a query such as "My PC suddenly stopped starting up. Even when I turn it on, the screen remains black," the server will extract keywords such as "My PC won't start up" and "The screen is black." Furthermore, it performs context analysis to understand the overall intent of the query.

[0332] The server then searches for past correspondence information and technical information in a knowledge base based on the extracted keywords. Databases used here include MySQL and PostgreSQL. The server uses high-speed search algorithms to quickly gather the necessary information.

[0333] The server then uses a generative AI model (e.g., GPT-3) to infer appropriate actions for the query and generate detailed instructions. Keywords and contextual information are provided as input during this generation process. For example, the server might provide the generative AI model with inputs like "PC won't start" and "black screen," which would generate instructions like "check the power cable connection," "remove and reinsert the battery," and "reset the BIOS."

[0334] Finally, the device presents the inferred action received from the server to the user. Front-end frameworks used here include "React" and "Vue.js," for example. The presented content is processed into a format that is intuitively easy for the user to understand. Specifically, the device displays instructions to the user such as, "Please make sure the power cable is connected properly. Then, remove the battery, wait a few seconds, and then reinsert it."

[0335] As a specific example, assume that a user makes the following inquiry to the service desk:

[0336] "My PC won't start. When I turn it on, the power light comes on, but the screen is black."

[0337] In response to this query, the server generates the following response:

[0338] "Check the power cable connection"

[0339] "Remove and reinsert the battery"

[0340] Please reset the BIOS settings.

[0341] This allows the system to provide quick and consistent responses to user inquiries, improving the efficiency and quality of service desk operations.

[0342] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0343] Step 1:

[0344] When a user makes an inquiry to the service desk, the inquiry is received by a terminal. Here, the inquiry entered by the user is sent to the terminal as text data. For example, suppose a user enters "My PC suddenly stopped starting up. Even when I turn it on, the screen remains black" and presses the send button. This input data is received by the terminal in the form of a string.

[0345] Step 2:

[0346] The terminal transfers the received inquiry data to the server. This process uses the HTTP or HTTPS protocol. Specifically, the terminal sends the user's input to the server as an API request. At this time, metadata such as the user ID and inquiry content are also transferred. The input data is the inquiry content in string format, and the output is the API request received on the server side.

[0347] Step 3:

[0348] The server analyzes the received query data using natural language processing (NLP) technology. NLP libraries such as "SpaCy" and "NLTK" are used for this process. The purpose of the analysis is to extract important keywords and context contained in the query. The input data is the text data of the query received as an API request, and the output is the extracted keywords and analyzed data. A specific example of how this works is when the server extracts keywords such as "My PC won't start" and "The screen is completely black."

[0349] Step 4:

[0350] Next, the server searches for past support information and technical information in the knowledge base based on the extracted keywords. This process uses databases such as "MySQL" and "PostgreSQL." The input data are the extracted keywords, and the output is related technical information and past troubleshooting cases. Specifically, the server searches for technical documents and solutions in the database that contain the keyword "PC won't start."

[0351] Step 5:

[0352] The server uses a generative AI model (for example, GPT-3) to predict the appropriate action for the inquiry and generate detailed instructions. The input data is searched technical information, past solution cases, and extracted keywords, and the output is the predicted action and response procedure. As a specific example of how it works, the server inputs keywords such as "PC won't start" and "black screen" into the generative AI model, and the model generates instructions such as "check the power cable connection," "remove and reinsert the battery," and "reset the BIOS."

[0353] Step 6:

[0354] Finally, the device presents the inferred action received from the server to the user. Front-end frameworks such as "React" and "Vue.js" are used here. The input data are the inferred action and response procedure, and the output is the action instruction displayed to the user. Specifically, the device displays the following instruction to the user: "Please make sure the power cable is connected properly. Then, remove the battery, wait a few seconds, and then reinsert it." In this way, the user can receive the appropriate response procedure in an intuitive and easy-to-understand format.

[0355] (Application example 1)

[0356] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0357] Modern brick-and-mortar stores are required to respond to customer inquiries quickly and accurately. However, because responses depend on the technical knowledge and experience of store staff, there is a problem of inconsistency in the quality and speed of responses. In addition, it is difficult for store staff to immediately come up with appropriate actions, which can lead to a decrease in customer satisfaction.

[0358] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0359] In this invention, the server includes means for receiving an inquiry from a user, means for analyzing the inquiry using natural language processing technology and extracting important keywords, means for searching past response information, solution cases, and general technical information based on the extracted keywords, means for inferring a necessary action from the searched information using a generative AI, means for processing the inferred action and presenting it to a respondent using an intuitive user interface, and means for the respondent to send an answer to the user using the presented action. This enables store staff to quickly determine the appropriate action and respond to customers with consistent quality.

[0360] The "means for receiving inquiries from users" refers to a system or device that receives inquiries when customers make inquiries to the service desk.

[0361] The "means of analyzing using natural language processing technology and extracting important keywords" is a technology for analyzing text-based inquiry data and identifying keywords that indicate the core of the problem.

[0362] "Means for searching past correspondence information, solution cases, and general technical information based on extracted keywords" is a method for efficiently searching for related information from knowledge bases and databases using important keywords.

[0363] "Means for inferring necessary actions using generative AI" refers to a system or algorithm that uses generative AI technology based on collected information to infer specific actions to resolve a user's problem.

[0364] The "means for processing the inferred actions and presenting them to the respondent using an intuitive user interface" refers to a method for converting the inferred actions into a visually easy-to-understand format and presenting them in an easy-to-understand manner to store staff and respondents.

[0365] "Means for responders to send answers to users using suggested actions" refers to a means that uses an intuitive interface to enable staff to quickly take appropriate actions and provide solutions to users.

[0366] The present invention relates to a service desk system for a brick-and-mortar store that responds to inquiries from customers quickly and accurately. Specifically, when a user makes an inquiry to a store service desk, the inquiry is received by a terminal via a smartphone.

[0367] The device forwards the received query to the server, which then analyzes the query using natural language processing technology (e.g., the NLTK library) and extracts important keywords related to the user's problem. For example, if a user enters the query "The cash register isn't working," the server extracts the keywords "cash register" and "not working."

[0368] Next, the server searches knowledge bases (e.g., using Tf-idf vectorization) that store past correspondence information and general technical information based on the extracted keywords. The server searches these databases at high speed to collect related information.

[0369] The server uses generative AI technology (e.g., Hugging Face's Transformers library) based on the collected information to predict the necessary action. For example, in response to a query such as "The cash register isn't working," the server predicts a specific action such as "Please turn the cash register off and then on again."

[0370] The terminal presents the predicted action to the user in an intuitively easy-to-understand user interface, allowing store staff to easily check the presented action and quickly instruct the user on the appropriate action.

[0371] For example, if a user types "The cash register won't work," the server extracts the keywords "cash register" and "not working," searches the knowledge base, and the AI ​​generates detailed instructions based on the information found. The terminal then displays an action to the user, such as "Turn the cash register off and then on again."

[0372] An example of an input prompt for a generative AI model is as follows:

[0373] "The customer has reported that their cash register is not working. Generate a detailed troubleshooting step."

[0374] As a result, the service desk system of the present invention enables quick and consistent responses to customer inquiries and can propose appropriate actions with high accuracy, regardless of the skills or experience of store staff.

[0375] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0376] Step 1:

[0377] A user inputs a query into a terminal. The input query is received by the terminal as character string data. The input at this time is the content of the user's query, and the output is the query received by the terminal.

[0378] Step 2:

[0379] The terminal transfers the received query to the server. The input here is the query data received by the terminal, and the output is the query data transferred to the server. In specific operations, the terminal transmits the query data to the server via network communication.

[0380] Step 3:

[0381] The server receives the query data and analyzes it using natural language processing techniques (e.g., the NLTK library). The input here is the query data transferred to the server, and the output is the important keywords extracted by the analysis. Specific operations include text tokenization and morphological analysis.

[0382] Step 4:

[0383] The server uses the extracted keywords to search a knowledge base that contains past correspondence information, solution cases, and general technical information. The input here is the extracted keywords, and the output is related information obtained from the knowledge base. Specific operations include Tf-idf vectorization and a search algorithm.

[0384] Step 5:

[0385] The server uses generative AI technology (e.g., Hugging Face's Transformers library) to infer the necessary actions based on the collected information. The input here is relevant information collected from a knowledge base, and the output is the specific inferred action. Specific actions include generating prompts for the generative AI model and inputting them into the model.

[0386] Step 6:

[0387] The server sends the predicted action to the terminal, which then presents it to the store staff in an intuitive user interface. The input here is the action sent from the server, and the output is the action displayed on the terminal's display. The specific operation is displayed in a visually easy-to-understand format using UI elements.

[0388] Step 7:

[0389] The store staff instructs the user on the appropriate action based on the content displayed on the terminal. The input here is the action displayed on the terminal display, and the output is a specific solution provided to the user. As a specific action, the staff performs the displayed procedure and conveys the instructions to the user.

[0390] This enables the system to respond quickly and accurately to customer inquiries.

[0391] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0392] Understood. The following is a description of the "Mode for carrying out the invention."

[0393] The present invention relates to a service desk system for quickly and accurately responding to inquiries from users. Specifically, the system combines a means for analyzing received inquiries and predicting appropriate actions with an emotion engine that recognizes user emotions.

[0394] Specifically, the system operates as follows: When a user makes an inquiry to the service desk, the inquiry is received by a terminal. Here, it is assumed that the inquiry is entered as character string data. At the same time, an emotion engine that identifies the user's emotions also operates, recognizing the user's emotional state.

[0395] The device forwards the received query and emotional information to the server. The server uses natural language processing technology to analyze the query and extract important keywords and phrases, while also analyzing the user's emotional state obtained from the emotion engine. For example, if a user enters a query such as "My PC suddenly stopped starting up. Even when I turn it on, the screen remains black," and the emotion engine recognizes that "the user is anxious," the server will extract keywords such as "My PC won't start up" and "The screen is black," and also record the user's emotional state.

[0396] The server then searches past correspondence information, solution cases, and general technical information based on the extracted keywords and emotional state. This uses a knowledge base of many technical information and past troubleshooting records. The server quickly searches these databases to gather the most relevant information.

[0397] The server uses generative AI technology based on the collected information and the user's emotional state to predict the necessary actions and generate detailed instructions. For example, in the case of the above inquiry, the server would predict specific actions such as "check the power cable connection," "remove and reinsert the battery," and "reset the BIOS," and for users who are feeling anxious, would include additional emotionally sensitive instructions such as "please stay calm and follow the steps."

[0398] The next predicted action is presented to the user by the device. At this time, the device processes the predicted action and displays it in an intuitive and easy-to-understand UI. This allows the respondent to easily confirm the presented steps and instruct the user on the appropriate action. Specifically, instructions such as "Make sure the power cable is connected correctly. Next, remove the battery, wait a few seconds, and then reinsert it. Please stay calm and proceed without rushing" are sent to the user.

[0399] In this way, the present invention enables quick and consistent responses to user inquiries, improving the efficiency and quality of service desk operations. Appropriate actions can be suggested with high accuracy, regardless of the skill or experience of the respondent, reducing operating time and quickly resolving problems. Furthermore, the use of an emotion engine enables responses that take into account the user's emotional state, contributing to an improved user experience.

[0400] The processing flow will be explained below.

[0401] Understood. The process flow will be explained in detail below.

[0402] Step 1:

[0403] A user inputs a query into the service desk interface. Data for extracting emotions from the user's voice and text is also input at the same time. For example, the user may input, "My PC suddenly stopped starting up. Even when I turn it on, the screen remains black." The voice tone indicates anxiety.

[0404] Step 2:

[0405] The device receives the user's query and emotion data. The device checks the input data and converts the format. Specifically, it converts the query string data into structured data and extracts emotion data using the emotion engine.

[0406] Step 3:

[0407] The device sends query data and emotion data to the server via an API. The data sent includes the query content and the user's emotional state.

[0408] Step 4:

[0409] The server analyzes the inquiry data it receives. Natural language processing technology is used to extract important keywords and phrases from the inquiry. At this time, an emotion engine is used to analyze the user's emotional state. For example, keywords such as "PC won't start" and "black screen" and the emotional state of "anxiety" are extracted.

[0410] Step 5:

[0411] The server searches for past support information, solution cases, and technical information based on the extracted keywords and the user's emotional state. The server issues a search query to the knowledge base to collect related information. For example, it searches the database for solutions to "PC startup problems" and "screen not displaying."

[0412] Step 6:

[0413] Based on the information collected by the server, a generative AI is used to predict the necessary actions. At this time, the user's emotional state is also taken into consideration, and actions including detailed explanations are generated for users who are feeling particularly anxious. Specifically, in addition to actions such as "check the power cable connection," "remove and reinsert the battery," and "reset the BIOS," the system also includes considerations such as "please remain calm while working."

[0414] Step 7:

[0415] The device receives the inferred action from the server. The device processes this in a format that is easy for the user to see and presents it to the respondent. The UI displays it in an intuitive and easy-to-understand format. For example, instructions such as "1. Make sure the power cable is properly connected. 2. Remove the battery, wait a few seconds, and then reinsert it. 3. Please remain calm as you work." are displayed.

[0416] Step 8:

[0417] The respondent can review the suggested actions displayed on their device, make any necessary adjustments, and then send the final answer to the user, allowing them to work on solving the problem with peace of mind.

[0418] Step 9:

[0419] The user takes specific action as instructed by the respondent. The user completes the steps and checks whether the issue is resolved. If the issue is not resolved, the user makes a new inquiry and starts again with a new step 1.

[0420] Example 2

[0421] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0422] Currently, service desks often provide answers without considering the user's emotional state, and appropriate solutions are needed. Furthermore, because answers depend on the skill and experience of the person answering the questions, it is difficult to provide consistently high-quality service. Therefore, there is a need for a system that can quickly and accurately propose appropriate solutions while taking the user's emotional state into consideration.

[0423] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving an inquiry from a user, means for analyzing the inquiry using natural language processing technology and extracting important keywords, means for transferring the inquiry and the emotional state to the server, means for the server to search past response information, solution cases, and general technical information based on the extracted keywords and the emotional state, means for predicting and generating a required action using a generative AI model based on the searched information, means for processing the predicted and generated action and intuitively presenting it to the user, and means for the user to execute the presented action. This enables consistent, high-quality responses while taking the user's emotional state into consideration.

[0424] "User" refers to any individual or corporation that makes an inquiry using this system.

[0425] An "inquiry" refers to text data such as a question, a trouble report, or an information request that a user sends to the service desk.

[0426] "Terminal" refers to an input device such as a computer, smartphone, or tablet that a user uses to make an inquiry.

[0427] "Server" refers to a central processing unit that receives queries and emotional states sent from terminals, analyzes, searches, and generates results.

[0428] "Natural language processing technology" refers to algorithms and methods for understanding, analyzing, and processing language data.

[0429] "Keywords" refer to words or phrases extracted from inquiries that express the essence of the problem.

[0430] "Emotional state" refers to the psychological state that a user exhibits at the time of inquiry, and includes emotions such as anxiety, anger, and joy.

[0431] "Transfer means" refers to the protocol or method by which the query and emotional state are communicated from the terminal to the server.

[0432] "Knowledge base" refers to a database that stores past response information, solution cases, and general technical information.

[0433] A "generative AI model" refers to a system that uses machine learning and artificial intelligence techniques to generate new information and procedures.

[0434] An "action" refers to a solution or approach to a specific problem.

[0435] "Intuitive presentation means" refers to interfaces and methods for visually displaying information in a way that is easy for users to understand and act on.

[0436] The "means of execution" refers to the procedure or method by which the user actually executes the countermeasure in accordance with the presented action.

[0437] The present invention relates to a system for quickly and accurately responding to user inquiries. Specifically, the system combines a means for analyzing received inquiries and predicting appropriate actions with an emotion engine that recognizes the user's emotions.

[0438] When a user makes an inquiry to the service desk, the inquiry is received by the terminal as string data. The terminal analyzes the content of the inquiry using natural language processing technology (e.g., Natural Language Toolkit (NLTK) or SpaCy) and extracts important keywords. At the same time, it uses an emotion engine (e.g., a general emotion analysis API) to analyze the user's emotional state and identify emotions such as "anxiety," "anger," and "joy."

[0439] The extracted keywords and emotional state are transferred from the device to the server. Based on the received data, the server uses a knowledge base search engine (e.g., ElasticSearch or Microsoft SQL Server) to search past correspondence information, solution cases, and general technical information to collect related information. At this time, a search query is generated based on the extracted keywords and emotional state.

[0440] Furthermore, the server uses a generative AI model (e.g., GPT-4) to infer the necessary actions from the collected information and generate specific instructions for the user in natural language. For example, it generates specific actions such as "check the power cable connection," "remove and reinsert the battery," and "reset the BIOS." It also adds explanations that take the user's emotions into consideration. For example, it includes advice such as "stay calm and follow the steps step by step."

[0441] The generated results are then processed by the device into an intuitive and easy-to-understand UI and presented to the user, allowing the user to easily understand and carry out the presented procedures.

[0442] As a concrete example, consider a case where a user makes an inquiry saying, "My PC suddenly won't start up. Even when I turn it on, the screen is black." and the emotion engine recognizes that the user is anxious. The server extracts keywords such as "PC won't start up" and "black screen" and identifies the emotional state as "anxiety." Based on this, it searches for related information from the knowledge base and uses a generative AI model to generate the following specific actions:

[0443] Step 1: Check if the power cable is properly connected.

[0444] Step 2: Remove the battery, wait a few seconds, then reinsert it.

[0445] Step 3: Try resetting your BIOS.

[0446] Emotionally sensitive side note: Stay calm and check things out step by step.

[0447] Example prompt sentence:

[0448] "The user says, 'My PC suddenly stopped starting up. Even when I turn it on, the screen is black.' The emotion engine recognizes that the user is anxious. In response to this inquiry, please generate an action that the user should take and an explanation that takes into account their emotions."

[0449] This system improves the efficiency of service desk operations and the user experience by quickly providing appropriate actions while taking into consideration the user's feelings.

[0450] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0451] Step 1:

[0452] A user makes an inquiry to the service desk. The input is text data that the user enters into a browser or a dedicated inquiry form. For example, the user might enter, "My PC suddenly stopped starting up. Even when I turn it on, the screen remains black." This string of data is received by the terminal. Specifically, the user types the inquiry content into the text box and clicks the "Send" button.

[0453] Step 2:

[0454] The terminal analyzes the received inquiry using natural language processing technology. The input here is the text data received in step 1. The terminal uses libraries such as the Natural Language Toolkit (NLTK) and SpaCy to extract important keywords. For example, keywords such as "PC won't start" and "black screen" are extracted. The output is a list of extracted keywords. Specifically, the terminal runs a Python script to perform text analysis.

[0455] Step 3:

[0456] The device uses an emotion engine to analyze the user's emotional state. The input is the text data received in step 1. The emotion engine (a general emotion analysis API) is used to identify the user's emotion from the text. For example, emotion recognition may result in "the user is anxious." The output is the analyzed emotional state. Specifically, the device calls the emotion analysis API and receives the results.

[0457] Step 4:

[0458] The device transfers the extracted keywords and emotional information to the server. The input is the keyword list extracted in step 2 and the emotional state analyzed in step 3. These data are sent to the server. The output is the data transferred to the server. Specifically, the device sends the data to the server using an HTTP POST request.

[0459] Step 5:

[0460] The server searches the knowledge base based on the received data. The input here is the keyword list and emotional state transferred in step 4. The server uses a search engine such as ElasticSearch or Microsoft SQL Server to search past correspondence information, solution cases, and general technical information. The output is a list of related information. Specifically, the server generates a search query and executes the search against the database.

[0461] Step 6:

[0462] The server uses a generative AI model to infer and generate the required actions. The input is the list of relevant information obtained in step 5. A generative AI model (e.g., GPT-4) is used to generate specific steps based on the relevant information. For example, actions such as "check the power cable connection," "remove and reinsert the battery," and "reset the BIOS" are generated. An emotion-sensitive explanation is also generated. The output is the generated steps and the emotion-sensitive explanation. Specifically, the server sends the prompt text to the generative AI model and receives the generated text.

[0463] Step 7:

[0464] The server sends the generated results to the terminal. The input is the procedure and explanation generated in step 6. It sends these data to the terminal. The output is the data sent to the terminal. In concrete terms, the server sends the data to the terminal using an HTTP POST request.

[0465] Step 8:

[0466] The terminal displays the received instructions and explanations in an intuitive and easy-to-understand UI. The input is the data received in step 7. The terminal generates a user interface (UI) and presents the instructions to the user in an easy-to-understand format. The output is the instructions and explanations displayed to the user. Specifically, the terminal uses HTML and JavaScript to update the user interface.

[0467] In this way, through each processing step, it is possible to respond quickly and accurately to user inquiries.

[0468] (Application example 2)

[0469] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0470] Conventional service desk systems have difficulty in providing prompt and appropriate responses to user inquiries, and have had problems with poor user experience because they do not take the user's emotions into consideration. Therefore, there is a demand for a service desk system that can recognize the user's emotions and provide prompt and appropriate responses.

[0471] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0472] In this invention, the server includes means for receiving an inquiry from a user, means for analyzing the inquiry using natural language processing technology and extracting important keywords, means for searching past response information, solved cases, and general technical information based on the extracted keywords and user emotion information, means for predicting a required action using a generative AI from the searched information and user emotion information, means for processing the predicted action and presenting it to a respondent, means for the respondent to send a response to the user using the presented action, and emotion analysis means for recognizing emotions from a user image. This makes it possible to quickly provide an appropriate response based on the content of the inquiry and the user emotion.

[0473] The "means for receiving inquiries from users" refers to a device or program for receiving information such as questions or requests made by users to the system.

[0474] "Natural language processing technology" is a technology that enables computers to understand, analyze, and generate human language.

[0475] The "means for extracting important keywords" is a device or program for identifying and extracting key words and phrases from the content of a user's inquiry.

[0476] "User emotion information" is data indicating the emotional state of the user when making an inquiry.

[0477] "Past response information" is a record of responses and responses to previous inquiries.

[0478] "Solution cases" are specific examples of problems that have occurred in the past and how they have been solved.

[0479] "General technical information" is information about widely used technologies and methods.

[0480] "Generative AI" is a technology that uses artificial intelligence to generate new information and answers.

[0481] The "means for predicting the necessary action" is a device or program for identifying the appropriate response procedure or process based on the extracted keywords and the user's emotions.

[0482] The "means for presenting to the respondent" is a device or program for visually or audibly presenting the inferred action.

[0483] The "means for transmitting a response to the user" is a device or program for transmitting the generated response procedure or information to the user.

[0484] "Emotion analysis means" refers to a device or program for analyzing and recognizing emotions from a user's facial expressions, voice, text, etc.

[0485] This invention is a service desk system that provides fast and accurate responses to user inquiries. This system combines natural language processing and sentiment analysis technology and uses generative AI models to improve the user experience and streamline problem resolution.

[0486] The system first receives a query from the user. The query is in text format and is sent to the system via a device such as a smartphone or PC. At this stage, image data from the user can also be received at the same time, and this image data will be used for sentiment analysis.

[0487] The received inquiry is analyzed by a natural language processing engine (for example, Google's Dialogflow or IBM's Watson). This analysis extracts important keywords and phrases from the inquiry. For example, from an inquiry such as "The product I ordered hasn't arrived," keywords such as "order," "product," and "didn't arrive" are extracted.

[0488] At the same time, the image data is subjected to emotion analysis using Microsoft Azure's Emotion API. This analysis identifies whether the user is in an emotional state such as anxiety, anger, or joy. For example, the received image may identify the user as "anxious."

[0489] The analyzed keywords and sentiment information are then transferred to a cloud server, which uses this information to search a knowledge base and retrieve past support information, case studies, and general technical information, using cloud services such as Amazon's AWS.

[0490] Next, generative AI technology (e.g., OpenAI's GPT-4) uses this information to infer the necessary action. Combining keywords and sentiment information, the most appropriate response steps are generated for the user. For example, the following prompt sentence can be used to generate the necessary response steps:

[0491] Inquiry: I haven't received my order, what should I do?

[0492] User Emotion: Anxiety

[0493] Generate appropriate response procedures.

[0494] The generated response procedures are presented to the user in an easy-to-understand manner. This is done through the device's UI. For example, the system may say, "Please check the delivery status. Also, please enter your order number and contact us again. Please remain calm." Additional explanations that take emotions into consideration are also included.

[0495] Ultimately, this system allows users to receive prompt and accurate responses, improving service quality. This system, which uses various hardware and software such as servers, terminals, natural language processing engines, sentiment analysis engines, and generative AI models, will greatly contribute to improving the user experience.

[0496] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0497] Step 1:

[0498] A user enters an inquiry using a device such as a smartphone or PC, and if there is image data, attaches it and sends it to the system. The input at this time is the inquiry content in text format and image data. The output is the raw inquiry data sent to the server.

[0499] Step 2:

[0500] The server passes the received inquiry to a natural language processing engine for analysis. This process extracts important keywords and phrases from the input text. For example, keywords such as "order," "product," and "not delivered" are extracted from the text. The output is a list of keywords as the analysis results.

[0501] Step 3:

[0502] The server passes the received image data to an emotion analysis engine to recognize the user's emotional state. The emotion analysis engine (for example, Microsoft Azure's Emotion API) analyzes the image data and identifies emotions such as "anxiety," "anger," and "joy." This emotional data is returned to the server as output.

[0503] Step 4:

[0504] The server searches a cloud knowledge base based on the extracted keywords and analyzed sentiment data. The knowledge base (e.g., Amazon AWS) stores past response information, solution cases, and general technical information. Highly relevant information is output as the search results.

[0505] Step 5:

[0506] The server combines the knowledge base search results with the user's emotional data and generates a response procedure using a generative AI model (e.g., OpenAI's GPT-4). Specifically, the server inputs the following prompt sentence into the generative AI:

[0507] Inquiry: I haven't received my order, what should I do?

[0508] User Emotion: Anxiety

[0509] Generate appropriate response procedures.

[0510] The generated response procedure is obtained as output.

[0511] Step 6:

[0512] The server transfers the generated response procedures to the terminal, which displays them in an easy-to-understand format for the user. For example, specific instructions such as "Please check the delivery status. Also, please enter your order number and contact us again. Please stay calm." are displayed. The output is the response procedures presented to the user.

[0513] Step 7:

[0514] The user follows the procedure displayed on the terminal and takes action to resolve the problem. The output at this stage is the execution of the action by the user.

[0515] 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 a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0516] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0517] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0518] [Third embodiment]

[0519] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0520] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0521] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0522] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0523] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0524] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0525] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0526] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0527] The specific processing program 56 is an example of a "program" according to the technology of the present 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.

[0528] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0529] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. 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 process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0530] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. 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."

[0531] Understood. The following is a description of the "Mode for carrying out the invention."

[0532] The present invention relates to a service desk system for quickly and accurately responding to inquiries from users, and more specifically, to a system including means for analyzing received inquiries and predicting appropriate actions.

[0533] Specifically, the system operates as follows: When a user makes an inquiry to the service desk, the inquiry is received by a terminal. Here, it is assumed that the inquiry is input as character string data.

[0534] The device forwards the received inquiry to the server, which then analyzes the inquiry using natural language processing technology. The purpose of the analysis is to extract important keywords contained in the inquiry and the intent of the problem. For example, if a user enters an inquiry such as "My PC suddenly stopped starting up. Even when I turn it on, the screen remains black," the server will extract keywords such as "My PC won't start up" and "The screen is black."

[0535] The server then searches for past support information, solution cases, and general technical information based on the extracted keywords. This uses a knowledge base containing a wealth of technical information and past troubleshooting records. The server quickly searches these databases to gather relevant information.

[0536] The server uses generative AI technology based on the collected information to predict the necessary actions and generate detailed instructions. For example, in the case of the above inquiry, the server would predict specific actions such as "check the power cable connection," "remove and reinsert the battery," and "reset the BIOS."

[0537] The next predicted action is presented to the user by the device. At this time, the device processes the predicted action and displays it in an intuitive and easy-to-understand UI. This allows the respondent to easily check the presented steps and instruct the user on the appropriate action. Specifically, instructions such as "Make sure the power cable is connected correctly. Next, remove the battery, wait a few seconds, and then reinsert it" are sent to the user.

[0538] In this way, the present invention enables quick and consistent responses to user inquiries, improving the efficiency and quality of service desk operations. Appropriate actions can be proposed with high accuracy, regardless of the skills or experience of the respondent, reducing operating time and enabling early problem resolution.

[0539] The processing flow will be explained below.

[0540] Step 1:

[0541] A user enters a query into the service desk interface. For example, the user enters, "My PC suddenly stopped starting up. When I turn it on, the screen remains black."

[0542] Step 2:

[0543] The terminal receives a query from a user. The terminal checks the input data and performs format conversion. Specifically, it converts the string data of the query into structured data.

[0544] Step 3:

[0545] The device sends the query data to the server, often via an API.

[0546] Step 4:

[0547] The server analyzes the inquiry data it receives. Natural language processing technology is used to extract important keywords and phrases from the inquiry. For example, important information such as "My PC won't start" or "The screen is black" is extracted here.

[0548] Step 5:

[0549] The server searches past correspondence information, solution cases, and technical information based on the extracted keywords. The server issues search queries to knowledge bases and related databases to extract the most relevant information.

[0550] Step 6:

[0551] The server uses generative AI based on the search results to predict the necessary actions. The generated actions are organized into specific steps. For example, actions such as "check the power cable connection," "remove and reinsert the battery," and "reset the BIOS" are predicted.

[0552] Step 7:

[0553] The device receives the inferred action from the server, processes it in a format that is easy for the user to view, and presents it to the respondent. It is displayed in an intuitive and easy-to-understand format using a UI.

[0554] Step 8:

[0555] The respondent checks the suggested actions displayed on the device and makes any necessary adjustments. The final answer is then sent to the user. For example, specific instructions such as "Make sure the power cable is connected properly. Then remove the battery, wait a few seconds, and then reinsert it."

[0556] Step 9:

[0557] The user performs specific actions as instructed by the respondent, checks whether the results resolve the issue, and if further inquiries are required, returns to step 1.

[0558] Example 1

[0559] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0560] With conventional service desk systems, it was difficult to respond to user inquiries appropriately and quickly, resulting in inconsistent response quality. Furthermore, because responses depended heavily on the skills and experience of the responders, it was difficult to provide efficient and consistent support.

[0561] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0562] In this invention, the server includes means for receiving inquiries from users, means for analyzing the inquiries using natural language processing technology and extracting important keywords, means for searching past response information, solution cases, and general technical information based on the extracted keywords, means for predicting necessary actions using a generative AI model, means for processing the predicted actions and displaying them on a terminal, and means for the terminal to present an answer to the user using the displayed actions. This enables quick and consistent responses, thereby improving the efficiency and quality of service desk operations.

[0563] An "inquiry" is a question or a trouble report that a user sends to the service desk.

[0564] "Natural language processing technology" is a technology that enables computers to understand, analyze, and process human natural language.

[0565] "Keywords" are important words or phrases extracted from a user's query.

[0566] "Past response information" refers to data on problems that the service desk has previously handled and how they were resolved.

[0567] A "solution case" is a specific method or procedure that solves a specific problem.

[0568] "General technical information" refers to information about technologies and methods that are widely and commonly used.

[0569] A "generative AI model" is an artificial intelligence model used to generate appropriate actions or responses to inquiries.

[0570] A "terminal" is a device (such as a computer or smartphone) that a user uses to access the service desk, send an inquiry, and receive a response.

[0571] "Analysis" refers to the process of understanding a user's query and extracting important information.

[0572] "Inference" means that the generative AI model predicts the appropriate action or response to a query.

[0573] A "knowledge base search engine" is a system for searching for necessary information from a database containing past correspondence information and technical information.

[0574] The present invention relates to a service desk system for quickly and accurately responding to inquiries from users, and more specifically, to a system including means for analyzing received inquiries and predicting appropriate actions.

[0575] The program for this system works as follows: When a user makes an inquiry to the service desk, the inquiry is received by a terminal. The terminal obtains the text data sent by the user in the form of a string. The hardware used here includes terminals such as PCs and smartphones, as well as high-performance database servers.

[0576] Next, the received query string data is transferred from the terminal to the server. The communication protocol used here is HTTP or HTTPS. The server analyzes the received query data using natural language processing (NLP). For this process, an NLP library such as "SpaCy" or "NLTK" is used.

[0577] The purpose of the analysis is to extract important keywords and context. For example, if a user enters a query such as "My PC suddenly stopped starting up. Even when I turn it on, the screen remains black," the server will extract keywords such as "My PC won't start up" and "The screen is black." Furthermore, it performs context analysis to understand the overall intent of the query.

[0578] The server then searches for past correspondence information and technical information in a knowledge base based on the extracted keywords. Databases used here include MySQL and PostgreSQL. The server uses high-speed search algorithms to quickly gather the necessary information.

[0579] The server then uses a generative AI model (e.g., GPT-3) to infer appropriate actions for the query and generate detailed instructions. Keywords and contextual information are provided as input during this generation process. For example, the server might provide the generative AI model with inputs like "PC won't start" and "black screen," which would generate instructions like "check the power cable connection," "remove and reinsert the battery," and "reset the BIOS."

[0580] Finally, the device presents the inferred action received from the server to the user. Front-end frameworks used here include "React" and "Vue.js," for example. The presented content is processed into a format that is intuitively easy for the user to understand. Specifically, the device displays instructions to the user such as, "Please make sure the power cable is connected properly. Then, remove the battery, wait a few seconds, and then reinsert it."

[0581] As a specific example, assume that a user makes the following inquiry to the service desk:

[0582] "My PC won't start. When I turn it on, the power light comes on, but the screen is black."

[0583] In response to this query, the server generates the following response:

[0584] "Check the power cable connection"

[0585] "Remove and reinsert the battery"

[0586] Please reset the BIOS settings.

[0587] This allows the system to provide quick and consistent responses to user inquiries, improving the efficiency and quality of service desk operations.

[0588] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0589] Step 1:

[0590] When a user makes an inquiry to the service desk, the inquiry is received by a terminal. Here, the inquiry entered by the user is sent to the terminal as text data. For example, suppose a user enters "My PC suddenly stopped starting up. Even when I turn it on, the screen remains black" and presses the send button. This input data is received by the terminal in the form of a string.

[0591] Step 2:

[0592] The terminal transfers the received inquiry data to the server. This process uses the HTTP or HTTPS protocol. Specifically, the terminal sends the user's input to the server as an API request. At this time, metadata such as the user ID and inquiry content are also transferred. The input data is the inquiry content in string format, and the output is the API request received on the server side.

[0593] Step 3:

[0594] The server analyzes the received query data using natural language processing (NLP) technology. NLP libraries such as "SpaCy" and "NLTK" are used for this process. The purpose of the analysis is to extract important keywords and context contained in the query. The input data is the text data of the query received as an API request, and the output is the extracted keywords and analyzed data. A specific example of how this works is when the server extracts keywords such as "My PC won't start" and "The screen is completely black."

[0595] Step 4:

[0596] Next, the server searches for past support information and technical information in the knowledge base based on the extracted keywords. This process uses databases such as "MySQL" and "PostgreSQL." The input data are the extracted keywords, and the output is related technical information and past troubleshooting cases. Specifically, the server searches for technical documents and solutions in the database that contain the keyword "PC won't start."

[0597] Step 5:

[0598] The server uses a generative AI model (for example, GPT-3) to predict the appropriate action for the inquiry and generate detailed instructions. The input data is searched technical information, past solution cases, and extracted keywords, and the output is the predicted action and response procedure. As a specific example of how it works, the server inputs keywords such as "PC won't start" and "black screen" into the generative AI model, and the model generates instructions such as "check the power cable connection," "remove and reinsert the battery," and "reset the BIOS."

[0599] Step 6:

[0600] Finally, the device presents the inferred action received from the server to the user. Front-end frameworks such as "React" and "Vue.js" are used here. The input data are the inferred action and response procedure, and the output is the action instruction displayed to the user. Specifically, the device displays the following instruction to the user: "Please make sure the power cable is connected properly. Then, remove the battery, wait a few seconds, and then reinsert it." In this way, the user can receive the appropriate response procedure in an intuitive and easy-to-understand format.

[0601] (Application example 1)

[0602] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0603] Modern brick-and-mortar stores are required to respond to customer inquiries quickly and accurately. However, because responses depend on the technical knowledge and experience of store staff, there is a problem of inconsistency in the quality and speed of responses. In addition, it is difficult for store staff to immediately come up with appropriate actions, which can lead to a decrease in customer satisfaction.

[0604] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0605] In this invention, the server includes means for receiving an inquiry from a user, means for analyzing the inquiry using natural language processing technology and extracting important keywords, means for searching past response information, solution cases, and general technical information based on the extracted keywords, means for inferring a necessary action from the searched information using a generative AI, means for processing the inferred action and presenting it to a respondent using an intuitive user interface, and means for the respondent to send an answer to the user using the presented action. This enables store staff to quickly determine the appropriate action and respond to customers with consistent quality.

[0606] The "means for receiving inquiries from users" refers to a system or device that receives inquiries when customers make inquiries to the service desk.

[0607] The "means of analyzing using natural language processing technology and extracting important keywords" is a technology for analyzing text-based inquiry data and identifying keywords that indicate the core of the problem.

[0608] "Means for searching past correspondence information, solution cases, and general technical information based on extracted keywords" is a method for efficiently searching for related information from knowledge bases and databases using important keywords.

[0609] "Means for inferring necessary actions using generative AI" refers to a system or algorithm that uses generative AI technology based on collected information to infer specific actions to resolve a user's problem.

[0610] The "means for processing the inferred actions and presenting them to the respondent using an intuitive user interface" refers to a method for converting the inferred actions into a visually easy-to-understand format and presenting them in an easy-to-understand manner to store staff and respondents.

[0611] "Means for responders to send answers to users using suggested actions" refers to a means that uses an intuitive interface to enable staff to quickly take appropriate actions and provide solutions to users.

[0612] The present invention relates to a service desk system for a brick-and-mortar store that responds to inquiries from customers quickly and accurately. Specifically, when a user makes an inquiry to a store service desk, the inquiry is received by a terminal via a smartphone.

[0613] The device forwards the received query to the server, which then analyzes the query using natural language processing technology (e.g., the NLTK library) and extracts important keywords related to the user's problem. For example, if a user enters the query "The cash register isn't working," the server extracts the keywords "cash register" and "not working."

[0614] Next, the server searches knowledge bases (e.g., using Tf-idf vectorization) that store past correspondence information and general technical information based on the extracted keywords. The server searches these databases at high speed to collect related information.

[0615] The server uses generative AI technology (e.g., Hugging Face's Transformers library) based on the collected information to predict the necessary action. For example, in response to a query such as "The cash register isn't working," the server predicts a specific action such as "Please turn the cash register off and then on again."

[0616] The terminal presents the predicted action to the user in an intuitively easy-to-understand user interface, allowing store staff to easily check the presented action and quickly instruct the user on the appropriate action.

[0617] For example, if a user types "The cash register won't work," the server extracts the keywords "cash register" and "not working," searches the knowledge base, and the AI ​​generates detailed instructions based on the information found. The terminal then displays an action to the user, such as "Turn the cash register off and then on again."

[0618] An example of an input prompt for a generative AI model is as follows:

[0619] "The customer has reported that their cash register is not working. Generate a detailed troubleshooting step."

[0620] As a result, the service desk system of the present invention enables quick and consistent responses to customer inquiries and can propose appropriate actions with high accuracy, regardless of the skills or experience of store staff.

[0621] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0622] Step 1:

[0623] A user inputs a query into a terminal. The input query is received by the terminal as character string data. The input at this time is the content of the user's query, and the output is the query received by the terminal.

[0624] Step 2:

[0625] The terminal transfers the received query to the server. The input here is the query data received by the terminal, and the output is the query data transferred to the server. In specific operations, the terminal transmits the query data to the server via network communication.

[0626] Step 3:

[0627] The server receives the query data and analyzes it using natural language processing techniques (e.g., the NLTK library). The input here is the query data transferred to the server, and the output is the important keywords extracted by the analysis. Specific operations include text tokenization and morphological analysis.

[0628] Step 4:

[0629] The server uses the extracted keywords to search a knowledge base that contains past correspondence information, solution cases, and general technical information. The input here is the extracted keywords, and the output is related information obtained from the knowledge base. Specific operations include Tf-idf vectorization and a search algorithm.

[0630] Step 5:

[0631] The server uses generative AI technology (e.g., Hugging Face's Transformers library) to infer the necessary actions based on the collected information. The input here is relevant information collected from a knowledge base, and the output is the specific inferred action. Specific actions include generating prompts for the generative AI model and inputting them into the model.

[0632] Step 6:

[0633] The server sends the predicted action to the terminal, which then presents it to the store staff in an intuitive user interface. The input here is the action sent from the server, and the output is the action displayed on the terminal's display. The specific operation is displayed in a visually easy-to-understand format using UI elements.

[0634] Step 7:

[0635] The store staff instructs the user on the appropriate action based on the content displayed on the terminal. The input here is the action displayed on the terminal display, and the output is a specific solution provided to the user. As a specific action, the staff performs the displayed procedure and conveys the instructions to the user.

[0636] This enables the system to respond quickly and accurately to customer inquiries.

[0637] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0638] Understood. The following is a description of the "Mode for carrying out the invention."

[0639] The present invention relates to a service desk system for quickly and accurately responding to inquiries from users. Specifically, the system combines a means for analyzing received inquiries and predicting appropriate actions with an emotion engine that recognizes user emotions.

[0640] Specifically, the system operates as follows: When a user makes an inquiry to the service desk, the inquiry is received by a terminal. Here, it is assumed that the inquiry is entered as character string data. At the same time, an emotion engine that identifies the user's emotions also operates, recognizing the user's emotional state.

[0641] The device forwards the received query and emotional information to the server. The server uses natural language processing technology to analyze the query and extract important keywords and phrases, while also analyzing the user's emotional state obtained from the emotion engine. For example, if a user enters a query such as "My PC suddenly stopped starting up. Even when I turn it on, the screen remains black," and the emotion engine recognizes that "the user is anxious," the server will extract keywords such as "My PC won't start up" and "The screen is black," and also record the user's emotional state.

[0642] The server then searches past correspondence information, solution cases, and general technical information based on the extracted keywords and emotional state. This uses a knowledge base of many technical information and past troubleshooting records. The server quickly searches these databases to gather the most relevant information.

[0643] The server uses generative AI technology based on the collected information and the user's emotional state to predict the necessary actions and generate detailed instructions. For example, in the case of the above inquiry, the server would predict specific actions such as "check the power cable connection," "remove and reinsert the battery," and "reset the BIOS," and for users who are feeling anxious, would include additional emotionally sensitive instructions such as "please stay calm and follow the steps."

[0644] The next predicted action is presented to the user by the device. At this time, the device processes the predicted action and displays it in an intuitive and easy-to-understand UI. This allows the respondent to easily confirm the presented steps and instruct the user on the appropriate action. Specifically, instructions such as "Make sure the power cable is connected correctly. Next, remove the battery, wait a few seconds, and then reinsert it. Please stay calm and proceed without rushing" are sent to the user.

[0645] In this way, the present invention enables quick and consistent responses to user inquiries, improving the efficiency and quality of service desk operations. Appropriate actions can be suggested with high accuracy, regardless of the skill or experience of the respondent, reducing operating time and quickly resolving problems. Furthermore, the use of an emotion engine enables responses that take into account the user's emotional state, contributing to an improved user experience.

[0646] The processing flow will be explained below.

[0647] Understood. The process flow will be explained in detail below.

[0648] Step 1:

[0649] A user inputs a query into the service desk interface. Data for extracting emotions from the user's voice and text is also input at the same time. For example, the user may input, "My PC suddenly stopped starting up. Even when I turn it on, the screen remains black." The voice tone indicates anxiety.

[0650] Step 2:

[0651] The device receives the user's query and emotion data. The device checks the input data and converts the format. Specifically, it converts the query string data into structured data and extracts emotion data using the emotion engine.

[0652] Step 3:

[0653] The device sends query data and emotion data to the server via an API. The data sent includes the query content and the user's emotional state.

[0654] Step 4:

[0655] The server analyzes the inquiry data it receives. Natural language processing technology is used to extract important keywords and phrases from the inquiry. At this time, an emotion engine is used to analyze the user's emotional state. For example, keywords such as "PC won't start" and "black screen" and the emotional state of "anxiety" are extracted.

[0656] Step 5:

[0657] The server searches for past support information, solution cases, and technical information based on the extracted keywords and the user's emotional state. The server issues a search query to the knowledge base to collect related information. For example, it searches the database for solutions to "PC startup problems" and "screen not displaying."

[0658] Step 6:

[0659] Based on the information collected by the server, a generative AI is used to predict the necessary actions. At this time, the user's emotional state is also taken into consideration, and actions including detailed explanations are generated for users who are feeling particularly anxious. Specifically, in addition to actions such as "check the power cable connection," "remove and reinsert the battery," and "reset the BIOS," the system also includes considerations such as "please remain calm while working."

[0660] Step 7:

[0661] The device receives the inferred action from the server. The device processes this in a format that is easy for the user to see and presents it to the respondent. The UI displays it in an intuitive and easy-to-understand format. For example, instructions such as "1. Make sure the power cable is properly connected. 2. Remove the battery, wait a few seconds, and then reinsert it. 3. Please remain calm as you work." are displayed.

[0662] Step 8:

[0663] The respondent can review the suggested actions displayed on their device, make any necessary adjustments, and then send the final answer to the user, allowing them to work on solving the problem with peace of mind.

[0664] Step 9:

[0665] The user takes specific action as instructed by the respondent. The user completes the steps and checks whether the issue is resolved. If the issue is not resolved, the user makes a new inquiry and starts again with a new step 1.

[0666] Example 2

[0667] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0668] Currently, service desks often provide answers without considering the user's emotional state, and appropriate solutions are needed. Furthermore, because answers depend on the skill and experience of the person answering the questions, it is difficult to provide consistently high-quality service. Therefore, there is a need for a system that can quickly and accurately propose appropriate solutions while taking the user's emotional state into consideration.

[0669] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving an inquiry from a user, means for analyzing the inquiry using natural language processing technology and extracting important keywords, means for transferring the inquiry and the emotional state to the server, means for the server to search past response information, solution cases, and general technical information based on the extracted keywords and the emotional state, means for predicting and generating a required action using a generative AI model based on the searched information, means for processing the predicted and generated action and intuitively presenting it to the user, and means for the user to execute the presented action. This enables consistent, high-quality responses while taking the user's emotional state into consideration.

[0670] "User" refers to any individual or corporation that makes an inquiry using this system.

[0671] An "inquiry" refers to text data such as a question, a trouble report, or an information request that a user sends to the service desk.

[0672] "Terminal" refers to an input device such as a computer, smartphone, or tablet that a user uses to make an inquiry.

[0673] "Server" refers to a central processing unit that receives queries and emotional states sent from terminals, analyzes, searches, and generates results.

[0674] "Natural language processing technology" refers to algorithms and methods for understanding, analyzing, and processing language data.

[0675] "Keywords" refer to words or phrases extracted from inquiries that express the essence of the problem.

[0676] "Emotional state" refers to the psychological state that a user exhibits at the time of inquiry, and includes emotions such as anxiety, anger, and joy.

[0677] "Transfer means" refers to the protocol or method by which the query and emotional state are communicated from the terminal to the server.

[0678] "Knowledge base" refers to a database that stores past response information, solution cases, and general technical information.

[0679] A "generative AI model" refers to a system that uses machine learning and artificial intelligence techniques to generate new information and procedures.

[0680] An "action" refers to a solution or approach to a specific problem.

[0681] "Intuitive presentation means" refers to interfaces and methods for visually displaying information in a way that is easy for users to understand and act on.

[0682] The "means of execution" refers to the procedure or method by which the user actually executes the countermeasure in accordance with the presented action.

[0683] The present invention relates to a system for quickly and accurately responding to user inquiries. Specifically, the system combines a means for analyzing received inquiries and predicting appropriate actions with an emotion engine that recognizes the user's emotions.

[0684] When a user makes an inquiry to the service desk, the inquiry is received by the terminal as string data. The terminal analyzes the content of the inquiry using natural language processing technology (e.g., Natural Language Toolkit (NLTK) or SpaCy) and extracts important keywords. At the same time, it uses an emotion engine (e.g., a general emotion analysis API) to analyze the user's emotional state and identify emotions such as "anxiety," "anger," and "joy."

[0685] The extracted keywords and emotional state are transferred from the device to the server. Based on the received data, the server uses a knowledge base search engine (e.g., ElasticSearch or Microsoft SQL Server) to search past correspondence information, solution cases, and general technical information to collect related information. At this time, a search query is generated based on the extracted keywords and emotional state.

[0686] Furthermore, the server uses a generative AI model (e.g., GPT-4) to infer the necessary actions from the collected information and generate specific instructions for the user in natural language. For example, it generates specific actions such as "check the power cable connection," "remove and reinsert the battery," and "reset the BIOS." It also adds explanations that take the user's emotions into consideration. For example, it includes advice such as "stay calm and follow the steps step by step."

[0687] The generated results are then processed by the device into an intuitive and easy-to-understand UI and presented to the user, allowing the user to easily understand and carry out the presented procedures.

[0688] As a concrete example, consider a case where a user makes an inquiry saying, "My PC suddenly won't start up. Even when I turn it on, the screen is black." and the emotion engine recognizes that the user is anxious. The server extracts keywords such as "PC won't start up" and "black screen" and identifies the emotional state as "anxiety." Based on this, it searches for related information from the knowledge base and uses a generative AI model to generate the following specific actions:

[0689] Step 1: Check if the power cable is properly connected.

[0690] Step 2: Remove the battery, wait a few seconds, then reinsert it.

[0691] Step 3: Try resetting your BIOS.

[0692] Emotionally sensitive side note: Stay calm and check things out step by step.

[0693] Example prompt sentence:

[0694] "The user says, 'My PC suddenly stopped starting up. Even when I turn it on, the screen is black.' The emotion engine recognizes that the user is anxious. In response to this inquiry, please generate an action that the user should take and an explanation that takes into account their emotions."

[0695] This system improves the efficiency of service desk operations and the user experience by quickly providing appropriate actions while taking into consideration the user's feelings.

[0696] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0697] Step 1:

[0698] A user makes an inquiry to the service desk. The input is text data that the user enters into a browser or a dedicated inquiry form. For example, the user might enter, "My PC suddenly stopped starting up. Even when I turn it on, the screen remains black." This string of data is received by the terminal. Specifically, the user types the inquiry content into the text box and clicks the "Send" button.

[0699] Step 2:

[0700] The terminal analyzes the received inquiry using natural language processing technology. The input here is the text data received in step 1. The terminal uses libraries such as the Natural Language Toolkit (NLTK) and SpaCy to extract important keywords. For example, keywords such as "PC won't start" and "black screen" are extracted. The output is a list of extracted keywords. Specifically, the terminal runs a Python script to perform text analysis.

[0701] Step 3:

[0702] The device uses an emotion engine to analyze the user's emotional state. The input is the text data received in step 1. The emotion engine (a general emotion analysis API) is used to identify the user's emotion from the text. For example, emotion recognition may result in "the user is anxious." The output is the analyzed emotional state. Specifically, the device calls the emotion analysis API and receives the results.

[0703] Step 4:

[0704] The device transfers the extracted keywords and emotional information to the server. The input is the keyword list extracted in step 2 and the emotional state analyzed in step 3. These data are sent to the server. The output is the data transferred to the server. Specifically, the device sends the data to the server using an HTTP POST request.

[0705] Step 5:

[0706] The server searches the knowledge base based on the received data. The input here is the keyword list and emotional state transferred in step 4. The server uses a search engine such as ElasticSearch or Microsoft SQL Server to search past correspondence information, solution cases, and general technical information. The output is a list of related information. Specifically, the server generates a search query and executes the search against the database.

[0707] Step 6:

[0708] The server uses a generative AI model to infer and generate the required actions. The input is the list of relevant information obtained in step 5. A generative AI model (e.g., GPT-4) is used to generate specific steps based on the relevant information. For example, actions such as "check the power cable connection," "remove and reinsert the battery," and "reset the BIOS" are generated. An emotion-sensitive explanation is also generated. The output is the generated steps and the emotion-sensitive explanation. Specifically, the server sends the prompt text to the generative AI model and receives the generated text.

[0709] Step 7:

[0710] The server sends the generated results to the terminal. The input is the procedure and explanation generated in step 6. It sends these data to the terminal. The output is the data sent to the terminal. In concrete terms, the server sends the data to the terminal using an HTTP POST request.

[0711] Step 8:

[0712] The terminal displays the received instructions and explanations in an intuitive and easy-to-understand UI. The input is the data received in step 7. The terminal generates a user interface (UI) and presents the instructions to the user in an easy-to-understand format. The output is the instructions and explanations displayed to the user. Specifically, the terminal uses HTML and JavaScript to update the user interface.

[0713] In this way, through each processing step, it is possible to respond quickly and accurately to user inquiries.

[0714] (Application example 2)

[0715] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0716] Conventional service desk systems have difficulty in providing prompt and appropriate responses to user inquiries, and have had problems with poor user experience because they do not take the user's emotions into consideration. Therefore, there is a demand for a service desk system that can recognize the user's emotions and provide prompt and appropriate responses.

[0717] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0718] In this invention, the server includes means for receiving an inquiry from a user, means for analyzing the inquiry using natural language processing technology and extracting important keywords, means for searching past response information, solved cases, and general technical information based on the extracted keywords and user emotion information, means for predicting a required action using a generative AI from the searched information and user emotion information, means for processing the predicted action and presenting it to a respondent, means for the respondent to send a response to the user using the presented action, and emotion analysis means for recognizing emotions from a user image. This makes it possible to quickly provide an appropriate response based on the content of the inquiry and the user emotion.

[0719] The "means for receiving inquiries from users" refers to a device or program for receiving information such as questions or requests made by users to the system.

[0720] "Natural language processing technology" is a technology that enables computers to understand, analyze, and generate human language.

[0721] The "means for extracting important keywords" is a device or program for identifying and extracting key words and phrases from the content of a user's inquiry.

[0722] "User emotion information" is data indicating the emotional state of the user when making an inquiry.

[0723] "Past response information" is a record of responses and responses to previous inquiries.

[0724] "Solution cases" are specific examples of problems that have occurred in the past and how they have been solved.

[0725] "General technical information" is information about widely used technologies and methods.

[0726] "Generative AI" is a technology that uses artificial intelligence to generate new information and answers.

[0727] The "means for predicting the necessary action" is a device or program for identifying the appropriate response procedure or process based on the extracted keywords and the user's emotions.

[0728] The "means for presenting to the respondent" is a device or program for visually or audibly presenting the inferred action.

[0729] The "means for transmitting a response to the user" is a device or program for transmitting the generated response procedure or information to the user.

[0730] "Emotion analysis means" refers to a device or program for analyzing and recognizing emotions from a user's facial expressions, voice, text, etc.

[0731] This invention is a service desk system that provides fast and accurate responses to user inquiries. This system combines natural language processing and sentiment analysis technology and uses generative AI models to improve the user experience and streamline problem resolution.

[0732] The system first receives a query from the user. The query is in text format and is sent to the system via a device such as a smartphone or PC. At this stage, image data from the user can also be received at the same time, and this image data will be used for sentiment analysis.

[0733] The received inquiry is analyzed by a natural language processing engine (for example, Google's Dialogflow or IBM's Watson). This analysis extracts important keywords and phrases from the inquiry. For example, from an inquiry such as "The product I ordered hasn't arrived," keywords such as "order," "product," and "didn't arrive" are extracted.

[0734] At the same time, the image data is subjected to emotion analysis using Microsoft Azure's Emotion API. This analysis identifies whether the user is in an emotional state such as anxiety, anger, or joy. For example, the received image may identify the user as "anxious."

[0735] The analyzed keywords and sentiment information are then transferred to a cloud server, which uses this information to search a knowledge base and retrieve past support information, case studies, and general technical information, using cloud services such as Amazon's AWS.

[0736] Next, generative AI technology (e.g., OpenAI's GPT-4) uses this information to infer the necessary action. Combining keywords and sentiment information, the most appropriate response steps are generated for the user. For example, the following prompt sentence can be used to generate the necessary response steps:

[0737] Inquiry: I haven't received my order, what should I do?

[0738] User Emotion: Anxiety

[0739] Generate appropriate response procedures.

[0740] The generated response procedures are presented to the user in an easy-to-understand manner. This is done through the device's UI. For example, the system may say, "Please check the delivery status. Also, please enter your order number and contact us again. Please remain calm." Additional explanations that take emotions into consideration are also included.

[0741] Ultimately, this system allows users to receive prompt and accurate responses, improving service quality. This system, which uses various hardware and software such as servers, terminals, natural language processing engines, sentiment analysis engines, and generative AI models, will greatly contribute to improving the user experience.

[0742] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0743] Step 1:

[0744] A user enters an inquiry using a device such as a smartphone or PC, and if there is image data, attaches it and sends it to the system. The input at this time is the inquiry content in text format and image data. The output is the raw inquiry data sent to the server.

[0745] Step 2:

[0746] The server passes the received inquiry to a natural language processing engine for analysis. This process extracts important keywords and phrases from the input text. For example, keywords such as "order," "product," and "not delivered" are extracted from the text. The output is a list of keywords as the analysis results.

[0747] Step 3:

[0748] The server passes the received image data to an emotion analysis engine to recognize the user's emotional state. The emotion analysis engine (for example, Microsoft Azure's Emotion API) analyzes the image data and identifies emotions such as "anxiety," "anger," and "joy." This emotional data is returned to the server as output.

[0749] Step 4:

[0750] The server searches a cloud knowledge base based on the extracted keywords and analyzed sentiment data. The knowledge base (e.g., Amazon AWS) stores past response information, solution cases, and general technical information. Highly relevant information is output as the search results.

[0751] Step 5:

[0752] The server combines the knowledge base search results with the user's emotional data and generates a response procedure using a generative AI model (e.g., OpenAI's GPT-4). Specifically, the server inputs the following prompt sentence into the generative AI:

[0753] Inquiry: I haven't received my order, what should I do?

[0754] User Emotion: Anxiety

[0755] Generate appropriate response procedures.

[0756] The generated response procedure is obtained as output.

[0757] Step 6:

[0758] The server transfers the generated response procedures to the terminal, which displays them in an easy-to-understand format for the user. For example, specific instructions such as "Please check the delivery status. Also, please enter your order number and contact us again. Please stay calm." are displayed. The output is the response procedures presented to the user.

[0759] Step 7:

[0760] The user follows the procedure displayed on the terminal and takes action to resolve the problem. The output at this stage is the execution of the action by the user.

[0761] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0762] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0763] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[0764] [Fourth embodiment]

[0765] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0766] 7, a 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.

[0767] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0768] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[0769] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0770] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0771] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0772] The control object 443 includes a display device, LEDs in the eyes, and motors for driving 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 emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[0773] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0774] The specific processing program 56 is an example of a "program" according to the technology of the present 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.

[0775] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0776] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. 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 process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0777] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[0778] Understood. The following is a description of the "Mode for carrying out the invention."

[0779] The present invention relates to a service desk system for quickly and accurately responding to inquiries from users, and more specifically, to a system including means for analyzing received inquiries and predicting appropriate actions.

[0780] Specifically, the system operates as follows: When a user makes an inquiry to the service desk, the inquiry is received by a terminal. Here, it is assumed that the inquiry is input as character string data.

[0781] The device forwards the received inquiry to the server, which then analyzes the inquiry using natural language processing technology. The purpose of the analysis is to extract important keywords contained in the inquiry and the intent of the problem. For example, if a user enters an inquiry such as "My PC suddenly stopped starting up. Even when I turn it on, the screen remains black," the server will extract keywords such as "My PC won't start up" and "The screen is black."

[0782] The server then searches for past support information, solution cases, and general technical information based on the extracted keywords. This uses a knowledge base containing a wealth of technical information and past troubleshooting records. The server quickly searches these databases to gather relevant information.

[0783] The server uses generative AI technology based on the collected information to predict the necessary actions and generate detailed instructions. For example, in the case of the above inquiry, the server would predict specific actions such as "check the power cable connection," "remove and reinsert the battery," and "reset the BIOS."

[0784] The next predicted action is presented to the user by the device. At this time, the device processes the predicted action and displays it in an intuitive and easy-to-understand UI. This allows the respondent to easily check the presented steps and instruct the user on the appropriate action. Specifically, instructions such as "Make sure the power cable is connected correctly. Next, remove the battery, wait a few seconds, and then reinsert it" are sent to the user.

[0785] In this way, the present invention enables quick and consistent responses to user inquiries, improving the efficiency and quality of service desk operations. Appropriate actions can be proposed with high accuracy, regardless of the skills or experience of the respondent, reducing operating time and enabling early problem resolution.

[0786] The processing flow will be explained below.

[0787] Step 1:

[0788] A user enters a query into the service desk interface. For example, the user enters, "My PC suddenly stopped starting up. When I turn it on, the screen remains black."

[0789] Step 2:

[0790] The terminal receives a query from a user. The terminal checks the input data and performs format conversion. Specifically, it converts the string data of the query into structured data.

[0791] Step 3:

[0792] The device sends the query data to the server, often via an API.

[0793] Step 4:

[0794] The server analyzes the inquiry data it receives. Natural language processing technology is used to extract important keywords and phrases from the inquiry. For example, important information such as "My PC won't start" or "The screen is black" is extracted here.

[0795] Step 5:

[0796] The server searches past correspondence information, solution cases, and technical information based on the extracted keywords. The server issues search queries to knowledge bases and related databases to extract the most relevant information.

[0797] Step 6:

[0798] The server uses generative AI based on the search results to predict the necessary actions. The generated actions are organized into specific steps. For example, actions such as "check the power cable connection," "remove and reinsert the battery," and "reset the BIOS" are predicted.

[0799] Step 7:

[0800] The device receives the inferred action from the server, processes it in a format that is easy for the user to view, and presents it to the respondent. It is displayed in an intuitive and easy-to-understand format using a UI.

[0801] Step 8:

[0802] The respondent checks the suggested actions displayed on the device and makes any necessary adjustments. The final answer is then sent to the user. For example, specific instructions such as "Make sure the power cable is connected properly. Then remove the battery, wait a few seconds, and then reinsert it."

[0803] Step 9:

[0804] The user performs specific actions as instructed by the respondent, checks whether the results resolve the issue, and if further inquiries are required, returns to step 1.

[0805] Example 1

[0806] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[0807] With conventional service desk systems, it was difficult to respond to user inquiries appropriately and quickly, resulting in inconsistent response quality. Furthermore, because responses depended heavily on the skills and experience of the responders, it was difficult to provide efficient and consistent support.

[0808] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0809] In this invention, the server includes means for receiving inquiries from users, means for analyzing the inquiries using natural language processing technology and extracting important keywords, means for searching past response information, solution cases, and general technical information based on the extracted keywords, means for predicting necessary actions using a generative AI model, means for processing the predicted actions and displaying them on a terminal, and means for the terminal to present an answer to the user using the displayed actions. This enables quick and consistent responses, thereby improving the efficiency and quality of service desk operations.

[0810] An "inquiry" is a question or a trouble report that a user sends to the service desk.

[0811] "Natural language processing technology" is a technology that enables computers to understand, analyze, and process human natural language.

[0812] "Keywords" are important words or phrases extracted from a user's query.

[0813] "Past response information" refers to data on problems that the service desk has previously handled and how they were resolved.

[0814] A "solution case" is a specific method or procedure that solves a specific problem.

[0815] "General technical information" refers to information about technologies and methods that are widely and commonly used.

[0816] A "generative AI model" is an artificial intelligence model used to generate appropriate actions or responses to inquiries.

[0817] A "terminal" is a device (such as a computer or smartphone) that a user uses to access the service desk, send an inquiry, and receive a response.

[0818] "Analysis" refers to the process of understanding a user's query and extracting important information.

[0819] "Inference" means that the generative AI model predicts the appropriate action or response to a query.

[0820] A "knowledge base search engine" is a system for searching for necessary information from a database containing past correspondence information and technical information.

[0821] The present invention relates to a service desk system for quickly and accurately responding to inquiries from users, and more specifically, to a system including means for analyzing received inquiries and predicting appropriate actions.

[0822] The program for this system works as follows: When a user makes an inquiry to the service desk, the inquiry is received by a terminal. The terminal obtains the text data sent by the user in the form of a string. The hardware used here includes terminals such as PCs and smartphones, as well as high-performance database servers.

[0823] Next, the received query string data is transferred from the terminal to the server. The communication protocol used here is HTTP or HTTPS. The server analyzes the received query data using natural language processing (NLP). For this process, an NLP library such as "SpaCy" or "NLTK" is used.

[0824] The purpose of the analysis is to extract important keywords and context. For example, if a user enters a query such as "My PC suddenly stopped starting up. Even when I turn it on, the screen remains black," the server will extract keywords such as "My PC won't start up" and "The screen is black." Furthermore, it performs context analysis to understand the overall intent of the query.

[0825] The server then searches for past correspondence information and technical information in a knowledge base based on the extracted keywords. Databases used here include MySQL and PostgreSQL. The server uses high-speed search algorithms to quickly gather the necessary information.

[0826] The server then uses a generative AI model (e.g., GPT-3) to infer appropriate actions for the query and generate detailed instructions. Keywords and contextual information are provided as input during this generation process. For example, the server might provide the generative AI model with inputs like "PC won't start" and "black screen," which would generate instructions like "check the power cable connection," "remove and reinsert the battery," and "reset the BIOS."

[0827] Finally, the device presents the inferred action received from the server to the user. Front-end frameworks used here include "React" and "Vue.js," for example. The presented content is processed into a format that is intuitively easy for the user to understand. Specifically, the device displays instructions to the user such as, "Please make sure the power cable is connected properly. Then, remove the battery, wait a few seconds, and then reinsert it."

[0828] As a specific example, assume that a user makes the following inquiry to the service desk:

[0829] "My PC won't start. When I turn it on, the power light comes on, but the screen is black."

[0830] In response to this query, the server generates the following response:

[0831] "Check the power cable connection"

[0832] "Remove and reinsert the battery"

[0833] Please reset the BIOS settings.

[0834] This allows the system to provide quick and consistent responses to user inquiries, improving the efficiency and quality of service desk operations.

[0835] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0836] Step 1:

[0837] When a user makes an inquiry to the service desk, the inquiry is received by a terminal. Here, the inquiry entered by the user is sent to the terminal as text data. For example, suppose a user enters "My PC suddenly stopped starting up. Even when I turn it on, the screen remains black" and presses the send button. This input data is received by the terminal in the form of a string.

[0838] Step 2:

[0839] The terminal transfers the received inquiry data to the server. This process uses the HTTP or HTTPS protocol. Specifically, the terminal sends the user's input to the server as an API request. At this time, metadata such as the user ID and inquiry content are also transferred. The input data is the inquiry content in string format, and the output is the API request received on the server side.

[0840] Step 3:

[0841] The server analyzes the received query data using natural language processing (NLP) technology. NLP libraries such as "SpaCy" and "NLTK" are used for this process. The purpose of the analysis is to extract important keywords and context contained in the query. The input data is the text data of the query received as an API request, and the output is the extracted keywords and analyzed data. A specific example of how this works is when the server extracts keywords such as "My PC won't start" and "The screen is completely black."

[0842] Step 4:

[0843] Next, the server searches for past support information and technical information in the knowledge base based on the extracted keywords. This process uses databases such as "MySQL" and "PostgreSQL." The input data are the extracted keywords, and the output is related technical information and past troubleshooting cases. Specifically, the server searches for technical documents and solutions in the database that contain the keyword "PC won't start."

[0844] Step 5:

[0845] The server uses a generative AI model (for example, GPT-3) to predict the appropriate action for the inquiry and generate detailed instructions. The input data is searched technical information, past solution cases, and extracted keywords, and the output is the predicted action and response procedure. As a specific example of how it works, the server inputs keywords such as "PC won't start" and "black screen" into the generative AI model, and the model generates instructions such as "check the power cable connection," "remove and reinsert the battery," and "reset the BIOS."

[0846] Step 6:

[0847] Finally, the device presents the inferred action received from the server to the user. Front-end frameworks such as "React" and "Vue.js" are used here. The input data are the inferred action and response procedure, and the output is the action instruction displayed to the user. Specifically, the device displays the following instruction to the user: "Please make sure the power cable is connected properly. Then, remove the battery, wait a few seconds, and then reinsert it." In this way, the user can receive the appropriate response procedure in an intuitive and easy-to-understand format.

[0848] (Application example 1)

[0849] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[0850] Modern brick-and-mortar stores are required to respond to customer inquiries quickly and accurately. However, because responses depend on the technical knowledge and experience of store staff, there is a problem of inconsistency in the quality and speed of responses. In addition, it is difficult for store staff to immediately come up with appropriate actions, which can lead to a decrease in customer satisfaction.

[0851] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0852] In this invention, the server includes means for receiving an inquiry from a user, means for analyzing the inquiry using natural language processing technology and extracting important keywords, means for searching past response information, solution cases, and general technical information based on the extracted keywords, means for inferring a necessary action from the searched information using a generative AI, means for processing the inferred action and presenting it to a respondent using an intuitive user interface, and means for the respondent to send an answer to the user using the presented action. This enables store staff to quickly determine the appropriate action and respond to customers with consistent quality.

[0853] The "means for receiving inquiries from users" refers to a system or device that receives inquiries when customers make inquiries to the service desk.

[0854] The "means of analyzing using natural language processing technology and extracting important keywords" is a technology for analyzing text-based inquiry data and identifying keywords that indicate the core of the problem.

[0855] "Means for searching past correspondence information, solution cases, and general technical information based on extracted keywords" is a method for efficiently searching for related information from knowledge bases and databases using important keywords.

[0856] "Means for inferring necessary actions using generative AI" refers to a system or algorithm that uses generative AI technology based on collected information to infer specific actions to resolve a user's problem.

[0857] The "means for processing the inferred actions and presenting them to the respondent using an intuitive user interface" refers to a method for converting the inferred actions into a visually easy-to-understand format and presenting them in an easy-to-understand manner to store staff and respondents.

[0858] "Means for responders to send answers to users using suggested actions" refers to a means that uses an intuitive interface to enable staff to quickly take appropriate actions and provide solutions to users.

[0859] The present invention relates to a service desk system for a brick-and-mortar store that responds to inquiries from customers quickly and accurately. Specifically, when a user makes an inquiry to a store service desk, the inquiry is received by a terminal via a smartphone.

[0860] The device forwards the received query to the server, which then analyzes the query using natural language processing technology (e.g., the NLTK library) and extracts important keywords related to the user's problem. For example, if a user enters the query "The cash register isn't working," the server extracts the keywords "cash register" and "not working."

[0861] Next, the server searches knowledge bases (e.g., using Tf-idf vectorization) that store past correspondence information and general technical information based on the extracted keywords. The server searches these databases at high speed to collect related information.

[0862] The server uses generative AI technology (e.g., Hugging Face's Transformers library) based on the collected information to predict the necessary action. For example, in response to a query such as "The cash register isn't working," the server predicts a specific action such as "Please turn the cash register off and then on again."

[0863] The terminal presents the predicted action to the user in an intuitively easy-to-understand user interface, allowing store staff to easily check the presented action and quickly instruct the user on the appropriate action.

[0864] For example, if a user types "The cash register won't work," the server extracts the keywords "cash register" and "not working," searches the knowledge base, and the AI ​​generates detailed instructions based on the information found. The terminal then displays an action to the user, such as "Turn the cash register off and then on again."

[0865] An example of an input prompt for a generative AI model is as follows:

[0866] "The customer has reported that their cash register is not working. Generate a detailed troubleshooting step."

[0867] As a result, the service desk system of the present invention enables quick and consistent responses to customer inquiries and can propose appropriate actions with high accuracy, regardless of the skills or experience of store staff.

[0868] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0869] Step 1:

[0870] A user inputs a query into a terminal. The input query is received by the terminal as character string data. The input at this time is the content of the user's query, and the output is the query received by the terminal.

[0871] Step 2:

[0872] The terminal transfers the received query to the server. The input here is the query data received by the terminal, and the output is the query data transferred to the server. In specific operations, the terminal transmits the query data to the server via network communication.

[0873] Step 3:

[0874] The server receives the query data and analyzes it using natural language processing techniques (e.g., the NLTK library). The input here is the query data transferred to the server, and the output is the important keywords extracted by the analysis. Specific operations include text tokenization and morphological analysis.

[0875] Step 4:

[0876] The server uses the extracted keywords to search a knowledge base that contains past correspondence information, solution cases, and general technical information. The input here is the extracted keywords, and the output is related information obtained from the knowledge base. Specific operations include Tf-idf vectorization and a search algorithm.

[0877] Step 5:

[0878] The server uses generative AI technology (e.g., Hugging Face's Transformers library) to infer the necessary actions based on the collected information. The input here is relevant information collected from a knowledge base, and the output is the specific inferred action. Specific actions include generating prompts for the generative AI model and inputting them into the model.

[0879] Step 6:

[0880] The server sends the predicted action to the terminal, which then presents it to the store staff in an intuitive user interface. The input here is the action sent from the server, and the output is the action displayed on the terminal's display. The specific operation is displayed in a visually easy-to-understand format using UI elements.

[0881] Step 7:

[0882] The store staff instructs the user on the appropriate action based on the content displayed on the terminal. The input here is the action displayed on the terminal display, and the output is a specific solution provided to the user. As a specific action, the staff performs the displayed procedure and conveys the instructions to the user.

[0883] This enables the system to respond quickly and accurately to customer inquiries.

[0884] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0885] Understood. The following is a description of the "Mode for carrying out the invention."

[0886] The present invention relates to a service desk system for quickly and accurately responding to inquiries from users. Specifically, the system combines a means for analyzing received inquiries and predicting appropriate actions with an emotion engine that recognizes user emotions.

[0887] Specifically, the system operates as follows: When a user makes an inquiry to the service desk, the inquiry is received by a terminal. Here, it is assumed that the inquiry is entered as character string data. At the same time, an emotion engine that identifies the user's emotions also operates, recognizing the user's emotional state.

[0888] The device forwards the received query and emotional information to the server. The server uses natural language processing technology to analyze the query and extract important keywords and phrases, while also analyzing the user's emotional state obtained from the emotion engine. For example, if a user enters a query such as "My PC suddenly stopped starting up. Even when I turn it on, the screen remains black," and the emotion engine recognizes that "the user is anxious," the server will extract keywords such as "My PC won't start up" and "The screen is black," and also record the user's emotional state.

[0889] The server then searches past correspondence information, solution cases, and general technical information based on the extracted keywords and emotional state. This uses a knowledge base of many technical information and past troubleshooting records. The server quickly searches these databases to gather the most relevant information.

[0890] The server uses generative AI technology based on the collected information and the user's emotional state to predict the necessary actions and generate detailed instructions. For example, in the case of the above inquiry, the server would predict specific actions such as "check the power cable connection," "remove and reinsert the battery," and "reset the BIOS," and for users who are feeling anxious, would include additional emotionally sensitive instructions such as "please stay calm and follow the steps."

[0891] The next predicted action is presented to the user by the device. At this time, the device processes the predicted action and displays it in an intuitive and easy-to-understand UI. This allows the respondent to easily confirm the presented steps and instruct the user on the appropriate action. Specifically, instructions such as "Make sure the power cable is connected correctly. Next, remove the battery, wait a few seconds, and then reinsert it. Please stay calm and proceed without rushing" are sent to the user.

[0892] In this way, the present invention enables quick and consistent responses to user inquiries, improving the efficiency and quality of service desk operations. Appropriate actions can be suggested with high accuracy, regardless of the skill or experience of the respondent, reducing operating time and quickly resolving problems. Furthermore, the use of an emotion engine enables responses that take into account the user's emotional state, contributing to an improved user experience.

[0893] The processing flow will be explained below.

[0894] Understood. The process flow will be explained in detail below.

[0895] Step 1:

[0896] A user inputs a query into the service desk interface. Data for extracting emotions from the user's voice and text is also input at the same time. For example, the user may input, "My PC suddenly stopped starting up. Even when I turn it on, the screen remains black." The voice tone indicates anxiety.

[0897] Step 2:

[0898] The device receives the user's query and emotion data. The device checks the input data and converts the format. Specifically, it converts the query string data into structured data and extracts emotion data using the emotion engine.

[0899] Step 3:

[0900] The device sends query data and emotion data to the server via an API. The data sent includes the query content and the user's emotional state.

[0901] Step 4:

[0902] The server analyzes the inquiry data it receives. Natural language processing technology is used to extract important keywords and phrases from the inquiry. At this time, an emotion engine is used to analyze the user's emotional state. For example, keywords such as "PC won't start" and "black screen" and the emotional state of "anxiety" are extracted.

[0903] Step 5:

[0904] The server searches for past support information, solution cases, and technical information based on the extracted keywords and the user's emotional state. The server issues a search query to the knowledge base to collect related information. For example, it searches the database for solutions to "PC startup problems" and "screen not displaying."

[0905] Step 6:

[0906] Based on the information collected by the server, a generative AI is used to predict the necessary actions. At this time, the user's emotional state is also taken into consideration, and actions including detailed explanations are generated for users who are feeling particularly anxious. Specifically, in addition to actions such as "check the power cable connection," "remove and reinsert the battery," and "reset the BIOS," the system also includes considerations such as "please remain calm while working."

[0907] Step 7:

[0908] The device receives the inferred action from the server. The device processes this in a format that is easy for the user to see and presents it to the respondent. The UI displays it in an intuitive and easy-to-understand format. For example, instructions such as "1. Make sure the power cable is properly connected. 2. Remove the battery, wait a few seconds, and then reinsert it. 3. Please remain calm as you work." are displayed.

[0909] Step 8:

[0910] The respondent can review the suggested actions displayed on their device, make any necessary adjustments, and then send the final answer to the user, allowing them to work on solving the problem with peace of mind.

[0911] Step 9:

[0912] The user takes specific action as instructed by the respondent. The user completes the steps and checks whether the issue is resolved. If the issue is not resolved, the user makes a new inquiry and starts again with a new step 1.

[0913] Example 2

[0914] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[0915] Currently, service desks often provide answers without considering the user's emotional state, and appropriate solutions are needed. Furthermore, because answers depend on the skill and experience of the person answering the questions, it is difficult to provide consistently high-quality service. Therefore, there is a need for a system that can quickly and accurately propose appropriate solutions while taking the user's emotional state into consideration.

[0916] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving an inquiry from a user, means for analyzing the inquiry using natural language processing technology and extracting important keywords, means for transferring the inquiry and the emotional state to the server, means for the server to search past response information, solution cases, and general technical information based on the extracted keywords and the emotional state, means for predicting and generating a required action using a generative AI model based on the searched information, means for processing the predicted and generated action and intuitively presenting it to the user, and means for the user to execute the presented action. This enables consistent, high-quality responses while taking the user's emotional state into consideration.

[0917] "User" refers to any individual or corporation that makes an inquiry using this system.

[0918] An "inquiry" refers to text data such as a question, a trouble report, or an information request that a user sends to the service desk.

[0919] "Terminal" refers to an input device such as a computer, smartphone, or tablet that a user uses to make an inquiry.

[0920] "Server" refers to a central processing unit that receives queries and emotional states sent from terminals, analyzes, searches, and generates results.

[0921] "Natural language processing technology" refers to algorithms and methods for understanding, analyzing, and processing language data.

[0922] "Keywords" refer to words or phrases extracted from inquiries that express the essence of the problem.

[0923] "Emotional state" refers to the psychological state that a user exhibits at the time of inquiry, and includes emotions such as anxiety, anger, and joy.

[0924] "Transfer means" refers to the protocol or method by which the query and emotional state are communicated from the terminal to the server.

[0925] "Knowledge base" refers to a database that stores past response information, solution cases, and general technical information.

[0926] A "generative AI model" refers to a system that uses machine learning and artificial intelligence techniques to generate new information and procedures.

[0927] An "action" refers to a solution or approach to a specific problem.

[0928] "Intuitive presentation means" refers to interfaces and methods for visually displaying information in a way that is easy for users to understand and act on.

[0929] The "means of execution" refers to the procedure or method by which the user actually executes the countermeasure in accordance with the presented action.

[0930] The present invention relates to a system for quickly and accurately responding to user inquiries. Specifically, the system combines a means for analyzing received inquiries and predicting appropriate actions with an emotion engine that recognizes the user's emotions.

[0931] When a user makes an inquiry to the service desk, the inquiry is received by the terminal as string data. The terminal analyzes the content of the inquiry using natural language processing technology (e.g., Natural Language Toolkit (NLTK) or SpaCy) and extracts important keywords. At the same time, it uses an emotion engine (e.g., a general emotion analysis API) to analyze the user's emotional state and identify emotions such as "anxiety," "anger," and "joy."

[0932] The extracted keywords and emotional state are transferred from the device to the server. Based on the received data, the server uses a knowledge base search engine (e.g., ElasticSearch or Microsoft SQL Server) to search past correspondence information, solution cases, and general technical information to collect related information. At this time, a search query is generated based on the extracted keywords and emotional state.

[0933] Furthermore, the server uses a generative AI model (e.g., GPT-4) to infer the necessary actions from the collected information and generate specific instructions for the user in natural language. For example, it generates specific actions such as "check the power cable connection," "remove and reinsert the battery," and "reset the BIOS." It also adds explanations that take the user's emotions into consideration. For example, it includes advice such as "stay calm and follow the steps step by step."

[0934] The generated results are then processed by the device into an intuitive and easy-to-understand UI and presented to the user, allowing the user to easily understand and carry out the presented procedures.

[0935] As a concrete example, consider a case where a user makes an inquiry saying, "My PC suddenly won't start up. Even when I turn it on, the screen is black." and the emotion engine recognizes that the user is anxious. The server extracts keywords such as "PC won't start up" and "black screen" and identifies the emotional state as "anxiety." Based on this, it searches for related information from the knowledge base and uses a generative AI model to generate the following specific actions:

[0936] Step 1: Check if the power cable is properly connected.

[0937] Step 2: Remove the battery, wait a few seconds, then reinsert it.

[0938] Step 3: Try resetting your BIOS.

[0939] Emotionally sensitive side note: Stay calm and check things out step by step.

[0940] Example prompt sentence:

[0941] "The user says, 'My PC suddenly stopped starting up. Even when I turn it on, the screen is black.' The emotion engine recognizes that the user is anxious. In response to this inquiry, please generate an action that the user should take and an explanation that takes into account their emotions."

[0942] This system improves the efficiency of service desk operations and the user experience by quickly providing appropriate actions while taking into consideration the user's feelings.

[0943] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0944] Step 1:

[0945] A user makes an inquiry to the service desk. The input is text data that the user enters into a browser or a dedicated inquiry form. For example, the user might enter, "My PC suddenly stopped starting up. Even when I turn it on, the screen remains black." This string of data is received by the terminal. Specifically, the user types the inquiry content into the text box and clicks the "Send" button.

[0946] Step 2:

[0947] The terminal analyzes the received inquiry using natural language processing technology. The input here is the text data received in step 1. The terminal uses libraries such as the Natural Language Toolkit (NLTK) and SpaCy to extract important keywords. For example, keywords such as "PC won't start" and "black screen" are extracted. The output is a list of extracted keywords. Specifically, the terminal runs a Python script to perform text analysis.

[0948] Step 3:

[0949] The device uses an emotion engine to analyze the user's emotional state. The input is the text data received in step 1. The emotion engine (a general emotion analysis API) is used to identify the user's emotion from the text. For example, emotion recognition may result in "the user is anxious." The output is the analyzed emotional state. Specifically, the device calls the emotion analysis API and receives the results.

[0950] Step 4:

[0951] The device transfers the extracted keywords and emotional information to the server. The input is the keyword list extracted in step 2 and the emotional state analyzed in step 3. These data are sent to the server. The output is the data transferred to the server. Specifically, the device sends the data to the server using an HTTP POST request.

[0952] Step 5:

[0953] The server searches the knowledge base based on the received data. The input here is the keyword list and emotional state transferred in step 4. The server uses a search engine such as ElasticSearch or Microsoft SQL Server to search past correspondence information, solution cases, and general technical information. The output is a list of related information. Specifically, the server generates a search query and executes the search against the database.

[0954] Step 6:

[0955] The server uses a generative AI model to infer and generate the required actions. The input is the list of relevant information obtained in step 5. A generative AI model (e.g., GPT-4) is used to generate specific steps based on the relevant information. For example, actions such as "check the power cable connection," "remove and reinsert the battery," and "reset the BIOS" are generated. An emotion-sensitive explanation is also generated. The output is the generated steps and the emotion-sensitive explanation. Specifically, the server sends the prompt text to the generative AI model and receives the generated text.

[0956] Step 7:

[0957] The server sends the generated results to the terminal. The input is the procedure and explanation generated in step 6. It sends these data to the terminal. The output is the data sent to the terminal. In concrete terms, the server sends the data to the terminal using an HTTP POST request.

[0958] Step 8:

[0959] The terminal displays the received instructions and explanations in an intuitive and easy-to-understand UI. The input is the data received in step 7. The terminal generates a user interface (UI) and presents the instructions to the user in an easy-to-understand format. The output is the instructions and explanations displayed to the user. Specifically, the terminal uses HTML and JavaScript to update the user interface.

[0960] In this way, through each processing step, it is possible to respond quickly and accurately to user inquiries.

[0961] (Application example 2)

[0962] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[0963] Conventional service desk systems have difficulty in providing prompt and appropriate responses to user inquiries, and have had problems with poor user experience because they do not take the user's emotions into consideration. Therefore, there is a demand for a service desk system that can recognize the user's emotions and provide prompt and appropriate responses.

[0964] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0965] In this invention, the server includes means for receiving an inquiry from a user, means for analyzing the inquiry using natural language processing technology and extracting important keywords, means for searching past response information, solved cases, and general technical information based on the extracted keywords and user emotion information, means for predicting a required action using a generative AI from the searched information and user emotion information, means for processing the predicted action and presenting it to a respondent, means for the respondent to send a response to the user using the presented action, and emotion analysis means for recognizing emotions from a user image. This makes it possible to quickly provide an appropriate response based on the content of the inquiry and the user emotion.

[0966] The "means for receiving inquiries from users" refers to a device or program for receiving information such as questions or requests made by users to the system.

[0967] "Natural language processing technology" is a technology that enables computers to understand, analyze, and generate human language.

[0968] The "means for extracting important keywords" is a device or program for identifying and extracting key words and phrases from the content of a user's inquiry.

[0969] "User emotion information" is data indicating the emotional state of the user when making an inquiry.

[0970] "Past response information" is a record of responses and responses to previous inquiries.

[0971] "Solution cases" are specific examples of problems that have occurred in the past and how they have been solved.

[0972] "General technical information" is information about widely used technologies and methods.

[0973] "Generative AI" is a technology that uses artificial intelligence to generate new information and answers.

[0974] The "means for predicting the necessary action" is a device or program for identifying the appropriate response procedure or process based on the extracted keywords and the user's emotions.

[0975] The "means for presenting to the respondent" is a device or program for visually or audibly presenting the inferred action.

[0976] The "means for transmitting a response to the user" is a device or program for transmitting the generated response procedure or information to the user.

[0977] "Emotion analysis means" refers to a device or program for analyzing and recognizing emotions from a user's facial expressions, voice, text, etc.

[0978] This invention is a service desk system that provides fast and accurate responses to user inquiries. This system combines natural language processing and sentiment analysis technology and uses generative AI models to improve the user experience and streamline problem resolution.

[0979] The system first receives a query from the user. The query is in text format and is sent to the system via a device such as a smartphone or PC. At this stage, image data from the user can also be received at the same time, and this image data will be used for sentiment analysis.

[0980] The received inquiry is analyzed by a natural language processing engine (for example, Google's Dialogflow or IBM's Watson). This analysis extracts important keywords and phrases from the inquiry. For example, from an inquiry such as "The product I ordered hasn't arrived," keywords such as "order," "product," and "didn't arrive" are extracted.

[0981] At the same time, the image data is subjected to emotion analysis using Microsoft Azure's Emotion API. This analysis identifies whether the user is in an emotional state such as anxiety, anger, or joy. For example, the received image may identify the user as "anxious."

[0982] The analyzed keywords and sentiment information are then transferred to a cloud server, which uses this information to search a knowledge base and retrieve past support information, case studies, and general technical information, using cloud services such as Amazon's AWS.

[0983] Next, generative AI technology (e.g., OpenAI's GPT-4) uses this information to infer the necessary action. Combining keywords and sentiment information, the most appropriate response steps are generated for the user. For example, the following prompt sentence can be used to generate the necessary response steps:

[0984] Inquiry: I haven't received my order, what should I do?

[0985] User Emotion: Anxiety

[0986] Generate appropriate response procedures.

[0987] The generated response procedures are presented to the user in an easy-to-understand manner. This is done through the device's UI. For example, the system may say, "Please check the delivery status. Also, please enter your order number and contact us again. Please remain calm." Additional explanations that take emotions into consideration are also included.

[0988] Ultimately, this system allows users to receive prompt and accurate responses, improving service quality. This system, which uses various hardware and software such as servers, terminals, natural language processing engines, sentiment analysis engines, and generative AI models, will greatly contribute to improving the user experience.

[0989] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0990] Step 1:

[0991] A user enters an inquiry using a device such as a smartphone or PC, and if there is image data, attaches it and sends it to the system. The input at this time is the inquiry content in text format and image data. The output is the raw inquiry data sent to the server.

[0992] Step 2:

[0993] The server passes the received inquiry to a natural language processing engine for analysis. This process extracts important keywords and phrases from the input text. For example, keywords such as "order," "product," and "not delivered" are extracted from the text. The output is a list of keywords as the analysis results.

[0994] Step 3:

[0995] The server passes the received image data to an emotion analysis engine to recognize the user's emotional state. The emotion analysis engine (for example, Microsoft Azure's Emotion API) analyzes the image data and identifies emotions such as "anxiety," "anger," and "joy." This emotional data is returned to the server as output.

[0996] Step 4:

[0997] The server searches a cloud knowledge base based on the extracted keywords and analyzed sentiment data. The knowledge base (e.g., Amazon AWS) stores past response information, solution cases, and general technical information. Highly relevant information is output as the search results.

[0998] Step 5:

[0999] The server combines the knowledge base search results with the user's emotional data and generates a response procedure using a generative AI model (e.g., OpenAI's GPT-4). Specifically, the server inputs the following prompt sentence into the generative AI:

[1000] Inquiry: I haven't received my order, what should I do?

[1001] User Emotion: Anxiety

[1002] Generate appropriate response procedures.

[1003] The generated response procedure is obtained as output.

[1004] Step 6:

[1005] The server transfers the generated response procedures to the terminal, which displays them in an easy-to-understand format for the user. For example, specific instructions such as "Please check the delivery status. Also, please enter your order number and contact us again. Please stay calm." are displayed. The output is the response procedures presented to the user.

[1006] Step 7:

[1007] The user follows the procedure displayed on the terminal and takes action to resolve the problem. The output at this stage is the execution of the action by the user.

[1008] 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 control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1009] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1010] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1011] The emotion identification model 59 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 an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1012] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1013] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1014] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1015] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1016] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs 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 a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1017] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1018] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1019] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1020] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1021] 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.

[1022] It is not necessary to store all 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 all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1023] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1024] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1025] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1026] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1027] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1028] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1029] The following is further disclosed regarding the above embodiment.

[1030] (Claim 1)

[1031] means for receiving a query from a user;

[1032] means for analyzing the query using natural language processing technology and extracting important keywords;

[1033] A means for searching past correspondence information, solution cases, and general technical information based on the extracted keywords;

[1034] A means for predicting a required action from the retrieved information using a generation AI;

[1035] a means for processing the inferred action and presenting it to the respondent;

[1036] A means for the respondent to send an answer to a user using the proposed action;

[1037] A system including:

[1038] (Claim 2)

[1039] 10. The system of claim 1, wherein the means for analyzing the query includes morphological analysis and dependency structure analysis.

[1040] (Claim 3)

[1041] The system of claim 1 , wherein the search means utilizes a knowledge-based search engine.

[1042] "Example 1"

[1043] (Claim 1)

[1044] means for receiving a query from a user;

[1045] means for analyzing the query using natural language processing technology and extracting important keywords;

[1046] A means for searching past correspondence information, solution cases, and general technical information based on the extracted keywords;

[1047] means for inferring a required action from the retrieved information using a generative AI model;

[1048] means for processing the inferred action and displaying it on a terminal;

[1049] means for the terminal to present an answer to the user using the displayed action;

[1050] A system including:

[1051] (Claim 2)

[1052] 10. The system of claim 1, wherein the means for analyzing the query includes morphological analysis and dependency structure analysis.

[1053] (Claim 3)

[1054] The system of claim 1 , wherein the search means utilizes a knowledge-based search engine.

[1055] "Application Example 1"

[1056] (Claim 1)

[1057] means for receiving a query from a user;

[1058] means for analyzing the query using natural language processing technology and extracting important keywords;

[1059] A means for searching past correspondence information, solution cases, and general technical information based on the extracted keywords;

[1060] A means for predicting a required action from the retrieved information using a generation AI;

[1061] a means for processing the inferred action and presenting it to the respondent in an intuitive user interface;

[1062] A means for the respondent to send an answer to a user using the proposed action;

[1063] A system including:

[1064] (Claim 2)

[1065] 10. The system of claim 1, wherein the query analyzing means includes morphological analysis and dependency structure analysis, and the searching means uses a weighted search algorithm.

[1066] (Claim 3)

[1067] The system of claim 1 , wherein the query is input as string data and the inferred action is displayed on a smartphone display.

[1068] "Example 2: Combining Emotion Engines"

[1069] (Claim 1)

[1070] means for receiving a query from a user;

[1071] means for analyzing the query using natural language processing technology and extracting important keywords;

[1072] means for transmitting said query and emotional state to a server;

[1073] A means for the server to search for past correspondence information, solution cases, and general technical information based on the extracted keywords and emotional state;

[1074] means for predicting and generating a required action using a generative AI model based on the retrieved information;

[1075] a means for processing the inferences and generated actions and presenting them to a user in an intuitive manner;

[1076] means for a user to perform the proposed action;

[1077] A system including:

[1078] (Claim 2)

[1079] 10. The system of claim 1, wherein the means for analyzing the query includes morphological analysis and dependency structure analysis.

[1080] (Claim 3)

[1081] The system of claim 1 , wherein the search means utilizes a knowledge-based search engine.

[1082] "Application example 2 when combining emotion engines"

[1083] (Claim 1)

[1084] means for receiving a query from a user;

[1085] means for analyzing the query using natural language processing technology and extracting important keywords;

[1086] A means for searching past response information, solution cases, and general technical information based on the extracted keywords and user emotion information;

[1087] A means for predicting a necessary action using a generation AI based on the retrieved information and the user's emotional information;

[1088] a means for processing the inferred action and presenting it to the respondent;

[1089] A means for the respondent to send an answer to a user using the proposed action;

[1090] emotion analysis means for recognizing emotions from a user image;

[1091] A system including:

[1092] (Claim 2)

[1093] 10. The system of claim 1, wherein the means for analyzing the query includes morphological analysis and dependency structure analysis.

[1094] (Claim 3)

[1095] The system of claim 1 , wherein the search means utilizes a knowledge-based search engine. [Explanation of symbols]

[1096] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means for receiving a query from a user; means for analyzing the query using natural language processing technology and extracting important keywords; A means for searching past correspondence information, solution cases, and general technical information based on the extracted keywords; A means for predicting a required action from the retrieved information using a generation AI; a means for processing the inferred action and presenting it to the respondent; A means for the respondent to send an answer to a user using the proposed action; A system including:

2. The system of claim 1 , wherein the means for analyzing the query includes morphological analysis and dependency structure analysis.

3. The system of claim 1 , wherein the search means utilizes a knowledge-based search engine.

Images