System

Abstract

A system is provided.SOLUTION: A system comprising: means for receiving a query; means for formatting the received query into a unified format; means for analyzing the query to extract key words; means for searching for similar queries in the past and identifying the most similar query; means for generating an automated response message using a generative AI; and means for transmitting the generated response message to a user.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] The current system for responding to inquiries has many contact points, resulting in a problem of dispersed knowledge. Furthermore, the free-form nature of inquiries means that communication and investigation take time, reducing the efficiency of inquiry responses. This situation increases the workload of both the user making the inquiry and the person handling the inquiry, resulting in a decline in productivity. [Means for solving the problem]

[0005] In order to solve the above-mentioned problems, the present invention provides the following means. Specifically, the present invention provides a system including a means for receiving inquiry content, a means for formatting the received inquiry content into a unified format, a means for analyzing the inquiry content and extracting key keywords, a means for searching past similar inquiries and identifying the most similar inquiry, a means for generating an automatic response message using a generation AI, and a means for sending the generated response message to a user. The present invention also provides a system that further includes a means for reanalyzing the inquiry content and identifying an appropriate department, and a means for forwarding the inquiry to the identified appropriate department. Furthermore, the present invention provides a system that also includes a means for providing a user interface for accessing an inquiry portal, thereby making it possible to improve the efficiency of inquiry responses and business productivity.

[0006] The "means for receiving the inquiry content" refers to a method or device for electronically receiving the inquiry data sent by the user.

[0007] The "means for converting data into a unified format" refers to a method or device for converting received raw query data into a predetermined format that is easy to analyze and store.

[0008] "Means for extracting key keywords" refers to a method or device for identifying and extracting important words and phrases from the content of an inquiry using natural language processing techniques or the like.

[0009] The "means for searching for similar past queries" refers to a method or device for referencing existing query data in a database and finding past cases similar to the received query.

[0010] "Means for generating automatic response messages using generation AI" refers to a method or device for automatically creating response content using artificial intelligence based on extracted keywords and past similar inquiry data.

[0011] The "means for transmitting to the user" refers to a method or device for transmitting the generated automatic response message to the user's terminal and transmitting it so that the user can check it.

[0012] The "means for identifying an appropriate department" refers to a method or device for reanalyzing the content of an inquiry and determining the most appropriate department or person in charge.

[0013] "Means for forwarding to an identified appropriate department" refers to a method or device for automatically redirecting an inquiry to a determined department or person so that that department or person can respond.

[0014] The "means for providing a user interface" refers to a method or device for providing a screen or an operation method that allows a user to easily send an inquiry. [Brief explanation of the drawings]

[0015] [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

[0016] 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.

[0017] First, the terms used in the following description will be explained.

[0018] 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).

[0019] 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.

[0020] 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.

[0021] 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.

[0022] 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."

[0023] [First embodiment]

[0024] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0025] 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.

[0026] 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).

[0027] 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.

[0028] 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.

[0029] 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.

[0030] 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.

[0031] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0032] 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.

[0033] 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.

[0034] 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.

[0035] 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."

[0036] This invention is a system that automates responses to inquiries from users, aiming to improve the efficiency and productivity of inquiry responses. This system receives inquiry content, formats it into a unified format, analyzes it to extract key keywords, searches for similar past inquiries, and uses generative AI to automatically generate a response message that can be sent to the user. It can also reanalyze the inquiry content to identify the appropriate department and redirect the inquiry to the identified department.

[0037] Program processing overview

[0038] 1. Receiving an Inquiry

[0039] Users access a dedicated inquiry portal and enter their questions in an inquiry form.

[0040] When the user presses the send button, the inquiry is sent to the server.

[0041] The server stores the received inquiry in a database.

[0042] 2. Formatting into a unified format

[0043] The server formats the raw query content it receives into a predefined format, which makes subsequent processing easier.

[0044] 3. Content Analysis

[0045] The server analyzes the formatted query using natural language processing technology and extracts key keywords.

[0046] 4. Search for similar queries from the past

[0047] The server searches a database for similar past queries and identifies the most similar queries.

[0048] The server scores the search results and selects those with the highest similarity.

[0049] 5. Generate Auto-Response Messages

[0050] The server uses a generation AI to generate an automatic response message based on the extracted keywords and selected past inquiries.

[0051] 6. Transmission to the User

[0052] The server sends the generated automatic response message to the user's terminal.

[0053] The user checks the sent response message on the terminal.

[0054] 7. Redirect to the appropriate department

[0055] The server re-analyzes the query content and identifies an appropriate department if one has not been set.

[0056] The server forwards the query to the specified department if redirection is necessary.

[0057] A terminal (in the appropriate department) receives the inquiry and provides the information the user requests.

[0058] Specific examples

[0059] Example 1: Inquiry about network trouble

[0060] A user enters a query into the portal, such as "What should I do if my network is unstable?" and presses the send button. The server receives this query and stores it in a database. The server then formats the query based on keywords such as "network," "unstable," and "how to deal with it," and analyzes it using natural language processing technology. It then searches the database for similar past queries and identifies the most similar case. Generative AI is used to generate an automated response message (e.g., "If your network is unstable, please follow the steps below...") and sends it to the user. When the user receives the response message, it becomes easier for them to solve the problem themselves.

[0061] Example 2: Redirection due to misidentification of inquiry content

[0062] If a user sends a query saying "My PC won't start," but the server initially analyzes it and tries to treat it as a network problem, the server will reanalyze it and determine that the query is actually a hardware problem. Therefore, the query will be automatically redirected to the IT support department. The terminal (the IT support department's terminal) will receive this query and provide an appropriate solution.

[0063] In this way, the present invention can dramatically improve the efficiency of responding to inquiries and reduce the man-hours of both users and responders.

[0064] The processing flow will be explained below.

[0065] Step 1:

[0066] The user accesses a dedicated inquiry portal and enters the question into the inquiry form. The user also fills in the necessary information in the input fields and presses the send button to send the inquiry.

[0067] Step 2:

[0068] The server receives the inquiry sent by the user, and the received data is stored in a database for record keeping.

[0069] Step 3:

[0070] The server formats the received queries into a unified format, converting raw data into a defined format for easier subsequent processing.

[0071] Step 4:

[0072] The server analyzes the query using natural language processing technology, specifically extracting key keywords and phrases and obtaining key information needed to understand the query.

[0073] Step 5:

[0074] The server searches the database for similar past queries, identifies the most similar past queries based on the extracted keywords, and calculates a relevance score.

[0075] Step 6:

[0076] The server selects the query with the highest similarity from the search results and prepares to generate a response message based on the selected past query data.

[0077] Step 7:

[0078] The server uses a generation AI to generate an automated response message, creating a specific and appropriate response message based on selected past inquiry data and extracted keywords.

[0079] Step 8:

[0080] The server then sends the generated automated response message to the user's terminal, allowing the user to receive an immediate response.

[0081] Step 9:

[0082] The user checks the automatic response message on the terminal, which allows the user to quickly find a way to resolve the problem.

[0083] Step 10:

[0084] The server re-parses the query and identifies the appropriate department if one is not already set, so that the query can be re-evaluated and handled by the correct department.

[0085] Step 11:

[0086] The server then forwards the query to the appropriate department, and the query is automatically redirected to the correct department.

[0087] Step 12:

[0088] The terminal (the terminal in the appropriate department) receives the inquiry and provides the information the user is looking for. The department staff responds, allowing the user to receive professional support.

[0089] Through these steps, inquiries from users are handled efficiently and quickly. By automating the entire inquiry response process, the burden on staff is reduced and users can have their problems resolved quickly.

[0090] Example 1

[0091] 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."

[0092] Conventional inquiry response systems analyzed the content of inquiries individually and responded manually, resulting in issues with efficiency and productivity. Furthermore, they did not redirect inquiries to the appropriate department or generate automated response messages, increasing the likelihood of incorrect responses. Furthermore, responding to inquiries required time and effort, which led to lower user satisfaction. This resulted in increased response workloads across the company and a decline in overall performance.

[0093] 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.

[0094] In this invention, the server includes: means for receiving inquiry content; means for formatting the received inquiry content into a unified format; means for analyzing the inquiry content and extracting key keywords; means for searching for similar past inquiries and identifying the most similar inquiry; means for generating an automatic response message using a generative AI model; means for sending the generated response message to the user; means for providing a redirect function to an appropriate department; means for reanalyzing the inquiry content and identifying an appropriate department; means for forwarding the inquiry to the identified appropriate department; means for providing a user interface for accessing the inquiry portal; means for scoring answer candidates and selecting the most appropriate answer; means for analyzing the inquiry content using natural language processing technology; and means for generating a prompt sentence based on the analysis result and inputting it into the generative AI model. This automates the inquiry content analysis and response process, enabling prompt and accurate responses. Furthermore, adding a redirect function to the appropriate department improves user satisfaction and reduces response time.

[0095] "Inquiry content" is text data of questions, requests, and inquiries sent by users through the system.

[0096] The "receiving means" refers to the technical means for inputting the contents of the inquiry sent by the user into the server.

[0097] The "means for formatting into a unified format" refers to a technical means for converting the received query content into a format that is easy to process for subsequent analysis.

[0098] The "analysis means" is a natural language processing technique for extracting key keywords from the content of the inquiry.

[0099] A "keyword" is a word or phrase that contains important information in the query.

[0100] The "means for searching for similar past queries" refers to a technical means for finding similar past queries from a database.

[0101] A "generative AI model" is a system that uses artificial intelligence technology to automatically generate new response messages.

[0102] A "means for generating an auto-response message" is a technical means for automatically creating a response to a user's inquiry using a generative AI model.

[0103] "Means for sending" refers to the technical means for delivering the generated auto-reply message to the user.

[0104] The "redirect function to the appropriate department" is a function that transfers an inquiry to the appropriate department based on the content of the inquiry.

[0105] "Reanalysis measures" are technical measures that perform additional analysis when the results of the initial analysis are unclear.

[0106] "Means of identification" refers to technical means for determining the appropriate department to respond based on reanalysis.

[0107] "Forwarding means" refers to the technical means by which the inquiry is redirected to a specified department.

[0108] "User Interface" means the means by which a user accesses the inquiry portal.

[0109] "Scoring means" refers to a technical means for evaluating detected answer candidates and selecting the most appropriate answer.

[0110] "Natural language processing technology" is an artificial intelligence technology for understanding, analyzing, and processing human language.

[0111] A "prompt sentence" is an instruction sentence that is input into a generative AI model to generate a response.

[0112] The present invention is a system that automates responses to user inquiries, thereby improving the efficiency and productivity of inquiry responses. This system functions through communication between a server, terminals, and users.

[0113] Hardware and software used

[0114] The server is a high-performance computer system, and uses a database management system (DBMS) such as MySQL (registered trademark) or PostgreSQL. For server-side natural language processing, a natural language processing engine such as Tensorflow (registered trademark) or Watson (registered trademark) is used. Furthermore, OpenAI's (registered trademark) GPT-3 (registered trademark) is used as the generative AI model.

[0115] A terminal is a device where a user inputs a query and receives a response message. This can be a PC, tablet, smartphone, or other device. The user interface is implemented as a web application or a mobile application.

[0116] Processing Overview

[0117] 1. Receiving an Inquiry

[0118] Users access a dedicated inquiry portal and enter their questions in the inquiry form. For example, they might enter, "Please tell me what to do if the network is unstable."

[0119] When the user presses the send button, the query content is sent to the server as an HTTP request.

[0120] The server stores the received inquiry in a database.

[0121] 2. Formatting into a unified format

[0122] The server formats the received query content into a specific format (e.g., JSON format), which makes subsequent processing easier.

[0123] 3. Content Analysis

[0124] The server analyzes the formatted query using a natural language processing engine (e.g., TensorFlow) and extracts key keywords, such as "network," "unstable," and "how to respond."

[0125] 4. Search for similar queries from the past

[0126] The server searches the database to identify similar past queries, and the search results are scored to select the most similar queries.

[0127] 5. Generate Auto-Response Messages

[0128] The server uses a generative AI model (e.g., GPT-3) to generate an automated response message based on the extracted keywords and selected past inquiries. The prompt text is "Please generate an automated response message in response to the user's inquiry, 'Please tell me what to do if the network is unstable.'"

[0129] 6. Transmission to the User

[0130] The server then sends the generated auto-reply message to the user's device via email or a notification API (e.g., Firebase Cloud Messaging).

[0131] The user checks the response message sent and uses it to solve the problem.

[0132] 7. Redirect to the appropriate department

[0133] The server re-analyzes the query and identifies the appropriate department.

[0134] The server redirects the inquiry to the specified department, for example, contacting the IT support department via email or an internal notification system.

[0135] The terminal (the terminal of the appropriate department) receives this inquiry and initiates specific actions.

[0136] Specific examples

[0137] Example 1: Inquiry about network trouble

[0138] When a user enters "What to do if the network is unstable" into the inquiry portal and submits it, the server receives the inquiry and stores it in a database. Next, the inquiry is formatted based on keywords such as "network," "unstable," and "how to deal with it," and analyzed using natural language processing technology. After that, similar past inquiries are searched for in the database to identify the most similar case. Using generative AI, an automatic response message is generated that reads, "If the network is unstable, please follow the steps below..." and sent to the user. When the user receives the response message, it becomes easier for them to solve the problem themselves.

[0139] Example 2: Redirection due to misidentification of inquiry content

[0140] If a user sends a query saying "My PC won't start," but the server initially analyzes it and treats it as a network problem, the server reanalyzes it and determines that the query is actually a hardware problem. Therefore, the query is automatically redirected to the IT support department. The terminal (the IT support department's terminal) receives this query and can provide an appropriate solution.

[0141] In this way, the present invention can dramatically improve the efficiency of responding to inquiries and reduce the man-hours of both users and responders.

[0142] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0143] Step 1:

[0144] The user accesses the inquiry portal, enters the question in the inquiry form, and presses the send button.

[0145] Input: The inquiry entered by the user (e.g., "Please tell me what to do if the network is unstable").

[0146] Output: The query stored in the database.

[0147] Specific operation: The server stores the query content received as an HTTP request in a database. In this case, a database management system such as MySQL or PostgreSQL is used.

[0148] Step 2:

[0149] The server formats the received query content into a specified format (e.g., JSON format).

[0150] Input: The raw query stored in the database.

[0151] Output: Pretty-printed JSON representation of the query.

[0152] What happens: The server converts the query to JSON format, removing line breaks and special characters. This formatting makes it easier to parse later.

[0153] Step 3:

[0154] The server analyzes the formatted query using a natural language processing (NLP) engine and extracts key keywords.

[0155] Input: Query content in JSON format.

[0156] Output: Extracted key keywords (e.g., "network," "unstable," "how to respond").

[0157] Specific operation: The server uses a natural language processing engine such as TensorFlow or Watson to analyze the query content and identify important keywords.

[0158] Step 4:

[0159] The server searches a database to identify similar past queries.

[0160] Input: Extracted main keywords.

[0161] Output: The most similar previous queries.

[0162] Specific operation: The server generates an SQL query based on the extracted keywords and searches the database. The search results are scored using a search engine such as ElasticSearch (registered trademark), and the results with the highest similarity are selected.

[0163] Step 5:

[0164] The server generates the automated response message using the generative AI model.

[0165] Input: The most similar previous query, extracted keywords, and prompt (e.g., "Please generate an auto-response message in response to the user's query 'What should I do if my network is unstable?'").

[0166] Output: Auto-reply message (e.g. "If your network is unstable, please try the following steps...").

[0167] Specific operation: The server inputs a prompt sentence into a generative AI model such as OpenAI's GPT-3 to generate an automated response message.

[0168] Step 6:

[0169] The server sends the generated automatic response message to the user's terminal.

[0170] Input: Auto-reply message.

[0171] Output: The response message that is displayed on the user's terminal.

[0172] Specific operation: The server sends the generated message to the user using a notification API such as email or Firebase Cloud Messaging. The user then checks the message on their device.

[0173] Step 7:

[0174] The server re-analyzes the query and identifies the appropriate department.

[0175] Input: Query content, reanalysis instructions.

[0176] Output: The appropriate department identified.

[0177] Specific operation: The server uses the natural language processing engine again to re-analyze the inquiry content and identify the appropriate department.

[0178] Step 8:

[0179] The server redirects the query to the specified department.

[0180] Input: Identified appropriate department, inquiry details.

[0181] Output: Query redirected to the appropriate department's terminal.

[0182] Specific operation: The server forwards the inquiry to the specified department using an internal notification system or email. The appropriate department's terminal receives the inquiry and begins to respond.

[0183] (Application example 1)

[0184] 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."

[0185] Current inquiry response systems have difficulty responding quickly and accurately to a wide variety of inquiries from users. They also face the problem of increasing the workload of responders, reducing the efficiency of customer service. Furthermore, it is difficult to provide consistent responses to similar inquiries, making it difficult to improve customer satisfaction.

[0186] 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.

[0187] In this invention, the server includes means for receiving the inquiry content, means for formatting the received inquiry content into a unified format, means for analyzing the inquiry content and extracting key keywords, means for searching past similar inquiries and identifying the most similar inquiry, means for generating a prompt sentence using a generative AI model and generating an automatic response message, means for transmitting the generated response message to an information processing device, and means for reanalyzing the inquiry content, identifying an appropriate department if an appropriate department has not been set, and transferring the inquiry to the appropriate department. This dramatically improves the efficiency of inquiry response and reduces the man-hours of both users and responders.

[0188] The "means for receiving the inquiry content" is a function for transmitting the inquiry content submitted by the user to the server via the network, and for the server to receive the content.

[0189] The "means for converting received inquiry content into a unified format" is a function for converting inquiry content sent in various formats into a consistent unified format.

[0190] "Means for analyzing the content of inquiries and extracting key keywords" refers to a function that utilizes natural language processing technology to identify and extract key keywords from the content of inquiries.

[0191] The "means for searching past similar queries and identifying the most similar query" is a function for comparing a query received from a database in the past with the most similar query and identifying the most similar query.

[0192] "Means for generating prompt sentences using a generative AI model and generating automatic response messages" refers to a function that uses a pre-trained generative AI model to generate prompt sentences required for a response and creates automatic response messages based on those sentences.

[0193] The "means for transmitting the generated response message to the information processing device" is a function for transferring the generated response message to the user's information processing device (for example, a smartphone or a computer) and displaying it.

[0194] "Means for reanalyzing the content of the inquiry, identifying an appropriate department if an appropriate department has not been set, and transferring the inquiry to the relevant department" is a function that reanalyzes the content of the inquiry, and if an appropriate department has not been set, identifies the most appropriate department and transfers the inquiry to that department.

[0195] MODE FOR CARRYING OUT THE INVENTION

[0196] System Configuration

[0197] The system implemented in this invention is broadly composed of the following main means.

[0198] 1. Means of receiving inquiries

[0199] 2. A means of formatting received inquiries into a unified format

[0200] 3. A method for analyzing inquiries and extracting key keywords

[0201] 4. A means to search for similar queries from the past and identify the most similar queries

[0202] 5. A means of generating prompts and auto-response messages using generative AI models

[0203] 6. Means for transmitting the generated response message to the information processing device

[0204] 7. A means to re-analyze the content of the inquiry, identify the appropriate department if one is not set, and transfer the inquiry to the appropriate department.

[0205] Program processing

[0206] Hardware and software usage

[0207] server:

[0208] Hardware: High-performance server

[0209] Software: Database management systems (e.g., MySQL), search engines (e.g., Elasticsearch), natural language processing libraries (e.g., Spacy, BERT), generative AI models (e.g., ChatGPT®, GPT-3)

[0210] User device:

[0211] Hardware: Smartphones, tablets, computers

[0212] Software: Mobile application, browser-based interface

[0213] Data processing and calculation flow

[0214] 1. Receiving an Inquiry

[0215] Users access the inquiry form using a smartphone or PC, enter their question, and submit it. The entered inquiry is sent to a server via a network and stored in a database.

[0216] 2. Formatting into a unified format

[0217] The server converts the received query content into a unified format, using a text formatting library (e.g., Python's NLTK) for this conversion process.

[0218] 3. Content Analysis

[0219] The server analyzes the query using natural language processing techniques (e.g., Spacy, BERT) to extract key keywords. This analysis identifies the core of the query.

[0220] 4. Search for similar queries from the past

[0221] The server compares the queries against previous queries in a database to find the most similar queries, using a search engine such as Elasticsearch to score the similarity and identify highly similar queries.

[0222] 5. Generate Auto-Response Messages

[0223] The server uses a generative AI model (e.g., ChatGPT, GPT-3) to generate an automated response message based on the prompt, which includes extracted keywords and information about similar past inquiries.

[0224] 6. Transmission to the User

[0225] The server generates a response message and sends it to the user's device. A real-time notification service (e.g., Firebase) is used to deliver the message to the user quickly and reliably.

[0226] 7. Redirect to the appropriate department

[0227] The server re-analyzes the inquiry and, if an appropriate department is not set, identifies the appropriate department and transfers the inquiry. This allows inquiries that require specialized assistance to be handled appropriately.

[0228] Specific examples

[0229] Example 1: Delivery enquiry

[0230] A user submits an inquiry saying, "The item I ordered hasn't arrived." This inquiry is received by the server, and key keywords such as "order," "item," and "not arrived" are extracted. Similar past inquiries are searched for, and the generative AI model generates a response message saying, "We will confirm your order number and let you know the delivery status. Delivery usually takes 5-7 days." This message is then promptly sent to the user's device.

[0231] Prompt Sentence Examples

[0232] "Please generate an auto-reply message for the following inquiry: 'My order hasn't arrived'"

[0233] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0234] Step 1:

[0235] The user accesses the smart mail-order support application, enters their question into the inquiry form, and presses the send button. At this time, the user may enter something like, "The item I ordered hasn't arrived." The entered data is saved in JSON format within the application and sent to the server via the network. The server receives the inquiry and simultaneously saves the information in a database. The input is the user's question, and the output is saved in the database.

[0236] Step 2:

[0237] The server formats the received raw query content into a specified unified format. This process uses a text formatting library (e.g., Python's NLTK). Specifically, it parses JSON-formatted data, removes unnecessary whitespace and special characters, and formats the query content. The input is the received raw query content, and the output is the data formatted in a unified format.

[0238] Step 3:

[0239] The server analyzes the formatted query using natural language processing technology (e.g., Spacy, BERT) and extracts key keywords. This results in keywords such as "order," "product," and "not delivered." The input is data formatted in a unified format, and the output is the key keywords.

[0240] Step 4:

[0241] The server uses a search engine (e.g., Elasticsearch) to search a database for similar past queries and identify the most similar queries. It scores the search results and selects queries with high similarity. The input is the main keywords, and the output is the past query data with high similarity.

[0242] Step 5:

[0243] The server uses a generative AI model (e.g., ChatGPT, GPT-3) to generate a prompt and generate an auto-response message. This prompt includes extracted keywords and information on the most similar past inquiries. For example, it could be in the format "Please generate an auto-response message in response to the following inquiry: 'The item I ordered has not arrived.'" The input is similar inquiry data, and the output is the generated response message.

[0244] Step 6:

[0245] The server sends the generated automatic response message to the user's device using a real-time notification service (e.g., Firebase). The user can check this response message within the application. The input is the generated response message, and the output is the notification and display on the user's device.

[0246] Step 7:

[0247] The server re-analyzes the query and identifies the appropriate department if one is not already set. If redirection is required, the server forwards the query to the appropriate department. The input is the user's query, and the output is a redirect to the identified appropriate department.

[0248] 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.

[0249] This invention is a system that automates responses to inquiries from users, aiming to improve the efficiency and productivity of inquiry responses. This system receives inquiry content, formats it into a unified format, analyzes it to extract key keywords, searches for similar past inquiries, and uses generative AI to generate an automatic response message that is sent to the user. It can also reanalyze the inquiry content to identify the appropriate department and redirect the inquiry to the identified department. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it also has the function of improving the quality of inquiry responses.

[0250] Program processing overview

[0251] 1. Receiving an Inquiry

[0252] Users access a dedicated inquiry portal, enter their questions into the inquiry form, and then press the send button to send the inquiry to the server.

[0253] The server checks the received inquiry and stores it in a database.

[0254] 2. Formatting into a unified format

[0255] The server converts the received raw query content into a unified format that can be easily parsed and processed.

[0256] 3. Content Analysis

[0257] The server uses NLP technology to analyze the query content, which has been formatted into a unified format, and extracts key keywords and phrases.

[0258] 4. Search for similar queries from the past

[0259] The server searches a database for similar queries from the past to identify the most similar queries, and the search results are evaluated based on a relevance score.

[0260] 5. Generate Auto-Response Messages

[0261] The server uses a generation AI to create an automated response message based on the extracted keywords and identified similar queries.

[0262] 6. Emotional Recognition and Regulation

[0263] The server uses an emotion engine to recognize emotions from the user's query content and expressions, and adjusts the tone and content of the generated response message according to the user's emotions.

[0264] For example, if the user is expressing anger or frustration, the response message will have a more polite and empathetic tone.

[0265] 7. Transmission to the User

[0266] The server sends the generated automatic response message to the user's terminal, so that the user can receive the response quickly.

[0267] 8. Redirect to the appropriate department

[0268] The server re-analyzes the query and, if the appropriate department is not set, forwards the query to the correct department.

[0269] The terminal (the terminal of the appropriate department) responds to the received inquiry and provides the user with an appropriate answer.

[0270] Specific examples

[0271] Example 1: Inquiry about network trouble

[0272] The user enters "Please tell me what to do if the network is unstable" into an inquiry form and submits it. The server receives the inquiry and saves it in a database. The server then formats the inquiry into a unified format and uses NLP technology to extract keywords such as "network," "unstable," and "how to deal with it." The server then searches the database for similar past inquiries and identifies the most similar one. It then uses generative AI to generate an automated response message, which the server then sends to the user. The server then uses an emotion engine to recognize the user's emotions. For example, if the user expresses frustration, the response message is adjusted to show empathy, such as "We apologize for the inconvenience. If the network is unstable, please follow the steps below..."

[0273] Example 2: Redirection due to misidentification of inquiry content

[0274] A user enters "My PC won't start" into an inquiry form and submits it. The server receives the inquiry and saves it in a database. The server analyzes the inquiry and extracts keywords such as "PC" and "won't start." An initial analysis identifies it as a network problem, but reanalysis identifies it as a hardware problem. In this case, the server automatically redirects the inquiry to the IT support department. A terminal (a terminal in the IT support department) receives the inquiry and provides an appropriate solution.

[0275] In this way, the present invention dramatically improves the efficiency of inquiries and reduces the man-hours required by both users and responders. The introduction of an emotion engine can provide an even higher level of user experience.

[0276] The processing flow will be explained below.

[0277] Step 1:

[0278] Users access a dedicated inquiry portal, enter their questions in the inquiry form, and then press the send button to send the inquiry.

[0279] Step 2:

[0280] The server receives the inquiry sent by the user, and the received inquiry is immediately saved in the database.

[0281] Step 3:

[0282] The server formats the raw query content it receives into a uniform format, ensuring data consistency and facilitating subsequent processing.

[0283] Step 4:

[0284] The server uses natural language processing technology to analyze the query, which has been formatted in a unified format, and extracts key keywords and phrases.

[0285] Step 5:

[0286] The server searches the database for similar queries from the past, and identifies the most similar query based on the extracted keywords.

[0287] Step 6:

[0288] The server evaluates the results of similar queries based on relevance scores and selects the most appropriate answers.

[0289] Step 7:

[0290] The server uses AI to generate an automated response message based on the selected answer, utilizing extracted keywords and past inquiry data.

[0291] Step 8:

[0292] The server uses an emotion engine to recognize the user's emotions and determines their emotions from the content of the user's query and the way they input it.

[0293] Step 9:

[0294] The server adjusts the tone and content of the automated response message depending on the emotion recognized by the emotion engine. For example, if the user expresses anger or frustration, the response message will be more polite and empathetic.

[0295] Step 10:

[0296] The server then sends the generated automated response message to the user's terminal, allowing the user to quickly obtain a solution.

[0297] Step 11:

[0298] The user checks the automatic response message on the terminal, which allows the user to immediately obtain a solution to the problem.

[0299] Step 12:

[0300] The server re-analyzes the inquiry content and identifies an appropriate department if one has not been set.

[0301] Step 13:

[0302] The server forwards the query to the appropriate identified department, and the query is automatically redirected to ensure it reaches the correct department.

[0303] Step 14:

[0304] The terminal (in the appropriate department) receives the inquiry and provides the information the user is looking for, allowing for a prompt and professional response.

[0305] By following the above steps, the system of the present invention can efficiently process user inquiries and provide appropriate responses. The introduction of the emotion engine will provide a higher quality user experience.

[0306] Example 2

[0307] 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."

[0308] In conventional inquiry response systems, user inquiries are often handled manually, which results in long response times and reduced user satisfaction. In addition, inquiries are sometimes not accurately routed to the appropriate department based on their content, reducing response efficiency. Furthermore, a uniform response is returned without considering the user's feelings, which also results in a poor user experience.

[0309] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0310] In this invention, the server includes means for receiving inquiry content, means for formatting the received inquiry content into a unified format, means for analyzing the inquiry content and extracting key keywords, means for searching past similar inquiries and identifying the most similar inquiry, means for generating an automatic response message using a generation AI, means for recognizing a user's emotion, means for adjusting the content of the automatic response message according to the recognized emotion, means for sending the generated response message to the user, means for reanalyzing the inquiry content and identifying an appropriate department, means for transferring the inquiry to the identified appropriate department, and means for providing a user interface for accessing the inquiry portal. This dramatically improves the efficiency of inquiry response, improves user satisfaction, enables accurate department transfers, and enables flexible responses according to the user's emotion.

[0311] "Inquiry content" refers to information that a user inputs and transmits to the system through the inquiry portal.

[0312] "Receiving" refers to the process in which the server acquires the inquiry content sent by the user.

[0313] A "uniform format" is a standardized data format that is formatted to facilitate analysis and processing.

[0314] "Analysis" is the process of extracting and understanding meaning and keywords from the content of a query.

[0315] "Keywords" refer to the main concepts or phrases in the query content.

[0316] A "similar query" refers to a query sent by a user in the past that is similar to the current query.

[0317] "Generative AI" refers to a system that uses artificial intelligence to generate automated response messages.

[0318] An "automatic response message" is a response created by a generation AI and automatically sent to the user.

[0319] "Emotion recognition" is a technology that recognizes the emotional state of a user from the content of their inquiry.

[0320] "Adjustment" is the process of changing the tone and content of a response message based on the results of emotion recognition.

[0321] "Sending" is the process by which the server delivers the generated automated response message to the user.

[0322] "Reanalysis" is the process of reviewing the results of an initial analysis to improve accuracy.

[0323] The "appropriate department" refers to the department that can respond to the inquiry most quickly and accurately based on the content of the inquiry.

[0324] "Forwarding" is the process by which the server routes the query to the appropriate department.

[0325] "User Interface" means the interface through which a user accesses the system and inputs and submits inquiries.

[0326] The present invention is a system for automating responses to user inquiries, and aims to improve the efficiency and productivity of inquiries. This system includes the following processes.

[0327] Receiving inquiries

[0328] The user accesses a dedicated inquiry portal, enters their question in the inquiry form, and submits it. For example, they might enter, "I forgot my password. Please tell me how to reset it." The server retrieves the received inquiry and stores it in a MySQL database or similar.

[0329] Formatting to a unified format

[0330] The server formats the received query content into a unified format, for example, by removing unnecessary line breaks and spaces from the received text and converting it to JSON format.

[0331] Content Analysis

[0332] The server uses a natural language processing (NLP) library (e.g., spaCy or NLTK) to parse the query, specifically by performing morphological analysis to extract key keywords and phrases such as "password," "forgot," and "reset."

[0333] Search for similar past inquiries

[0334] The server generates a search query and searches a database for similar queries, calculating a similarity score using techniques such as TF-IDF and Word2Vec, identifying the most relevant query, and retrieving its content and answer.

[0335] Generate auto-response messages

[0336] The server creates a generation prompt for the generative AI model (for example, OpenAI's GPT-3). The prompt text includes the extracted keywords and the content of similar queries. For example, a prompt text such as "A user has asked how to reset their forgotten password. Please follow the steps below to reset it..." is input to the generative AI. The generated automated response message is then obtained.

[0337] Emotion Recognition and Regulation

[0338] The server uses an emotion engine (for example, Microsoft® Azure® Text Analytics API) to recognize emotions from the user's inquiry. For example, if the user expresses dissatisfaction or anger, the server adjusts the tone and content of the generated response message. For example, it might make the message more polite, such as "We apologize for the inconvenience. Please follow the steps below to reset your password."

[0339] Send to user

[0340] The server sends the generated automatic response message to the user's terminal, so that the user can receive the response quickly.

[0341] Redirecting to the appropriate department

[0342] The server re-analyzes the inquiry, and if the appropriate department is not set, it automatically redirects the inquiry to the appropriate department. For example, if an inquiry that was initially misidentified as a network problem is re-analyzed to be a hardware problem, the server redirects the inquiry to the IT support department. The terminal (for example, a terminal in the IT support department) responds appropriately to the received inquiry and provides the user with a solution.

[0343] The above is an embodiment of the present invention, and this system dramatically improves the efficiency of inquiries and reduces the man-hours of both users and responders. The introduction of an emotion engine can provide an even higher level of user experience.

[0344] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0345] Step 1: Receiving an inquiry

[0346] A user accesses a dedicated inquiry portal, enters a question in the inquiry form, and submits it. For example, the user might enter, "I forgot my password, please tell me how to reset it." The server receives the HTTP request sent from the portal and extracts its contents. The input is the user's inquiry text, and the output is the received inquiry text.

[0347] Step 2: Formatting into a unified format

[0348] The server formats the received query content into a unified format. Specifically, it removes unnecessary line breaks and spaces from the received text and converts it into JSON format. The input is the received query text, and the output is the formatted JSON data.

[0349] Step 3: Content Analysis

[0350] The server uses a natural language processing (NLP) library (e.g., spaCy or NLTK) to parse the query formatted in a unified format. Specifically, it performs morphological analysis to extract key keywords and phrases such as "password," "forgot," and "reset." The input is formatted JSON data, and the output is a list of extracted keywords.

[0351] Step 4: Search for similar past queries

[0352] The server generates a search query and searches a database for similar queries, calculating a similarity score using methods such as TF-IDF or Word2Vec. The input is a list of extracted keywords, and the output is related queries and their answers.

[0353] Step 5: Generate an auto-response message

[0354] The server creates a generation prompt for the generative AI model (for example, OpenAI's GPT-3). The prompt text includes the extracted keywords and the content of similar queries. For example, a prompt text such as "A user has asked how to reset their forgotten password. Please follow the steps below to reset it..." is input to the generative AI. The generative AI generates a response message, and the server receives that message. The input is the prompt text, and the output is the generated automated response message.

[0355] Step 6: Emotional awareness and regulation

[0356] The server uses an emotion engine (for example, Microsoft Azure's Text Analytics API) to recognize emotions from the user's inquiry. Specifically, it performs emotion analysis to identify the emotion expressed by the user (for example, anger, frustration, or joy). If the user expresses frustration or anger, it adjusts the tone and content of the generated response message. The input is the generated automated response message and the user's emotion data, and the output is the adjusted response message.

[0357] Step 7: Send to users

[0358] The server sends the generated automatic response message to the user's terminal. Specifically, it returns the generated message to the user's terminal as an HTTP response. The input is the adjusted response message, and the output is the response message sent to the user.

[0359] Step 8: Redirect to the appropriate department

[0360] The server re-analyzes the inquiry content, and if the appropriate department is not set, it automatically identifies the department and redirects the user. Specifically, it performs NLP analysis again to extract keywords to identify the appropriate department. It then sends an email or notification to that department. The input is the inquiry content and the re-analysis results, and the output is the redirected notification. The terminal (for example, a terminal in the IT support department) responds appropriately to the received inquiry and provides the user with a solution.

[0361] (Application example 2)

[0362] 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."

[0363] In modern brick-and-mortar stores, it is important to respond to customer inquiries quickly and appropriately, but this requires a lot of manpower and time, which is a challenge. Furthermore, it is also necessary to improve customer satisfaction by appropriately recognizing customer emotions and providing appropriate responses. However, a system for doing this efficiently has not yet been fully developed.

[0364] 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.

[0365] In this invention, the server includes means for receiving the inquiry content, means for formatting the received inquiry content into a unified format, means for analyzing the inquiry content and extracting key keywords, means for searching for similar past inquiries and identifying the most similar inquiry, means for generating an automatic response message using a generative AI, means for recognizing the user's emotions using a sentiment analysis engine and adjusting the response message, means for sending the generated response message to the user, means for reanalyzing the inquiry content and identifying an appropriate department, means for forwarding the inquiry to the identified appropriate department, means for providing a user interface for accessing the inquiry portal, means for quickly and accurately interacting with users in physical stores using terminals, means for using a generative AI model to generate a response message based on the customer's emotions, and means for generating the automatic response message using prompt sentences, thereby enabling efficiency and productivity improvement in inquiry response.

[0366] The "means for receiving the inquiry content" is an interface for the system to take in the inquiry content input by the user.

[0367] The "means for converting the received query content into a unified format" is a process for converting the received query content into a format that is easy to parse and process.

[0368] "Means for analyzing the content of the query and extracting key keywords" refers to a means for extracting important information from the query using natural language processing technology.

[0369] The "means for searching past similar queries and identifying the most similar query" is a technique for finding an example that is closest to the current query from past queries in a database.

[0370] "Means for generating automated response messages using generative AI" refers to a process that uses AI technology to generate appropriate response messages in response to inquiries.

[0371] "Means for recognizing a user's emotions using an emotion analysis engine and adjusting a response message" is a technology for detecting emotions from a user's input text and adjusting the tone and content of a response message.

[0372] The "means for transmitting the generated response message to the user" refers to a communication means for transmitting the generated response message to the user's terminal.

[0373] The "means for re-analyzing the inquiry content and identifying the appropriate department" is a process for re-analyzing the inquiry content and automatically identifying the appropriate department in charge.

[0374] The "means for transferring the inquiry to the specified appropriate department" is a technique for redirecting the inquiry content to the specified department.

[0375] "Means for providing a user interface for accessing the inquiry portal" refers to a method for providing an interface for a user to submit an inquiry to the system.

[0376] "Means for quickly and accurately communicating with users in a physical store using a terminal" refers to a method for efficiently communicating with customers via a terminal located in a physical store.

[0377] "Means for using a generative AI model to generate a response message based on customer sentiment" refers to an AI model for generating a response that takes into account customer sentiment information.

[0378] The "means for generating an automatic response message using a prompt sentence" is a method for generating an appropriate automatic response message based on a specified prompt sentence.

[0379] In order to carry out the present invention, it is important to build a system that combines a server and a terminal when dealing with customers in a brick-and-mortar store.

[0380] Server processing overview

[0381] The server efficiently processes customer inquiries using the following methods:

[0382] 1. Receiving and storing your inquiry:

[0383] The user enters the inquiry using a terminal installed in the physical store and sends it to the server. The server receives the inquiry and stores it in a database. This step requires a network connection and database software (e.g., MySQL).

[0384] 2. Formatting to Unified Format:

[0385] The server formats the received query content into a unified format for easy parsing. This is done using a script (e.g., a Python script) that performs text preprocessing.

[0386] 3. Extracting Key Keywords:

[0387] The server analyzes the query content using NLP techniques (e.g., NLTK library) and extracts key keywords, which are then used in the next process.

[0388] 4. Search for similar queries from the past:

[0389] The server searches the database for similar queries and identifies the most similar queries. This step utilizes a search algorithm (e.g., Elasticsearch).

[0390] 5. Generate auto-response messages:

[0391] The server generates an automated response message using a generative AI (e.g., GPT-2 model) based on the extracted keywords and identified similar queries, calling the model using an API (e.g., Hugging Face Transformers).

[0392] 6. Emotion recognition and regulation:

[0393] The server uses an emotion analysis engine (e.g., SentimentIntensityAnalyzer) to recognize the user's emotions and adjust the tone of the response message: if the user is expressing frustration, the response message will have a more polite and empathetic tone.

[0394] 7. Sending a response message:

[0395] The server then sends the generated automatic response message back to the user's terminal, allowing the user to receive a prompt reply.

[0396] 8. Redirect to the appropriate department:

[0397] The server re-parses the query and, if the appropriate department is not set, forwards the query to the correct department, along with the information to whom it should be assigned.

[0398] Use of concrete examples and prompts

[0399] For example, if a user types "I can't find the product in the store. Help me!" into a terminal in a physical store, the server receives the content, formats it into a unified format, and extracts keywords such as "product," "can't find," and "help." Sentiment analysis determines that the user's emotion is frustration. Based on this information, the following prompt sentence is provided to the generation AI:

[0400] Prompt: "Generate sentences containing the following keywords: product, not found, help. Sentiment: negative."

[0401] The response message generated using this prompt is "We apologize for the inconvenience. If you cannot find the product, please speak to a store staff member or check this link." The server then sends this message to the user.

[0402] As described above, the present invention is a system that can improve the efficiency and productivity of responding to inquiries and increase customer satisfaction in physical stores.

[0403] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0404] Step 1:

[0405] The user inputs the inquiry details using a terminal installed in the physical store, and then sends the input inquiry details to the server by pressing the send button.

[0406] Input: User's query (e.g. "I can't find the item in your store. Please help!")

[0407] Output: Received query data

[0408] Step 2:

[0409] The server formats the received query into a unified format, preprocessing the text to convert the query to lower case and removing special characters and unnecessary spaces.

[0410] Input: Received inquiry data

[0411] Output: Query data formatted in a unified format

[0412] Step 3:

[0413] The server uses NLP techniques to parse the query and extract key keywords, then uses the NLTK library to tokenize it and extract important nouns and verbs.

[0414] Input: Query data formatted in a unified format

[0415] Output: Extracted keywords (e.g. "product", "cannot find", "help")

[0416] Step 4:

[0417] The server searches the database for similar queries from the past, identifies the most similar queries, calculates a similarity score based on the extracted keywords using Elasticsearch, and identifies the query with the highest score.

[0418] Input: Extracted keywords

[0419] Output: Most similar previous queries

[0420] Step 5:

[0421] The server uses generative AI to generate an automated response message based on the most similar past queries, generating prompt sentences and feeding them into a GPT-2 model to generate an appropriate response message.

[0422] Input: Most similar previous query, prompt (e.g., "Generate a sentence containing the following keywords: product, not found, help. Sentiment: negative.")

[0423] Output: Generated automated response message (e.g., "We apologize for the inconvenience. If you cannot find the product, please contact a store staff member or check this link.")

[0424] Step 6:

[0425] The server uses a sentiment analysis engine to recognize the user's emotions and adjusts the response message. It calculates an emotion score using the SentimentIntensityAnalyzer and changes the tone and content of the response message as needed.

[0426] Input: Generated auto-response message, user inquiry

[0427] Output: Reconciled response message

[0428] Step 7:

[0429] The server then sends the generated response message back to the user's terminal, using a communication protocol to transfer the message so that the user can receive a response quickly.

[0430] Input: Tailored response message

[0431] Output: The response message displayed on the user's terminal.

[0432] Step 8:

[0433] The server reparses the query and redirects the query to the correct department if the appropriate department is not set. Based on the results of the reparse, the server redirects the query.

[0434] Input: Tailored response message, inquiry content

[0435] Output: Enquiry data redirected to the appropriate department

[0436] 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.

[0437] 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 (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.

[0438] 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.

[0439] [Second embodiment]

[0440] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0441] 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.

[0442] 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).

[0443] 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.

[0444] 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.

[0445] 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).

[0446] 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.

[0447] 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.

[0448] 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.

[0449] 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.

[0450] 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.

[0451] 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."

[0452] This invention is a system that automates responses to inquiries from users, aiming to improve the efficiency and productivity of inquiry responses. This system receives inquiry content, formats it into a unified format, analyzes it to extract key keywords, searches for similar past inquiries, and uses generative AI to automatically generate a response message that can be sent to the user. It can also reanalyze the inquiry content to identify the appropriate department and redirect the inquiry to the identified department.

[0453] Program processing overview

[0454] 1. Receiving an Inquiry

[0455] Users access a dedicated inquiry portal and enter their questions in an inquiry form.

[0456] When the user presses the send button, the inquiry is sent to the server.

[0457] The server stores the received inquiry in a database.

[0458] 2. Formatting into a unified format

[0459] The server formats the raw query content it receives into a predefined format, which makes subsequent processing easier.

[0460] 3. Content Analysis

[0461] The server analyzes the formatted query using natural language processing technology and extracts key keywords.

[0462] 4. Search for similar queries from the past

[0463] The server searches a database for similar past queries and identifies the most similar queries.

[0464] The server scores the search results and selects those with the highest similarity.

[0465] 5. Generate Auto-Response Messages

[0466] The server uses a generation AI to generate an automatic response message based on the extracted keywords and selected past inquiries.

[0467] 6. Transmission to the User

[0468] The server sends the generated automatic response message to the user's terminal.

[0469] The user checks the sent response message on the terminal.

[0470] 7. Redirect to the appropriate department

[0471] The server re-analyzes the query content and identifies an appropriate department if one has not been set.

[0472] The server forwards the query to the specified department if redirection is necessary.

[0473] A terminal (in the appropriate department) receives the inquiry and provides the information the user requests.

[0474] Specific examples

[0475] Example 1: Inquiry about network trouble

[0476] A user enters a query into the portal, such as "What should I do if my network is unstable?" and presses the send button. The server receives this query and stores it in a database. The server then formats the query based on keywords such as "network," "unstable," and "how to deal with it," and analyzes it using natural language processing technology. It then searches the database for similar past queries and identifies the most similar case. Generative AI is used to generate an automated response message (e.g., "If your network is unstable, please follow the steps below...") and sends it to the user. When the user receives the response message, it becomes easier for them to solve the problem themselves.

[0477] Example 2: Redirection due to misidentification of inquiry content

[0478] If a user sends a query saying "My PC won't start," but the server initially analyzes it and tries to treat it as a network problem, the server will reanalyze it and determine that the query is actually a hardware problem. Therefore, the query will be automatically redirected to the IT support department. The terminal (the IT support department's terminal) will receive this query and provide an appropriate solution.

[0479] In this way, the present invention can dramatically improve the efficiency of responding to inquiries and reduce the man-hours of both users and responders.

[0480] The processing flow will be explained below.

[0481] Step 1:

[0482] The user accesses a dedicated inquiry portal and enters the question into the inquiry form. The user also fills in the necessary information in the input fields and presses the send button to send the inquiry.

[0483] Step 2:

[0484] The server receives the inquiry sent by the user, and the received data is stored in a database for record keeping.

[0485] Step 3:

[0486] The server formats the received queries into a unified format, converting raw data into a defined format for easier subsequent processing.

[0487] Step 4:

[0488] The server analyzes the query using natural language processing technology, specifically extracting key keywords and phrases and obtaining key information needed to understand the query.

[0489] Step 5:

[0490] The server searches the database for similar past queries, identifies the most similar past queries based on the extracted keywords, and calculates a relevance score.

[0491] Step 6:

[0492] The server selects the query with the highest similarity from the search results and prepares to generate a response message based on the selected past query data.

[0493] Step 7:

[0494] The server uses a generation AI to generate an automated response message, creating a specific and appropriate response message based on selected past inquiry data and extracted keywords.

[0495] Step 8:

[0496] The server then sends the generated automated response message to the user's terminal, allowing the user to receive an immediate response.

[0497] Step 9:

[0498] The user checks the automatic response message on the terminal, which allows the user to quickly find a way to resolve the problem.

[0499] Step 10:

[0500] The server re-parses the query and identifies the appropriate department if one is not already set, so that the query can be re-evaluated and handled by the correct department.

[0501] Step 11:

[0502] The server then forwards the query to the appropriate department, and the query is automatically redirected to the correct department.

[0503] Step 12:

[0504] The terminal (the terminal in the appropriate department) receives the inquiry and provides the information the user is looking for. The department staff responds, allowing the user to receive professional support.

[0505] Through these steps, inquiries from users are handled efficiently and quickly. By automating the entire inquiry response process, the burden on staff is reduced and users can have their problems resolved quickly.

[0506] Example 1

[0507] 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."

[0508] Conventional inquiry response systems analyzed the content of inquiries individually and responded manually, resulting in issues with efficiency and productivity. Furthermore, they did not redirect inquiries to the appropriate department or generate automated response messages, increasing the likelihood of incorrect responses. Furthermore, responding to inquiries required time and effort, which led to lower user satisfaction. This resulted in increased response workloads across the company and a decline in overall performance.

[0509] 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.

[0510] In this invention, the server includes: means for receiving inquiry content; means for formatting the received inquiry content into a unified format; means for analyzing the inquiry content and extracting key keywords; means for searching for similar past inquiries and identifying the most similar inquiry; means for generating an automatic response message using a generative AI model; means for sending the generated response message to the user; means for providing a redirect function to an appropriate department; means for reanalyzing the inquiry content and identifying an appropriate department; means for forwarding the inquiry to the identified appropriate department; means for providing a user interface for accessing the inquiry portal; means for scoring answer candidates and selecting the most appropriate answer; means for analyzing the inquiry content using natural language processing technology; and means for generating a prompt sentence based on the analysis result and inputting it into the generative AI model. This automates the inquiry content analysis and response process, enabling prompt and accurate responses. Furthermore, adding a redirect function to the appropriate department improves user satisfaction and reduces response time.

[0511] "Inquiry content" is text data of questions, requests, and inquiries sent by users through the system.

[0512] The "receiving means" refers to the technical means for inputting the contents of the inquiry sent by the user into the server.

[0513] The "means for formatting into a unified format" refers to a technical means for converting the received query content into a format that is easy to process for subsequent analysis.

[0514] The "analysis means" is a natural language processing technique for extracting key keywords from the content of the inquiry.

[0515] A "keyword" is a word or phrase that contains important information in the query.

[0516] The "means for searching for similar past queries" refers to a technical means for finding similar past queries from a database.

[0517] A "generative AI model" is a system that uses artificial intelligence technology to automatically generate new response messages.

[0518] A "means for generating an auto-response message" is a technical means for automatically creating a response to a user's inquiry using a generative AI model.

[0519] "Means for sending" refers to the technical means for delivering the generated auto-reply message to the user.

[0520] The "redirect function to the appropriate department" is a function that transfers an inquiry to the appropriate department based on the content of the inquiry.

[0521] "Reanalysis measures" are technical measures that perform additional analysis when the results of the initial analysis are unclear.

[0522] "Means of identification" refers to technical means for determining the appropriate department to respond based on reanalysis.

[0523] "Forwarding means" refers to the technical means by which the inquiry is redirected to a specified department.

[0524] "User Interface" means the means by which a user accesses the inquiry portal.

[0525] "Scoring means" refers to a technical means for evaluating detected answer candidates and selecting the most appropriate answer.

[0526] "Natural language processing technology" is an artificial intelligence technology for understanding, analyzing, and processing human language.

[0527] A "prompt sentence" is an instruction sentence that is input into a generative AI model to generate a response.

[0528] The present invention is a system that automates responses to user inquiries, thereby improving the efficiency and productivity of inquiry responses. This system functions through communication between a server, terminals, and users.

[0529] Hardware and software used

[0530] The server is a high-performance computer system, and uses a database management system (DBMS) such as MySQL or PostgreSQL. Natural language processing engines such as TensorFlow and Watson are used for server-side natural language processing. Furthermore, OpenAI's GPT-3 and other generative AI models are used.

[0531] A terminal is a device where a user inputs a query and receives a response message. This can be a PC, tablet, smartphone, or other device. The user interface is implemented as a web application or a mobile application.

[0532] Processing Overview

[0533] 1. Receiving an Inquiry

[0534] Users access a dedicated inquiry portal and enter their questions in the inquiry form. For example, they might enter, "Please tell me what to do if the network is unstable."

[0535] When the user presses the send button, the query content is sent to the server as an HTTP request.

[0536] The server stores the received inquiry in a database.

[0537] 2. Formatting into a unified format

[0538] The server formats the received query content into a specific format (e.g., JSON format), which makes subsequent processing easier.

[0539] 3. Content Analysis

[0540] The server analyzes the formatted query using a natural language processing engine (e.g., TensorFlow) and extracts key keywords, such as "network," "unstable," and "how to respond."

[0541] 4. Search for similar queries from the past

[0542] The server searches the database to identify similar past queries, and the search results are scored to select the most similar queries.

[0543] 5. Generate Auto-Response Messages

[0544] The server uses a generative AI model (e.g., GPT-3) to generate an automated response message based on the extracted keywords and selected past inquiries. The prompt text is "Please generate an automated response message in response to the user's inquiry, 'Please tell me what to do if the network is unstable.'"

[0545] 6. Transmission to the User

[0546] The server then sends the generated auto-reply message to the user's device via email or a notification API (e.g., Firebase Cloud Messaging).

[0547] The user checks the response message sent and uses it to solve the problem.

[0548] 7. Redirect to the appropriate department

[0549] The server re-analyzes the query and identifies the appropriate department.

[0550] The server redirects the inquiry to the specified department, for example, contacting the IT support department via email or an internal notification system.

[0551] The terminal (the terminal of the appropriate department) receives this inquiry and initiates specific actions.

[0552] Specific examples

[0553] Example 1: Inquiry about network trouble

[0554] When a user enters "What to do if the network is unstable" into the inquiry portal and submits it, the server receives the inquiry and stores it in a database. Next, the inquiry is formatted based on keywords such as "network," "unstable," and "how to deal with it," and analyzed using natural language processing technology. After that, similar past inquiries are searched for in the database to identify the most similar case. Using generative AI, an automatic response message is generated that reads, "If the network is unstable, please follow the steps below..." and sent to the user. When the user receives the response message, it becomes easier for them to solve the problem themselves.

[0555] Example 2: Redirection due to misidentification of inquiry content

[0556] If a user sends a query saying "My PC won't start," but the server initially analyzes it and treats it as a network problem, the server reanalyzes it and determines that the query is actually a hardware problem. Therefore, the query is automatically redirected to the IT support department. The terminal (the IT support department's terminal) receives this query and can provide an appropriate solution.

[0557] In this way, the present invention can dramatically improve the efficiency of responding to inquiries and reduce the man-hours of both users and responders.

[0558] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0559] Step 1:

[0560] The user accesses the inquiry portal, enters the question in the inquiry form, and presses the send button.

[0561] Input: The inquiry entered by the user (e.g., "Please tell me what to do if the network is unstable").

[0562] Output: The query stored in the database.

[0563] Specific operation: The server stores the query content received as an HTTP request in a database. In this case, a database management system such as MySQL or PostgreSQL is used.

[0564] Step 2:

[0565] The server formats the received query content into a specified format (e.g., JSON format).

[0566] Input: The raw query stored in the database.

[0567] Output: Pretty-printed JSON representation of the query.

[0568] What happens: The server converts the query to JSON format, removing line breaks and special characters. This formatting makes it easier to parse later.

[0569] Step 3:

[0570] The server analyzes the formatted query using a natural language processing (NLP) engine and extracts key keywords.

[0571] Input: Query content in JSON format.

[0572] Output: Extracted key keywords (e.g., "network," "unstable," "how to respond").

[0573] Specific operation: The server uses a natural language processing engine such as TensorFlow or Watson to analyze the query content and identify important keywords.

[0574] Step 4:

[0575] The server searches a database to identify similar past queries.

[0576] Input: Extracted main keywords.

[0577] Output: The most similar previous queries.

[0578] Specific operation: The server generates an SQL query based on the extracted keywords and searches the database. The search results are scored using a search engine such as Elasticsearch, and the most similar results are selected.

[0579] Step 5:

[0580] The server generates the automated response message using the generative AI model.

[0581] Input: The most similar previous query, extracted keywords, and prompt (e.g., "Please generate an auto-response message in response to the user's query 'What should I do if my network is unstable?'").

[0582] Output: Auto-reply message (e.g. "If your network is unstable, please try the following steps...").

[0583] Specific operation: The server inputs a prompt sentence into a generative AI model such as OpenAI's GPT-3 to generate an automated response message.

[0584] Step 6:

[0585] The server sends the generated automatic response message to the user's terminal.

[0586] Input: Auto-reply message.

[0587] Output: The response message that is displayed on the user's terminal.

[0588] Specific operation: The server sends the generated message to the user using a notification API such as email or Firebase Cloud Messaging. The user then checks the message on their device.

[0589] Step 7:

[0590] The server re-analyzes the query and identifies the appropriate department.

[0591] Input: Query content, reanalysis instructions.

[0592] Output: The appropriate department identified.

[0593] Specific operation: The server uses the natural language processing engine again to re-analyze the inquiry content and identify the appropriate department.

[0594] Step 8:

[0595] The server redirects the query to the specified department.

[0596] Input: Identified appropriate department, inquiry details.

[0597] Output: Query redirected to the appropriate department's terminal.

[0598] Specific operation: The server forwards the inquiry to the specified department using an internal notification system or email. The appropriate department's terminal receives the inquiry and begins to respond.

[0599] (Application example 1)

[0600] 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."

[0601] Current inquiry response systems have difficulty responding quickly and accurately to a wide variety of inquiries from users. They also face the problem of increasing the workload of responders, reducing the efficiency of customer service. Furthermore, it is difficult to provide consistent responses to similar inquiries, making it difficult to improve customer satisfaction.

[0602] 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.

[0603] In this invention, the server includes means for receiving the inquiry content, means for formatting the received inquiry content into a unified format, means for analyzing the inquiry content and extracting key keywords, means for searching past similar inquiries and identifying the most similar inquiry, means for generating a prompt sentence using a generative AI model and generating an automatic response message, means for transmitting the generated response message to an information processing device, and means for reanalyzing the inquiry content, identifying an appropriate department if an appropriate department has not been set, and transferring the inquiry to the appropriate department. This dramatically improves the efficiency of inquiry response and reduces the man-hours of both users and responders.

[0604] The "means for receiving the inquiry content" is a function for transmitting the inquiry content submitted by the user to the server via the network, and for the server to receive the content.

[0605] The "means for converting received inquiry content into a unified format" is a function for converting inquiry content sent in various formats into a consistent unified format.

[0606] "Means for analyzing the content of inquiries and extracting key keywords" refers to a function that utilizes natural language processing technology to identify and extract key keywords from the content of inquiries.

[0607] The "means for searching past similar queries and identifying the most similar query" is a function for comparing a query received from a database in the past with the most similar query and identifying the most similar query.

[0608] "Means for generating prompt sentences using a generative AI model and generating automatic response messages" refers to a function that uses a pre-trained generative AI model to generate prompt sentences required for a response and creates automatic response messages based on those sentences.

[0609] The "means for transmitting the generated response message to the information processing device" is a function for transferring the generated response message to the user's information processing device (for example, a smartphone or a computer) and displaying it.

[0610] "Means for reanalyzing the content of the inquiry, identifying an appropriate department if an appropriate department has not been set, and transferring the inquiry to the relevant department" is a function that reanalyzes the content of the inquiry, and if an appropriate department has not been set, identifies the most appropriate department and transfers the inquiry to that department.

[0611] MODE FOR CARRYING OUT THE INVENTION

[0612] System Configuration

[0613] The system implemented in this invention is broadly composed of the following main means.

[0614] 1. Means of receiving inquiries

[0615] 2. A means of formatting received inquiries into a unified format

[0616] 3. A method for analyzing inquiries and extracting key keywords

[0617] 4. A means to search for similar queries from the past and identify the most similar queries

[0618] 5. A means of generating prompts and auto-response messages using generative AI models

[0619] 6. Means for transmitting the generated response message to the information processing device

[0620] 7. A means to re-analyze the content of the inquiry, identify the appropriate department if one is not set, and transfer the inquiry to the appropriate department.

[0621] Program processing

[0622] Hardware and software usage

[0623] server:

[0624] Hardware: High-performance server

[0625] Software: Database management systems (e.g., MySQL), search engines (e.g., Elasticsearch), natural language processing libraries (e.g., Spacy, BERT), generative AI models (e.g., ChatGPT, GPT-3)

[0626] User device:

[0627] Hardware: Smartphones, tablets, computers

[0628] Software: Mobile application, browser-based interface

[0629] Data processing and calculation flow

[0630] 1. Receiving an Inquiry

[0631] Users access the inquiry form using a smartphone or PC, enter their question, and submit it. The entered inquiry is sent to a server via a network and stored in a database.

[0632] 2. Formatting into a unified format

[0633] The server converts the received query content into a unified format, using a text formatting library (e.g., Python's NLTK) for this conversion process.

[0634] 3. Content Analysis

[0635] The server analyzes the query using natural language processing techniques (e.g., Spacy, BERT) to extract key keywords. This analysis identifies the core of the query.

[0636] 4. Search for similar queries from the past

[0637] The server compares the queries against previous queries in a database to find the most similar queries, using a search engine such as Elasticsearch to score the similarity and identify highly similar queries.

[0638] 5. Generate Auto-Response Messages

[0639] The server uses a generative AI model (e.g., ChatGPT, GPT-3) to generate an automated response message based on the prompt, which includes extracted keywords and information about similar past inquiries.

[0640] 6. Transmission to the User

[0641] The server generates a response message and sends it to the user's device. A real-time notification service (e.g., Firebase) is used to deliver the message to the user quickly and reliably.

[0642] 7. Redirect to the appropriate department

[0643] The server re-analyzes the inquiry and, if an appropriate department is not set, identifies the appropriate department and transfers the inquiry. This allows inquiries that require specialized assistance to be handled appropriately.

[0644] Specific examples

[0645] Example 1: Delivery enquiry

[0646] A user submits an inquiry saying, "The item I ordered hasn't arrived." This inquiry is received by the server, and key keywords such as "order," "item," and "not arrived" are extracted. Similar past inquiries are searched for, and the generative AI model generates a response message saying, "We will confirm your order number and let you know the delivery status. Delivery usually takes 5-7 days." This message is then promptly sent to the user's device.

[0647] Prompt Sentence Examples

[0648] "Please generate an auto-reply message for the following inquiry: 'My order hasn't arrived'"

[0649] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0650] Step 1:

[0651] The user accesses the smart mail-order support application, enters their question into the inquiry form, and presses the send button. At this time, the user may enter something like, "The item I ordered hasn't arrived." The entered data is saved in JSON format within the application and sent to the server via the network. The server receives the inquiry and simultaneously saves the information in a database. The input is the user's question, and the output is saved in the database.

[0652] Step 2:

[0653] The server formats the received raw query content into a specified unified format. This process uses a text formatting library (e.g., Python's NLTK). Specifically, it parses JSON-formatted data, removes unnecessary whitespace and special characters, and formats the query content. The input is the received raw query content, and the output is the data formatted in a unified format.

[0654] Step 3:

[0655] The server analyzes the formatted query using natural language processing technology (e.g., Spacy, BERT) and extracts key keywords. This results in keywords such as "order," "product," and "not delivered." The input is data formatted in a unified format, and the output is the key keywords.

[0656] Step 4:

[0657] The server uses a search engine (e.g., Elasticsearch) to search a database for similar past queries and identify the most similar queries. It scores the search results and selects queries with high similarity. The input is the main keywords, and the output is the past query data with high similarity.

[0658] Step 5:

[0659] The server uses a generative AI model (e.g., ChatGPT, GPT-3) to generate a prompt and generate an auto-response message. This prompt includes extracted keywords and information on the most similar past inquiries. For example, it could be in the format "Please generate an auto-response message in response to the following inquiry: 'The item I ordered has not arrived.'" The input is similar inquiry data, and the output is the generated response message.

[0660] Step 6:

[0661] The server sends the generated automatic response message to the user's device using a real-time notification service (e.g., Firebase). The user can check this response message within the application. The input is the generated response message, and the output is the notification and display on the user's device.

[0662] Step 7:

[0663] The server re-analyzes the query and identifies the appropriate department if one is not already set. If redirection is required, the server forwards the query to the appropriate department. The input is the user's query, and the output is a redirect to the identified appropriate department.

[0664] 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.

[0665] This invention is a system that automates responses to inquiries from users, aiming to improve the efficiency and productivity of inquiry responses. This system receives inquiry content, formats it into a unified format, analyzes it to extract key keywords, searches for similar past inquiries, and uses generative AI to generate an automatic response message that is sent to the user. It can also reanalyze the inquiry content to identify the appropriate department and redirect the inquiry to the identified department. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it also has the function of improving the quality of inquiry responses.

[0666] Program processing overview

[0667] 1. Receiving an Inquiry

[0668] Users access a dedicated inquiry portal, enter their questions into the inquiry form, and then press the send button to send the inquiry to the server.

[0669] The server checks the received inquiry and stores it in a database.

[0670] 2. Formatting into a unified format

[0671] The server converts the received raw query content into a unified format that can be easily parsed and processed.

[0672] 3. Content Analysis

[0673] The server uses NLP technology to analyze the query content, which has been formatted into a unified format, and extracts key keywords and phrases.

[0674] 4. Search for similar queries from the past

[0675] The server searches a database for similar queries from the past to identify the most similar queries, and the search results are evaluated based on a relevance score.

[0676] 5. Generate Auto-Response Messages

[0677] The server uses a generation AI to create an automated response message based on the extracted keywords and identified similar queries.

[0678] 6. Emotional Recognition and Regulation

[0679] The server uses an emotion engine to recognize emotions from the user's query content and expressions, and adjusts the tone and content of the generated response message according to the user's emotions.

[0680] For example, if the user is expressing anger or frustration, the response message will have a more polite and empathetic tone.

[0681] 7. Transmission to the User

[0682] The server sends the generated automatic response message to the user's terminal, so that the user can receive the response quickly.

[0683] 8. Redirect to the appropriate department

[0684] The server re-analyzes the query and, if the appropriate department is not set, forwards the query to the correct department.

[0685] The terminal (the terminal of the appropriate department) responds to the received inquiry and provides the user with an appropriate answer.

[0686] Specific examples

[0687] Example 1: Inquiry about network trouble

[0688] The user enters "Please tell me what to do if the network is unstable" into an inquiry form and submits it. The server receives the inquiry and saves it in a database. The server then formats the inquiry into a unified format and uses NLP technology to extract keywords such as "network," "unstable," and "how to deal with it." The server then searches the database for similar past inquiries and identifies the most similar one. It then uses generative AI to generate an automated response message, which the server then sends to the user. The server then uses an emotion engine to recognize the user's emotions. For example, if the user expresses frustration, the response message is adjusted to show empathy, such as "We apologize for the inconvenience. If the network is unstable, please follow the steps below..."

[0689] Example 2: Redirection due to misidentification of inquiry content

[0690] A user enters "My PC won't start" into an inquiry form and submits it. The server receives the inquiry and saves it in a database. The server analyzes the inquiry and extracts keywords such as "PC" and "won't start." An initial analysis identifies it as a network problem, but reanalysis identifies it as a hardware problem. In this case, the server automatically redirects the inquiry to the IT support department. A terminal (a terminal in the IT support department) receives the inquiry and provides an appropriate solution.

[0691] In this way, the present invention dramatically improves the efficiency of inquiries and reduces the man-hours required by both users and responders. The introduction of an emotion engine can provide an even higher level of user experience.

[0692] The processing flow will be explained below.

[0693] Step 1:

[0694] Users access a dedicated inquiry portal, enter their questions in the inquiry form, and then press the send button to send the inquiry.

[0695] Step 2:

[0696] The server receives the inquiry sent by the user, and the received inquiry is immediately saved in the database.

[0697] Step 3:

[0698] The server formats the raw query content it receives into a uniform format, ensuring data consistency and facilitating subsequent processing.

[0699] Step 4:

[0700] The server uses natural language processing technology to analyze the query, which has been formatted in a unified format, and extracts key keywords and phrases.

[0701] Step 5:

[0702] The server searches the database for similar queries from the past, and identifies the most similar query based on the extracted keywords.

[0703] Step 6:

[0704] The server evaluates the results of similar queries based on relevance scores and selects the most appropriate answers.

[0705] Step 7:

[0706] The server uses AI to generate an automated response message based on the selected answer, utilizing extracted keywords and past inquiry data.

[0707] Step 8:

[0708] The server uses an emotion engine to recognize the user's emotions and determines their emotions from the content of the user's query and the way they input it.

[0709] Step 9:

[0710] The server adjusts the tone and content of the automated response message depending on the emotion recognized by the emotion engine. For example, if the user expresses anger or frustration, the response message will be more polite and empathetic.

[0711] Step 10:

[0712] The server then sends the generated automated response message to the user's terminal, allowing the user to quickly obtain a solution.

[0713] Step 11:

[0714] The user checks the automatic response message on the terminal, which allows the user to immediately obtain a solution to the problem.

[0715] Step 12:

[0716] The server re-analyzes the inquiry content and identifies an appropriate department if one has not been set.

[0717] Step 13:

[0718] The server forwards the query to the appropriate identified department, and the query is automatically redirected to ensure it reaches the correct department.

[0719] Step 14:

[0720] The terminal (in the appropriate department) receives the inquiry and provides the information the user is looking for, allowing for a prompt and professional response.

[0721] By following the above steps, the system of the present invention can efficiently process user inquiries and provide appropriate responses. The introduction of the emotion engine will provide a higher quality user experience.

[0722] Example 2

[0723] 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."

[0724] In conventional inquiry response systems, user inquiries are often handled manually, which results in long response times and reduced user satisfaction. In addition, inquiries are sometimes not accurately routed to the appropriate department based on their content, reducing response efficiency. Furthermore, a uniform response is returned without considering the user's feelings, which also results in a poor user experience.

[0725] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0726] In this invention, the server includes means for receiving inquiry content, means for formatting the received inquiry content into a unified format, means for analyzing the inquiry content and extracting key keywords, means for searching past similar inquiries and identifying the most similar inquiry, means for generating an automatic response message using a generation AI, means for recognizing a user's emotion, means for adjusting the content of the automatic response message according to the recognized emotion, means for sending the generated response message to the user, means for reanalyzing the inquiry content and identifying an appropriate department, means for transferring the inquiry to the identified appropriate department, and means for providing a user interface for accessing the inquiry portal. This dramatically improves the efficiency of inquiry response, improves user satisfaction, enables accurate department transfers, and enables flexible responses according to the user's emotion.

[0727] "Inquiry content" refers to information that a user inputs and transmits to the system through the inquiry portal.

[0728] "Receiving" refers to the process in which the server acquires the inquiry content sent by the user.

[0729] A "uniform format" is a standardized data format that is formatted to facilitate analysis and processing.

[0730] "Analysis" is the process of extracting and understanding meaning and keywords from the content of a query.

[0731] "Keywords" refer to the main concepts or phrases in the query content.

[0732] A "similar query" refers to a query sent by a user in the past that is similar to the current query.

[0733] "Generative AI" refers to a system that uses artificial intelligence to generate automated response messages.

[0734] An "automatic response message" is a response created by a generation AI and automatically sent to the user.

[0735] "Emotion recognition" is a technology that recognizes the emotional state of a user from the content of their inquiry.

[0736] "Adjustment" is the process of changing the tone and content of a response message based on the results of emotion recognition.

[0737] "Sending" is the process by which the server delivers the generated automated response message to the user.

[0738] "Reanalysis" is the process of reviewing the results of an initial analysis to improve accuracy.

[0739] The "appropriate department" refers to the department that can respond to the inquiry most quickly and accurately based on the content of the inquiry.

[0740] "Forwarding" is the process by which the server routes the query to the appropriate department.

[0741] "User Interface" means the interface through which a user accesses the system and inputs and submits inquiries.

[0742] The present invention is a system for automating responses to user inquiries, and aims to improve the efficiency and productivity of inquiries. This system includes the following processes.

[0743] Receiving inquiries

[0744] The user accesses a dedicated inquiry portal, enters their question in the inquiry form, and submits it. For example, they might enter, "I forgot my password. Please tell me how to reset it." The server retrieves the received inquiry and stores it in a MySQL database or similar.

[0745] Formatting to a unified format

[0746] The server formats the received query content into a unified format, for example, by removing unnecessary line breaks and spaces from the received text and converting it to JSON format.

[0747] Content Analysis

[0748] The server uses a natural language processing (NLP) library (e.g., spaCy or NLTK) to parse the query, specifically by performing morphological analysis to extract key keywords and phrases such as "password," "forgot," and "reset."

[0749] Search for similar past inquiries

[0750] The server generates a search query and searches a database for similar queries, calculating a similarity score using techniques such as TF-IDF and Word2Vec, identifying the most relevant query, and retrieving its content and answer.

[0751] Generate auto-response messages

[0752] The server creates a generation prompt for the generative AI model (for example, OpenAI's GPT-3). The prompt text includes the extracted keywords and the content of similar queries. For example, a prompt text such as "A user has asked how to reset their forgotten password. Please follow the steps below to reset it..." is input to the generative AI. The generated automated response message is then obtained.

[0753] Emotion Recognition and Regulation

[0754] The server uses an emotion engine (for example, Microsoft Azure's Text Analytics API) to recognize emotions from the user's inquiry. For example, if the user expresses dissatisfaction or anger, the server adjusts the tone and content of the generated response message. For example, it might make the message more polite, such as, "We apologize for the inconvenience. Please follow the steps below to reset your password."

[0755] Send to user

[0756] The server sends the generated automatic response message to the user's terminal, so that the user can receive the response quickly.

[0757] Redirecting to the appropriate department

[0758] The server re-analyzes the inquiry, and if the appropriate department is not set, it automatically redirects the inquiry to the appropriate department. For example, if an inquiry that was initially misidentified as a network problem is re-analyzed to be a hardware problem, the server redirects the inquiry to the IT support department. The terminal (for example, a terminal in the IT support department) responds appropriately to the received inquiry and provides the user with a solution.

[0759] The above is an embodiment of the present invention, and this system dramatically improves the efficiency of inquiries and reduces the man-hours of both users and responders. The introduction of an emotion engine can provide an even higher level of user experience.

[0760] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0761] Step 1: Receiving an inquiry

[0762] A user accesses a dedicated inquiry portal, enters a question in the inquiry form, and submits it. For example, the user might enter, "I forgot my password, please tell me how to reset it." The server receives the HTTP request sent from the portal and extracts its contents. The input is the user's inquiry text, and the output is the received inquiry text.

[0763] Step 2: Formatting into a unified format

[0764] The server formats the received query content into a unified format. Specifically, it removes unnecessary line breaks and spaces from the received text and converts it into JSON format. The input is the received query text, and the output is the formatted JSON data.

[0765] Step 3: Content Analysis

[0766] The server uses a natural language processing (NLP) library (e.g., spaCy or NLTK) to parse the query formatted in a unified format. Specifically, it performs morphological analysis to extract key keywords and phrases such as "password," "forgot," and "reset." The input is formatted JSON data, and the output is a list of extracted keywords.

[0767] Step 4: Search for similar past queries

[0768] The server generates a search query and searches a database for similar queries, calculating a similarity score using methods such as TF-IDF or Word2Vec. The input is a list of extracted keywords, and the output is related queries and their answers.

[0769] Step 5: Generate an auto-response message

[0770] The server creates a generation prompt for the generative AI model (for example, OpenAI's GPT-3). The prompt text includes the extracted keywords and the content of similar queries. For example, a prompt text such as "A user has asked how to reset their forgotten password. Please follow the steps below to reset it..." is input to the generative AI. The generative AI generates a response message, and the server receives that message. The input is the prompt text, and the output is the generated automated response message.

[0771] Step 6: Emotional awareness and regulation

[0772] The server uses an emotion engine (for example, Microsoft Azure's Text Analytics API) to recognize emotions from the user's inquiry. Specifically, it performs emotion analysis to identify the emotion expressed by the user (for example, anger, frustration, or joy). If the user expresses frustration or anger, it adjusts the tone and content of the generated response message. The input is the generated automated response message and the user's emotion data, and the output is the adjusted response message.

[0773] Step 7: Send to users

[0774] The server sends the generated automatic response message to the user's terminal. Specifically, it returns the generated message to the user's terminal as an HTTP response. The input is the adjusted response message, and the output is the response message sent to the user.

[0775] Step 8: Redirect to the appropriate department

[0776] The server re-analyzes the inquiry content, and if the appropriate department is not set, it automatically identifies the department and redirects the user. Specifically, it performs NLP analysis again to extract keywords to identify the appropriate department. It then sends an email or notification to that department. The input is the inquiry content and the re-analysis results, and the output is the redirected notification. The terminal (for example, a terminal in the IT support department) responds appropriately to the received inquiry and provides the user with a solution.

[0777] (Application example 2)

[0778] 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."

[0779] In modern brick-and-mortar stores, it is important to respond to customer inquiries quickly and appropriately, but this requires a lot of manpower and time, which is a challenge. Furthermore, it is also necessary to improve customer satisfaction by appropriately recognizing customer emotions and providing appropriate responses. However, a system for doing this efficiently has not yet been fully developed.

[0780] 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.

[0781] In this invention, the server includes means for receiving the inquiry content, means for formatting the received inquiry content into a unified format, means for analyzing the inquiry content and extracting key keywords, means for searching for similar past inquiries and identifying the most similar inquiry, means for generating an automatic response message using a generative AI, means for recognizing the user's emotions using a sentiment analysis engine and adjusting the response message, means for sending the generated response message to the user, means for reanalyzing the inquiry content and identifying an appropriate department, means for forwarding the inquiry to the identified appropriate department, means for providing a user interface for accessing the inquiry portal, means for quickly and accurately interacting with users in physical stores using terminals, means for using a generative AI model to generate a response message based on the customer's emotions, and means for generating the automatic response message using prompt sentences, thereby enabling efficiency and productivity improvement in inquiry response.

[0782] The "means for receiving the inquiry content" is an interface for the system to take in the inquiry content input by the user.

[0783] The "means for converting the received query content into a unified format" is a process for converting the received query content into a format that is easy to parse and process.

[0784] "Means for analyzing the content of the query and extracting key keywords" refers to a means for extracting important information from the query using natural language processing technology.

[0785] The "means for searching past similar queries and identifying the most similar query" is a technique for finding an example that is closest to the current query from past queries in a database.

[0786] "Means for generating automated response messages using generative AI" refers to a process that uses AI technology to generate appropriate response messages in response to inquiries.

[0787] "Means for recognizing a user's emotions using an emotion analysis engine and adjusting a response message" is a technology for detecting emotions from a user's input text and adjusting the tone and content of a response message.

[0788] The "means for transmitting the generated response message to the user" refers to a communication means for transmitting the generated response message to the user's terminal.

[0789] The "means for re-analyzing the inquiry content and identifying the appropriate department" is a process for re-analyzing the inquiry content and automatically identifying the appropriate department in charge.

[0790] The "means for transferring the inquiry to the specified appropriate department" is a technique for redirecting the inquiry content to the specified department.

[0791] "Means for providing a user interface for accessing the inquiry portal" refers to a method for providing an interface for a user to submit an inquiry to the system.

[0792] "Means for quickly and accurately communicating with users in a physical store using a terminal" refers to a method for efficiently communicating with customers via a terminal located in a physical store.

[0793] "Means for using a generative AI model to generate a response message based on customer sentiment" refers to an AI model for generating a response that takes into account customer sentiment information.

[0794] The "means for generating an automatic response message using a prompt sentence" is a method for generating an appropriate automatic response message based on a specified prompt sentence.

[0795] In order to carry out the present invention, it is important to build a system that combines a server and a terminal when dealing with customers in a brick-and-mortar store.

[0796] Server processing overview

[0797] The server efficiently processes customer inquiries using the following methods:

[0798] 1. Receiving and storing your inquiry:

[0799] The user enters the inquiry using a terminal installed in the physical store and sends it to the server. The server receives the inquiry and stores it in a database. This step requires a network connection and database software (e.g., MySQL).

[0800] 2. Formatting to Unified Format:

[0801] The server formats the received query content into a unified format for easy parsing. This is done using a script (e.g., a Python script) that performs text preprocessing.

[0802] 3. Extracting Key Keywords:

[0803] The server analyzes the query content using NLP techniques (e.g., NLTK library) and extracts key keywords, which are then used in the next process.

[0804] 4. Search for similar queries from the past:

[0805] The server searches the database for similar queries and identifies the most similar queries. This step utilizes a search algorithm (e.g., Elasticsearch).

[0806] 5. Generate auto-response messages:

[0807] The server generates an automated response message using a generative AI (e.g., GPT-2 model) based on the extracted keywords and identified similar queries, calling the model using an API (e.g., Hugging Face Transformers).

[0808] 6. Emotion recognition and regulation:

[0809] The server uses an emotion analysis engine (e.g., SentimentIntensityAnalyzer) to recognize the user's emotions and adjust the tone of the response message: if the user is expressing frustration, the response message will have a more polite and empathetic tone.

[0810] 7. Sending a response message:

[0811] The server then sends the generated automatic response message back to the user's terminal, allowing the user to receive a prompt reply.

[0812] 8. Redirect to the appropriate department:

[0813] The server re-parses the query and, if the appropriate department is not set, forwards the query to the correct department, along with the information to whom it should be assigned.

[0814] Use of concrete examples and prompts

[0815] For example, if a user types "I can't find the product in the store. Help me!" into a terminal in a physical store, the server receives the content, formats it into a unified format, and extracts keywords such as "product," "can't find," and "help." Sentiment analysis determines that the user's emotion is frustration. Based on this information, the following prompt sentence is provided to the generation AI:

[0816] Prompt: "Generate sentences containing the following keywords: product, not found, help. Sentiment: negative."

[0817] The response message generated using this prompt is "We apologize for the inconvenience. If you cannot find the product, please speak to a store staff member or check this link." The server then sends this message to the user.

[0818] As described above, the present invention is a system that can improve the efficiency and productivity of responding to inquiries and increase customer satisfaction in physical stores.

[0819] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0820] Step 1:

[0821] The user inputs the inquiry details using a terminal installed in the physical store, and then sends the input inquiry details to the server by pressing the send button.

[0822] Input: User's query (e.g. "I can't find the item in your store. Please help!")

[0823] Output: Received query data

[0824] Step 2:

[0825] The server formats the received query into a unified format, preprocessing the text to convert the query to lower case and removing special characters and unnecessary spaces.

[0826] Input: Received inquiry data

[0827] Output: Query data formatted in a unified format

[0828] Step 3:

[0829] The server uses NLP techniques to parse the query and extract key keywords, then uses the NLTK library to tokenize it and extract important nouns and verbs.

[0830] Input: Query data formatted in a unified format

[0831] Output: Extracted keywords (e.g. "product", "cannot find", "help")

[0832] Step 4:

[0833] The server searches the database for similar queries from the past, identifies the most similar queries, calculates a similarity score based on the extracted keywords using Elasticsearch, and identifies the query with the highest score.

[0834] Input: Extracted keywords

[0835] Output: Most similar previous queries

[0836] Step 5:

[0837] The server uses generative AI to generate an automated response message based on the most similar past queries, generating prompt sentences and feeding them into a GPT-2 model to generate an appropriate response message.

[0838] Input: Most similar previous query, prompt (e.g., "Generate a sentence containing the following keywords: product, not found, help. Sentiment: negative.")

[0839] Output: Generated automated response message (e.g., "We apologize for the inconvenience. If you cannot find the product, please contact a store staff member or check this link.")

[0840] Step 6:

[0841] The server uses a sentiment analysis engine to recognize the user's emotions and adjusts the response message. It calculates an emotion score using the SentimentIntensityAnalyzer and changes the tone and content of the response message as needed.

[0842] Input: Generated auto-response message, user inquiry

[0843] Output: Reconciled response message

[0844] Step 7:

[0845] The server then sends the generated response message back to the user's terminal, using a communication protocol to transfer the message so that the user can receive a response quickly.

[0846] Input: Tailored response message

[0847] Output: The response message displayed on the user's terminal.

[0848] Step 8:

[0849] The server reparses the query and redirects the query to the correct department if the appropriate department is not set. Based on the results of the reparse, the server redirects the query.

[0850] Input: Tailored response message, inquiry content

[0851] Output: Enquiry data redirected to the appropriate department

[0852] 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.

[0853] 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.

[0854] 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.

[0855] [Third embodiment]

[0856] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0857] 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.

[0858] 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).

[0859] 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.

[0860] 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.

[0861] 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).

[0862] 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.

[0863] 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.

[0864] 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.

[0865] 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.

[0866] 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.

[0867] 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."

[0868] This invention is a system that automates responses to inquiries from users, aiming to improve the efficiency and productivity of inquiry responses. This system receives inquiry content, formats it into a unified format, analyzes it to extract key keywords, searches for similar past inquiries, and uses generative AI to automatically generate a response message that can be sent to the user. It can also reanalyze the inquiry content to identify the appropriate department and redirect the inquiry to the identified department.

[0869] Program processing overview

[0870] 1. Receiving an Inquiry

[0871] Users access a dedicated inquiry portal and enter their questions in an inquiry form.

[0872] When the user presses the send button, the inquiry is sent to the server.

[0873] The server stores the received inquiry in a database.

[0874] 2. Formatting into a unified format

[0875] The server formats the raw query content it receives into a predefined format, which makes subsequent processing easier.

[0876] 3. Content Analysis

[0877] The server analyzes the formatted query using natural language processing technology and extracts key keywords.

[0878] 4. Search for similar queries from the past

[0879] The server searches a database for similar past queries and identifies the most similar queries.

[0880] The server scores the search results and selects those with the highest similarity.

[0881] 5. Generate Auto-Response Messages

[0882] The server uses a generation AI to generate an automatic response message based on the extracted keywords and selected past inquiries.

[0883] 6. Transmission to the User

[0884] The server sends the generated automatic response message to the user's terminal.

[0885] The user checks the sent response message on the terminal.

[0886] 7. Redirect to the appropriate department

[0887] The server re-analyzes the query content and identifies an appropriate department if one has not been set.

[0888] The server forwards the query to the specified department if redirection is necessary.

[0889] A terminal (in the appropriate department) receives the inquiry and provides the information the user requests.

[0890] Specific examples

[0891] Example 1: Inquiry about network trouble

[0892] A user enters a query into the portal, such as "What should I do if my network is unstable?" and presses the send button. The server receives this query and stores it in a database. The server then formats the query based on keywords such as "network," "unstable," and "how to deal with it," and analyzes it using natural language processing technology. It then searches the database for similar past queries and identifies the most similar case. Generative AI is used to generate an automated response message (e.g., "If your network is unstable, please follow the steps below...") and sends it to the user. When the user receives the response message, it becomes easier for them to solve the problem themselves.

[0893] Example 2: Redirection due to misidentification of inquiry content

[0894] If a user sends a query saying "My PC won't start," but the server initially analyzes it and tries to treat it as a network problem, the server will reanalyze it and determine that the query is actually a hardware problem. Therefore, the query will be automatically redirected to the IT support department. The terminal (the IT support department's terminal) will receive this query and provide an appropriate solution.

[0895] In this way, the present invention can dramatically improve the efficiency of responding to inquiries and reduce the man-hours of both users and responders.

[0896] The processing flow will be explained below.

[0897] Step 1:

[0898] The user accesses a dedicated inquiry portal and enters the question into the inquiry form. The user also fills in the necessary information in the input fields and presses the send button to send the inquiry.

[0899] Step 2:

[0900] The server receives the inquiry sent by the user, and the received data is stored in a database for record keeping.

[0901] Step 3:

[0902] The server formats the received queries into a unified format, converting raw data into a defined format for easier subsequent processing.

[0903] Step 4:

[0904] The server analyzes the query using natural language processing technology, specifically extracting key keywords and phrases and obtaining key information needed to understand the query.

[0905] Step 5:

[0906] The server searches the database for similar past queries, identifies the most similar past queries based on the extracted keywords, and calculates a relevance score.

[0907] Step 6:

[0908] The server selects the query with the highest similarity from the search results and prepares to generate a response message based on the selected past query data.

[0909] Step 7:

[0910] The server uses a generation AI to generate an automated response message, creating a specific and appropriate response message based on selected past inquiry data and extracted keywords.

[0911] Step 8:

[0912] The server then sends the generated automated response message to the user's terminal, allowing the user to receive an immediate response.

[0913] Step 9:

[0914] The user checks the automatic response message on the terminal, which allows the user to quickly find a way to resolve the problem.

[0915] Step 10:

[0916] The server re-parses the query and identifies the appropriate department if one is not already set, so that the query can be re-evaluated and handled by the correct department.

[0917] Step 11:

[0918] The server then forwards the query to the appropriate department, and the query is automatically redirected to the correct department.

[0919] Step 12:

[0920] The terminal (the terminal in the appropriate department) receives the inquiry and provides the information the user is looking for. The department staff responds, allowing the user to receive professional support.

[0921] Through these steps, inquiries from users are handled efficiently and quickly. By automating the entire inquiry response process, the burden on staff is reduced and users can have their problems resolved quickly.

[0922] Example 1

[0923] 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."

[0924] Conventional inquiry response systems analyzed the content of inquiries individually and responded manually, resulting in issues with efficiency and productivity. Furthermore, they did not redirect inquiries to the appropriate department or generate automated response messages, increasing the likelihood of incorrect responses. Furthermore, responding to inquiries required time and effort, which led to lower user satisfaction. This resulted in increased response workloads across the company and a decline in overall performance.

[0925] 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.

[0926] In this invention, the server includes: means for receiving inquiry content; means for formatting the received inquiry content into a unified format; means for analyzing the inquiry content and extracting key keywords; means for searching for similar past inquiries and identifying the most similar inquiry; means for generating an automatic response message using a generative AI model; means for sending the generated response message to the user; means for providing a redirect function to an appropriate department; means for reanalyzing the inquiry content and identifying an appropriate department; means for forwarding the inquiry to the identified appropriate department; means for providing a user interface for accessing the inquiry portal; means for scoring answer candidates and selecting the most appropriate answer; means for analyzing the inquiry content using natural language processing technology; and means for generating a prompt sentence based on the analysis result and inputting it into the generative AI model. This automates the inquiry content analysis and response process, enabling prompt and accurate responses. Furthermore, adding a redirect function to the appropriate department improves user satisfaction and reduces response time.

[0927] "Inquiry content" is text data of questions, requests, and inquiries sent by users through the system.

[0928] The "receiving means" refers to the technical means for inputting the contents of the inquiry sent by the user into the server.

[0929] The "means for formatting into a unified format" refers to a technical means for converting the received query content into a format that is easy to process for subsequent analysis.

[0930] The "analysis means" is a natural language processing technique for extracting key keywords from the content of the inquiry.

[0931] A "keyword" is a word or phrase that contains important information in the query.

[0932] The "means for searching for similar past queries" refers to a technical means for finding similar past queries from a database.

[0933] A "generative AI model" is a system that uses artificial intelligence technology to automatically generate new response messages.

[0934] A "means for generating an auto-response message" is a technical means for automatically creating a response to a user's inquiry using a generative AI model.

[0935] "Means for sending" refers to the technical means for delivering the generated auto-reply message to the user.

[0936] The "redirect function to the appropriate department" is a function that transfers an inquiry to the appropriate department based on the content of the inquiry.

[0937] "Reanalysis measures" are technical measures that perform additional analysis when the results of the initial analysis are unclear.

[0938] "Means of identification" refers to technical means for determining the appropriate department to respond based on reanalysis.

[0939] "Forwarding means" refers to the technical means by which the inquiry is redirected to a specified department.

[0940] "User Interface" means the means by which a user accesses the inquiry portal.

[0941] "Scoring means" refers to a technical means for evaluating detected answer candidates and selecting the most appropriate answer.

[0942] "Natural language processing technology" is an artificial intelligence technology for understanding, analyzing, and processing human language.

[0943] A "prompt sentence" is an instruction sentence that is input into a generative AI model to generate a response.

[0944] The present invention is a system that automates responses to user inquiries, thereby improving the efficiency and productivity of inquiry responses. This system functions through communication between a server, terminals, and users.

[0945] Hardware and software used

[0946] The server is a high-performance computer system, and uses a database management system (DBMS) such as MySQL or PostgreSQL. Natural language processing engines such as TensorFlow and Watson are used for server-side natural language processing. Furthermore, OpenAI's GPT-3 and other generative AI models are used.

[0947] A terminal is a device where a user inputs a query and receives a response message. This can be a PC, tablet, smartphone, or other device. The user interface is implemented as a web application or a mobile application.

[0948] Processing Overview

[0949] 1. Receiving an Inquiry

[0950] Users access a dedicated inquiry portal and enter their questions in the inquiry form. For example, they might enter, "Please tell me what to do if the network is unstable."

[0951] When the user presses the send button, the query content is sent to the server as an HTTP request.

[0952] The server stores the received inquiry in a database.

[0953] 2. Formatting into a unified format

[0954] The server formats the received query content into a specific format (e.g., JSON format), which makes subsequent processing easier.

[0955] 3. Content Analysis

[0956] The server analyzes the formatted query using a natural language processing engine (e.g., TensorFlow) and extracts key keywords, such as "network," "unstable," and "how to respond."

[0957] 4. Search for similar queries from the past

[0958] The server searches the database to identify similar past queries, and the search results are scored to select the most similar queries.

[0959] 5. Generate Auto-Response Messages

[0960] The server uses a generative AI model (e.g., GPT-3) to generate an automated response message based on the extracted keywords and selected past inquiries. The prompt text is "Please generate an automated response message in response to the user's inquiry, 'Please tell me what to do if the network is unstable.'"

[0961] 6. Transmission to the User

[0962] The server then sends the generated auto-reply message to the user's device via email or a notification API (e.g., Firebase Cloud Messaging).

[0963] The user checks the response message sent and uses it to solve the problem.

[0964] 7. Redirect to the appropriate department

[0965] The server re-analyzes the query and identifies the appropriate department.

[0966] The server redirects the inquiry to the specified department, for example, contacting the IT support department via email or an internal notification system.

[0967] The terminal (the terminal of the appropriate department) receives this inquiry and initiates specific actions.

[0968] Specific examples

[0969] Example 1: Inquiry about network trouble

[0970] When a user enters "What to do if the network is unstable" into the inquiry portal and submits it, the server receives the inquiry and stores it in a database. Next, the inquiry is formatted based on keywords such as "network," "unstable," and "how to deal with it," and analyzed using natural language processing technology. After that, similar past inquiries are searched for in the database to identify the most similar case. Using generative AI, an automatic response message is generated that reads, "If the network is unstable, please follow the steps below..." and sent to the user. When the user receives the response message, it becomes easier for them to solve the problem themselves.

[0971] Example 2: Redirection due to misidentification of inquiry content

[0972] If a user sends a query saying "My PC won't start," but the server initially analyzes it and treats it as a network problem, the server reanalyzes it and determines that the query is actually a hardware problem. Therefore, the query is automatically redirected to the IT support department. The terminal (the IT support department's terminal) receives this query and can provide an appropriate solution.

[0973] In this way, the present invention can dramatically improve the efficiency of responding to inquiries and reduce the man-hours of both users and responders.

[0974] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0975] Step 1:

[0976] The user accesses the inquiry portal, enters the question in the inquiry form, and presses the send button.

[0977] Input: The inquiry entered by the user (e.g., "Please tell me what to do if the network is unstable").

[0978] Output: The query stored in the database.

[0979] Specific operation: The server stores the query content received as an HTTP request in a database. In this case, a database management system such as MySQL or PostgreSQL is used.

[0980] Step 2:

[0981] The server formats the received query content into a specified format (e.g., JSON format).

[0982] Input: The raw query stored in the database.

[0983] Output: Pretty-printed JSON representation of the query.

[0984] What happens: The server converts the query to JSON format, removing line breaks and special characters. This formatting makes it easier to parse later.

[0985] Step 3:

[0986] The server analyzes the formatted query using a natural language processing (NLP) engine and extracts key keywords.

[0987] Input: Query content in JSON format.

[0988] Output: Extracted key keywords (e.g., "network," "unstable," "how to respond").

[0989] Specific operation: The server uses a natural language processing engine such as TensorFlow or Watson to analyze the query content and identify important keywords.

[0990] Step 4:

[0991] The server searches a database to identify similar past queries.

[0992] Input: Extracted main keywords.

[0993] Output: The most similar previous queries.

[0994] Specific operation: The server generates an SQL query based on the extracted keywords and searches the database. The search results are scored using a search engine such as Elasticsearch, and the most similar results are selected.

[0995] Step 5:

[0996] The server generates the automated response message using the generative AI model.

[0997] Input: The most similar previous query, extracted keywords, and prompt (e.g., "Please generate an auto-response message in response to the user's query 'What should I do if my network is unstable?'").

[0998] Output: Auto-reply message (e.g. "If your network is unstable, please try the following steps...").

[0999] Specific operation: The server inputs a prompt sentence into a generative AI model such as OpenAI's GPT-3 to generate an automated response message.

[1000] Step 6:

[1001] The server sends the generated automatic response message to the user's terminal.

[1002] Input: Auto-reply message.

[1003] Output: The response message that is displayed on the user's terminal.

[1004] Specific operation: The server sends the generated message to the user using a notification API such as email or Firebase Cloud Messaging. The user then checks the message on their device.

[1005] Step 7:

[1006] The server re-analyzes the query and identifies the appropriate department.

[1007] Input: Query content, reanalysis instructions.

[1008] Output: The appropriate department identified.

[1009] Specific operation: The server uses the natural language processing engine again to re-analyze the inquiry content and identify the appropriate department.

[1010] Step 8:

[1011] The server redirects the query to the specified department.

[1012] Input: Identified appropriate department, inquiry details.

[1013] Output: Query redirected to the appropriate department's terminal.

[1014] Specific operation: The server forwards the inquiry to the specified department using an internal notification system or email. The appropriate department's terminal receives the inquiry and begins to respond.

[1015] (Application example 1)

[1016] 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."

[1017] Current inquiry response systems have difficulty responding quickly and accurately to a wide variety of inquiries from users. They also face the problem of increasing the workload of responders, reducing the efficiency of customer service. Furthermore, it is difficult to provide consistent responses to similar inquiries, making it difficult to improve customer satisfaction.

[1018] 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.

[1019] In this invention, the server includes means for receiving the inquiry content, means for formatting the received inquiry content into a unified format, means for analyzing the inquiry content and extracting key keywords, means for searching past similar inquiries and identifying the most similar inquiry, means for generating a prompt sentence using a generative AI model and generating an automatic response message, means for transmitting the generated response message to an information processing device, and means for reanalyzing the inquiry content, identifying an appropriate department if an appropriate department has not been set, and transferring the inquiry to the appropriate department. This dramatically improves the efficiency of inquiry response and reduces the man-hours of both users and responders.

[1020] The "means for receiving the inquiry content" is a function for transmitting the inquiry content submitted by the user to the server via the network, and for the server to receive the content.

[1021] The "means for converting received inquiry content into a unified format" is a function for converting inquiry content sent in various formats into a consistent unified format.

[1022] "Means for analyzing the content of inquiries and extracting key keywords" refers to a function that utilizes natural language processing technology to identify and extract key keywords from the content of inquiries.

[1023] The "means for searching past similar queries and identifying the most similar query" is a function for comparing a query received from a database in the past with the most similar query and identifying the most similar query.

[1024] "Means for generating prompt sentences using a generative AI model and generating automatic response messages" refers to a function that uses a pre-trained generative AI model to generate prompt sentences required for a response and creates automatic response messages based on those sentences.

[1025] The "means for transmitting the generated response message to the information processing device" is a function for transferring the generated response message to the user's information processing device (for example, a smartphone or a computer) and displaying it.

[1026] "Means for reanalyzing the content of the inquiry, identifying an appropriate department if an appropriate department has not been set, and transferring the inquiry to the relevant department" is a function that reanalyzes the content of the inquiry, and if an appropriate department has not been set, identifies the most appropriate department and transfers the inquiry to that department.

[1027] MODE FOR CARRYING OUT THE INVENTION

[1028] System Configuration

[1029] The system implemented in this invention is broadly composed of the following main means.

[1030] 1. Means of receiving inquiries

[1031] 2. A means of formatting received inquiries into a unified format

[1032] 3. A method for analyzing inquiries and extracting key keywords

[1033] 4. A means to search for similar queries from the past and identify the most similar queries

[1034] 5. A means of generating prompts and auto-response messages using generative AI models

[1035] 6. Means for transmitting the generated response message to the information processing device

[1036] 7. A means to re-analyze the content of the inquiry, identify the appropriate department if one is not set, and transfer the inquiry to the appropriate department.

[1037] Program processing

[1038] Hardware and software usage

[1039] server:

[1040] Hardware: High-performance server

[1041] Software: Database management systems (e.g., MySQL), search engines (e.g., Elasticsearch), natural language processing libraries (e.g., Spacy, BERT), generative AI models (e.g., ChatGPT, GPT-3)

[1042] User device:

[1043] Hardware: Smartphones, tablets, computers

[1044] Software: Mobile application, browser-based interface

[1045] Data processing and calculation flow

[1046] 1. Receiving an Inquiry

[1047] Users access the inquiry form using a smartphone or PC, enter their question, and submit it. The entered inquiry is sent to a server via a network and stored in a database.

[1048] 2. Formatting into a unified format

[1049] The server converts the received query content into a unified format, using a text formatting library (e.g., Python's NLTK) for this conversion process.

[1050] 3. Content Analysis

[1051] The server analyzes the query using natural language processing techniques (e.g., Spacy, BERT) to extract key keywords. This analysis identifies the core of the query.

[1052] 4. Search for similar queries from the past

[1053] The server compares the queries against previous queries in a database to find the most similar queries, using a search engine such as Elasticsearch to score the similarity and identify highly similar queries.

[1054] 5. Generate Auto-Response Messages

[1055] The server uses a generative AI model (e.g., ChatGPT, GPT-3) to generate an automated response message based on the prompt, which includes extracted keywords and information about similar past inquiries.

[1056] 6. Transmission to the User

[1057] The server generates a response message and sends it to the user's device. A real-time notification service (e.g., Firebase) is used to deliver the message to the user quickly and reliably.

[1058] 7. Redirect to the appropriate department

[1059] The server re-analyzes the inquiry and, if an appropriate department is not set, identifies the appropriate department and transfers the inquiry. This allows inquiries that require specialized assistance to be handled appropriately.

[1060] Specific examples

[1061] Example 1: Delivery enquiry

[1062] A user submits an inquiry saying, "The item I ordered hasn't arrived." This inquiry is received by the server, and key keywords such as "order," "item," and "not arrived" are extracted. Similar past inquiries are searched for, and the generative AI model generates a response message saying, "We will confirm your order number and let you know the delivery status. Delivery usually takes 5-7 days." This message is then promptly sent to the user's device.

[1063] Prompt Sentence Examples

[1064] "Please generate an auto-reply message for the following inquiry: 'My order hasn't arrived'"

[1065] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1066] Step 1:

[1067] The user accesses the smart mail-order support application, enters their question into the inquiry form, and presses the send button. At this time, the user may enter something like, "The item I ordered hasn't arrived." The entered data is saved in JSON format within the application and sent to the server via the network. The server receives the inquiry and simultaneously saves the information in a database. The input is the user's question, and the output is saved in the database.

[1068] Step 2:

[1069] The server formats the received raw query content into a specified unified format. This process uses a text formatting library (e.g., Python's NLTK). Specifically, it parses JSON-formatted data, removes unnecessary whitespace and special characters, and formats the query content. The input is the received raw query content, and the output is the data formatted in a unified format.

[1070] Step 3:

[1071] The server analyzes the formatted query using natural language processing technology (e.g., Spacy, BERT) and extracts key keywords. This results in keywords such as "order," "product," and "not delivered." The input is data formatted in a unified format, and the output is the key keywords.

[1072] Step 4:

[1073] The server uses a search engine (e.g., Elasticsearch) to search a database for similar past queries and identify the most similar queries. It scores the search results and selects queries with high similarity. The input is the main keywords, and the output is the past query data with high similarity.

[1074] Step 5:

[1075] The server uses a generative AI model (e.g., ChatGPT, GPT-3) to generate a prompt and generate an auto-response message. This prompt includes extracted keywords and information on the most similar past inquiries. For example, it could be in the format "Please generate an auto-response message in response to the following inquiry: 'The item I ordered has not arrived.'" The input is similar inquiry data, and the output is the generated response message.

[1076] Step 6:

[1077] The server sends the generated automatic response message to the user's device using a real-time notification service (e.g., Firebase). The user can check this response message within the application. The input is the generated response message, and the output is the notification and display on the user's device.

[1078] Step 7:

[1079] The server re-analyzes the query and identifies the appropriate department if one is not already set. If redirection is required, the server forwards the query to the appropriate department. The input is the user's query, and the output is a redirect to the identified appropriate department.

[1080] 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.

[1081] This invention is a system that automates responses to inquiries from users, aiming to improve the efficiency and productivity of inquiry responses. This system receives inquiry content, formats it into a unified format, analyzes it to extract key keywords, searches for similar past inquiries, and uses generative AI to generate an automatic response message that is sent to the user. It can also reanalyze the inquiry content to identify the appropriate department and redirect the inquiry to the identified department. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it also has the function of improving the quality of inquiry responses.

[1082] Program processing overview

[1083] 1. Receiving an Inquiry

[1084] Users access a dedicated inquiry portal, enter their questions into the inquiry form, and then press the send button to send the inquiry to the server.

[1085] The server checks the received inquiry and stores it in a database.

[1086] 2. Formatting into a unified format

[1087] The server converts the received raw query content into a unified format that can be easily parsed and processed.

[1088] 3. Content Analysis

[1089] The server uses NLP technology to analyze the query content, which has been formatted into a unified format, and extracts key keywords and phrases.

[1090] 4. Search for similar queries from the past

[1091] The server searches a database for similar queries from the past to identify the most similar queries, and the search results are evaluated based on a relevance score.

[1092] 5. Generate Auto-Response Messages

[1093] The server uses a generation AI to create an automated response message based on the extracted keywords and identified similar queries.

[1094] 6. Emotional Recognition and Regulation

[1095] The server uses an emotion engine to recognize emotions from the user's query content and expressions, and adjusts the tone and content of the generated response message according to the user's emotions.

[1096] For example, if the user is expressing anger or frustration, the response message will have a more polite and empathetic tone.

[1097] 7. Transmission to the User

[1098] The server sends the generated automatic response message to the user's terminal, so that the user can receive the response quickly.

[1099] 8. Redirect to the appropriate department

[1100] The server re-analyzes the query and, if the appropriate department is not set, forwards the query to the correct department.

[1101] The terminal (the terminal of the appropriate department) responds to the received inquiry and provides the user with an appropriate answer.

[1102] Specific examples

[1103] Example 1: Inquiry about network trouble

[1104] The user enters "Please tell me what to do if the network is unstable" into an inquiry form and submits it. The server receives the inquiry and saves it in a database. The server then formats the inquiry into a unified format and uses NLP technology to extract keywords such as "network," "unstable," and "how to deal with it." The server then searches the database for similar past inquiries and identifies the most similar one. It then uses generative AI to generate an automated response message, which the server then sends to the user. The server then uses an emotion engine to recognize the user's emotions. For example, if the user expresses frustration, the response message is adjusted to show empathy, such as "We apologize for the inconvenience. If the network is unstable, please follow the steps below..."

[1105] Example 2: Redirection due to misidentification of inquiry content

[1106] A user enters "My PC won't start" into an inquiry form and submits it. The server receives the inquiry and saves it in a database. The server analyzes the inquiry and extracts keywords such as "PC" and "won't start." An initial analysis identifies it as a network problem, but reanalysis identifies it as a hardware problem. In this case, the server automatically redirects the inquiry to the IT support department. A terminal (a terminal in the IT support department) receives the inquiry and provides an appropriate solution.

[1107] In this way, the present invention dramatically improves the efficiency of inquiries and reduces the man-hours required by both users and responders. The introduction of an emotion engine can provide an even higher level of user experience.

[1108] The processing flow will be explained below.

[1109] Step 1:

[1110] Users access a dedicated inquiry portal, enter their questions in the inquiry form, and then press the send button to send the inquiry.

[1111] Step 2:

[1112] The server receives the inquiry sent by the user, and the received inquiry is immediately saved in the database.

[1113] Step 3:

[1114] The server formats the raw query content it receives into a uniform format, ensuring data consistency and facilitating subsequent processing.

[1115] Step 4:

[1116] The server uses natural language processing technology to analyze the query, which has been formatted in a unified format, and extracts key keywords and phrases.

[1117] Step 5:

[1118] The server searches the database for similar queries from the past, and identifies the most similar query based on the extracted keywords.

[1119] Step 6:

[1120] The server evaluates the results of similar queries based on relevance scores and selects the most appropriate answers.

[1121] Step 7:

[1122] The server uses AI to generate an automated response message based on the selected answer, utilizing extracted keywords and past inquiry data.

[1123] Step 8:

[1124] The server uses an emotion engine to recognize the user's emotions and determines their emotions from the content of the user's query and the way they input it.

[1125] Step 9:

[1126] The server adjusts the tone and content of the automated response message depending on the emotion recognized by the emotion engine. For example, if the user expresses anger or frustration, the response message will be more polite and empathetic.

[1127] Step 10:

[1128] The server then sends the generated automated response message to the user's terminal, allowing the user to quickly obtain a solution.

[1129] Step 11:

[1130] The user checks the automatic response message on the terminal, which allows the user to immediately obtain a solution to the problem.

[1131] Step 12:

[1132] The server re-analyzes the inquiry content and identifies an appropriate department if one has not been set.

[1133] Step 13:

[1134] The server forwards the query to the appropriate identified department, and the query is automatically redirected to ensure it reaches the correct department.

[1135] Step 14:

[1136] The terminal (in the appropriate department) receives the inquiry and provides the information the user is looking for, allowing for a prompt and professional response.

[1137] By following the above steps, the system of the present invention can efficiently process user inquiries and provide appropriate responses. The introduction of the emotion engine will provide a higher quality user experience.

[1138] Example 2

[1139] 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."

[1140] In conventional inquiry response systems, user inquiries are often handled manually, which results in long response times and reduced user satisfaction. In addition, inquiries are sometimes not accurately routed to the appropriate department based on their content, reducing response efficiency. Furthermore, a uniform response is returned without considering the user's feelings, which also results in a poor user experience.

[1141] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1142] In this invention, the server includes means for receiving inquiry content, means for formatting the received inquiry content into a unified format, means for analyzing the inquiry content and extracting key keywords, means for searching past similar inquiries and identifying the most similar inquiry, means for generating an automatic response message using a generation AI, means for recognizing a user's emotion, means for adjusting the content of the automatic response message according to the recognized emotion, means for sending the generated response message to the user, means for reanalyzing the inquiry content and identifying an appropriate department, means for transferring the inquiry to the identified appropriate department, and means for providing a user interface for accessing the inquiry portal. This dramatically improves the efficiency of inquiry response, improves user satisfaction, enables accurate department transfers, and enables flexible responses according to the user's emotion.

[1143] "Inquiry content" refers to information that a user inputs and transmits to the system through the inquiry portal.

[1144] "Receiving" refers to the process in which the server acquires the inquiry content sent by the user.

[1145] A "uniform format" is a standardized data format that is formatted to facilitate analysis and processing.

[1146] "Analysis" is the process of extracting and understanding meaning and keywords from the content of a query.

[1147] "Keywords" refer to the main concepts or phrases in the query content.

[1148] A "similar query" refers to a query sent by a user in the past that is similar to the current query.

[1149] "Generative AI" refers to a system that uses artificial intelligence to generate automated response messages.

[1150] An "automatic response message" is a response created by a generation AI and automatically sent to the user.

[1151] "Emotion recognition" is a technology that recognizes the emotional state of a user from the content of their inquiry.

[1152] "Adjustment" is the process of changing the tone and content of a response message based on the results of emotion recognition.

[1153] "Sending" is the process by which the server delivers the generated automated response message to the user.

[1154] "Reanalysis" is the process of reviewing the results of an initial analysis to improve accuracy.

[1155] The "appropriate department" refers to the department that can respond to the inquiry most quickly and accurately based on the content of the inquiry.

[1156] "Forwarding" is the process by which the server routes the query to the appropriate department.

[1157] "User Interface" means the interface through which a user accesses the system and inputs and submits inquiries.

[1158] The present invention is a system for automating responses to user inquiries, and aims to improve the efficiency and productivity of inquiries. This system includes the following processes.

[1159] Receiving inquiries

[1160] The user accesses a dedicated inquiry portal, enters their question in the inquiry form, and submits it. For example, they might enter, "I forgot my password. Please tell me how to reset it." The server retrieves the received inquiry and stores it in a MySQL database or similar.

[1161] Formatting to a unified format

[1162] The server formats the received query content into a unified format, for example, by removing unnecessary line breaks and spaces from the received text and converting it to JSON format.

[1163] Content Analysis

[1164] The server uses a natural language processing (NLP) library (e.g., spaCy or NLTK) to parse the query, specifically by performing morphological analysis to extract key keywords and phrases such as "password," "forgot," and "reset."

[1165] Search for similar past inquiries

[1166] The server generates a search query and searches a database for similar queries, calculating a similarity score using techniques such as TF-IDF and Word2Vec, identifying the most relevant query, and retrieving its content and answer.

[1167] Generate auto-response messages

[1168] The server creates a generation prompt for the generative AI model (for example, OpenAI's GPT-3). The prompt text includes the extracted keywords and the content of similar queries. For example, a prompt text such as "A user has asked how to reset their forgotten password. Please follow the steps below to reset it..." is input to the generative AI. The generated automated response message is then obtained.

[1169] Emotion Recognition and Regulation

[1170] The server uses an emotion engine (for example, Microsoft Azure's Text Analytics API) to recognize emotions from the user's inquiry. For example, if the user expresses dissatisfaction or anger, the server adjusts the tone and content of the generated response message. For example, it might make the message more polite, such as, "We apologize for the inconvenience. Please follow the steps below to reset your password."

[1171] Send to user

[1172] The server sends the generated automatic response message to the user's terminal, so that the user can receive the response quickly.

[1173] Redirecting to the appropriate department

[1174] The server re-analyzes the inquiry, and if the appropriate department is not set, it automatically redirects the inquiry to the appropriate department. For example, if an inquiry that was initially misidentified as a network problem is re-analyzed to be a hardware problem, the server redirects the inquiry to the IT support department. The terminal (for example, a terminal in the IT support department) responds appropriately to the received inquiry and provides the user with a solution.

[1175] The above is an embodiment of the present invention, and this system dramatically improves the efficiency of inquiries and reduces the man-hours of both users and responders. The introduction of an emotion engine can provide an even higher level of user experience.

[1176] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1177] Step 1: Receiving an inquiry

[1178] A user accesses a dedicated inquiry portal, enters a question in the inquiry form, and submits it. For example, the user might enter, "I forgot my password, please tell me how to reset it." The server receives the HTTP request sent from the portal and extracts its contents. The input is the user's inquiry text, and the output is the received inquiry text.

[1179] Step 2: Formatting into a unified format

[1180] The server formats the received query content into a unified format. Specifically, it removes unnecessary line breaks and spaces from the received text and converts it into JSON format. The input is the received query text, and the output is the formatted JSON data.

[1181] Step 3: Content Analysis

[1182] The server uses a natural language processing (NLP) library (e.g., spaCy or NLTK) to parse the query formatted in a unified format. Specifically, it performs morphological analysis to extract key keywords and phrases such as "password," "forgot," and "reset." The input is formatted JSON data, and the output is a list of extracted keywords.

[1183] Step 4: Search for similar past queries

[1184] The server generates a search query and searches a database for similar queries, calculating a similarity score using methods such as TF-IDF or Word2Vec. The input is a list of extracted keywords, and the output is related queries and their answers.

[1185] Step 5: Generate an auto-response message

[1186] The server creates a generation prompt for the generative AI model (for example, OpenAI's GPT-3). The prompt text includes the extracted keywords and the content of similar queries. For example, a prompt text such as "A user has asked how to reset their forgotten password. Please follow the steps below to reset it..." is input to the generative AI. The generative AI generates a response message, and the server receives that message. The input is the prompt text, and the output is the generated automated response message.

[1187] Step 6: Emotional awareness and regulation

[1188] The server uses an emotion engine (for example, Microsoft Azure's Text Analytics API) to recognize emotions from the user's inquiry. Specifically, it performs emotion analysis to identify the emotion expressed by the user (for example, anger, frustration, or joy). If the user expresses frustration or anger, it adjusts the tone and content of the generated response message. The input is the generated automated response message and the user's emotion data, and the output is the adjusted response message.

[1189] Step 7: Send to users

[1190] The server sends the generated automatic response message to the user's terminal. Specifically, it returns the generated message to the user's terminal as an HTTP response. The input is the adjusted response message, and the output is the response message sent to the user.

[1191] Step 8: Redirect to the appropriate department

[1192] The server re-analyzes the inquiry content, and if the appropriate department is not set, it automatically identifies the department and redirects the user. Specifically, it performs NLP analysis again to extract keywords to identify the appropriate department. It then sends an email or notification to that department. The input is the inquiry content and the re-analysis results, and the output is the redirected notification. The terminal (for example, a terminal in the IT support department) responds appropriately to the received inquiry and provides the user with a solution.

[1193] (Application example 2)

[1194] 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."

[1195] In modern brick-and-mortar stores, it is important to respond to customer inquiries quickly and appropriately, but this requires a lot of manpower and time, which is a challenge. Furthermore, it is also necessary to improve customer satisfaction by appropriately recognizing customer emotions and providing appropriate responses. However, a system for doing this efficiently has not yet been fully developed.

[1196] 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.

[1197] In this invention, the server includes means for receiving the inquiry content, means for formatting the received inquiry content into a unified format, means for analyzing the inquiry content and extracting key keywords, means for searching for similar past inquiries and identifying the most similar inquiry, means for generating an automatic response message using a generative AI, means for recognizing the user's emotions using a sentiment analysis engine and adjusting the response message, means for sending the generated response message to the user, means for reanalyzing the inquiry content and identifying an appropriate department, means for forwarding the inquiry to the identified appropriate department, means for providing a user interface for accessing the inquiry portal, means for quickly and accurately interacting with users in physical stores using terminals, means for using a generative AI model to generate a response message based on the customer's emotions, and means for generating the automatic response message using prompt sentences, thereby enabling efficiency and productivity improvement in inquiry response.

[1198] The "means for receiving the inquiry content" is an interface for the system to take in the inquiry content input by the user.

[1199] The "means for converting the received query content into a unified format" is a process for converting the received query content into a format that is easy to parse and process.

[1200] "Means for analyzing the content of the query and extracting key keywords" refers to a means for extracting important information from the query using natural language processing technology.

[1201] The "means for searching past similar queries and identifying the most similar query" is a technique for finding an example that is closest to the current query from past queries in a database.

[1202] "Means for generating automated response messages using generative AI" refers to a process that uses AI technology to generate appropriate response messages in response to inquiries.

[1203] "Means for recognizing a user's emotions using an emotion analysis engine and adjusting a response message" is a technology for detecting emotions from a user's input text and adjusting the tone and content of a response message.

[1204] The "means for transmitting the generated response message to the user" refers to a communication means for transmitting the generated response message to the user's terminal.

[1205] The "means for re-analyzing the inquiry content and identifying the appropriate department" is a process for re-analyzing the inquiry content and automatically identifying the appropriate department in charge.

[1206] The "means for transferring the inquiry to the specified appropriate department" is a technique for redirecting the inquiry content to the specified department.

[1207] "Means for providing a user interface for accessing the inquiry portal" refers to a method for providing an interface for a user to submit an inquiry to the system.

[1208] "Means for quickly and accurately communicating with users in a physical store using a terminal" refers to a method for efficiently communicating with customers via a terminal located in a physical store.

[1209] "Means for using a generative AI model to generate a response message based on customer sentiment" refers to an AI model for generating a response that takes into account customer sentiment information.

[1210] The "means for generating an automatic response message using a prompt sentence" is a method for generating an appropriate automatic response message based on a specified prompt sentence.

[1211] In order to carry out the present invention, it is important to build a system that combines a server and a terminal when dealing with customers in a brick-and-mortar store.

[1212] Server processing overview

[1213] The server efficiently processes customer inquiries using the following methods:

[1214] 1. Receiving and storing your inquiry:

[1215] The user enters the inquiry using a terminal installed in the physical store and sends it to the server. The server receives the inquiry and stores it in a database. This step requires a network connection and database software (e.g., MySQL).

[1216] 2. Formatting to Unified Format:

[1217] The server formats the received query content into a unified format for easy parsing. This is done using a script (e.g., a Python script) that performs text preprocessing.

[1218] 3. Extracting Key Keywords:

[1219] The server analyzes the query content using NLP techniques (e.g., NLTK library) and extracts key keywords, which are then used in the next process.

[1220] 4. Search for similar queries from the past:

[1221] The server searches the database for similar queries and identifies the most similar queries. This step utilizes a search algorithm (e.g., Elasticsearch).

[1222] 5. Generate auto-response messages:

[1223] The server generates an automated response message using a generative AI (e.g., GPT-2 model) based on the extracted keywords and identified similar queries, calling the model using an API (e.g., Hugging Face Transformers).

[1224] 6. Emotion recognition and regulation:

[1225] The server uses an emotion analysis engine (e.g., SentimentIntensityAnalyzer) to recognize the user's emotions and adjust the tone of the response message: if the user is expressing frustration, the response message will have a more polite and empathetic tone.

[1226] 7. Sending a response message:

[1227] The server then sends the generated automatic response message back to the user's terminal, allowing the user to receive a prompt reply.

[1228] 8. Redirect to the appropriate department:

[1229] The server re-parses the query and, if the appropriate department is not set, forwards the query to the correct department, along with the information to whom it should be assigned.

[1230] Use of concrete examples and prompts

[1231] For example, if a user types "I can't find the product in the store. Help me!" into a terminal in a physical store, the server receives the content, formats it into a unified format, and extracts keywords such as "product," "can't find," and "help." Sentiment analysis determines that the user's emotion is frustration. Based on this information, the following prompt sentence is provided to the generation AI:

[1232] Prompt: "Generate sentences containing the following keywords: product, not found, help. Sentiment: negative."

[1233] The response message generated using this prompt is "We apologize for the inconvenience. If you cannot find the product, please speak to a store staff member or check this link." The server then sends this message to the user.

[1234] As described above, the present invention is a system that can improve the efficiency and productivity of responding to inquiries and increase customer satisfaction in physical stores.

[1235] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1236] Step 1:

[1237] The user inputs the inquiry details using a terminal installed in the physical store, and then sends the input inquiry details to the server by pressing the send button.

[1238] Input: User's query (e.g. "I can't find the item in your store. Please help!")

[1239] Output: Received query data

[1240] Step 2:

[1241] The server formats the received query into a unified format, preprocessing the text to convert the query to lower case and removing special characters and unnecessary spaces.

[1242] Input: Received inquiry data

[1243] Output: Query data formatted in a unified format

[1244] Step 3:

[1245] The server uses NLP techniques to parse the query and extract key keywords, then uses the NLTK library to tokenize it and extract important nouns and verbs.

[1246] Input: Query data formatted in a unified format

[1247] Output: Extracted keywords (e.g. "product", "cannot find", "help")

[1248] Step 4:

[1249] The server searches the database for similar queries from the past, identifies the most similar queries, calculates a similarity score based on the extracted keywords using Elasticsearch, and identifies the query with the highest score.

[1250] Input: Extracted keywords

[1251] Output: Most similar previous queries

[1252] Step 5:

[1253] The server uses generative AI to generate an automated response message based on the most similar past queries, generating prompt sentences and feeding them into a GPT-2 model to generate an appropriate response message.

[1254] Input: Most similar previous query, prompt (e.g., "Generate a sentence containing the following keywords: product, not found, help. Sentiment: negative.")

[1255] Output: Generated automated response message (e.g., "We apologize for the inconvenience. If you cannot find the product, please contact a store staff member or check this link.")

[1256] Step 6:

[1257] The server uses a sentiment analysis engine to recognize the user's emotions and adjusts the response message. It calculates an emotion score using the SentimentIntensityAnalyzer and changes the tone and content of the response message as needed.

[1258] Input: Generated auto-response message, user inquiry

[1259] Output: Reconciled response message

[1260] Step 7:

[1261] The server then sends the generated response message back to the user's terminal, using a communication protocol to transfer the message so that the user can receive a response quickly.

[1262] Input: Tailored response message

[1263] Output: The response message displayed on the user's terminal.

[1264] Step 8:

[1265] The server reparses the query and redirects the query to the correct department if the appropriate department is not set. Based on the results of the reparse, the server redirects the query.

[1266] Input: Tailored response message, inquiry content

[1267] Output: Enquiry data redirected to the appropriate department

[1268] 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.

[1269] 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.

[1270] 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.

[1271] [Fourth embodiment]

[1272] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1273] 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.

[1274] 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).

[1275] 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.

[1276] 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.

[1277] 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).

[1278] 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.

[1279] 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.

[1280] 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.

[1281] 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.

[1282] 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.

[1283] 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.

[1284] 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."

[1285] This invention is a system that automates responses to inquiries from users, aiming to improve the efficiency and productivity of inquiry responses. This system receives inquiry content, formats it into a unified format, analyzes it to extract key keywords, searches for similar past inquiries, and uses generative AI to automatically generate a response message that can be sent to the user. It can also reanalyze the inquiry content to identify the appropriate department and redirect the inquiry to the identified department.

[1286] Program processing overview

[1287] 1. Receiving an Inquiry

[1288] Users access a dedicated inquiry portal and enter their questions in an inquiry form.

[1289] When the user presses the send button, the inquiry is sent to the server.

[1290] The server stores the received inquiry in a database.

[1291] 2. Formatting into a unified format

[1292] The server formats the raw query content it receives into a predefined format, which makes subsequent processing easier.

[1293] 3. Content Analysis

[1294] The server analyzes the formatted query using natural language processing technology and extracts key keywords.

[1295] 4. Search for similar queries from the past

[1296] The server searches a database for similar past queries and identifies the most similar queries.

[1297] The server scores the search results and selects those with the highest similarity.

[1298] 5. Generate Auto-Response Messages

[1299] The server uses a generation AI to generate an automatic response message based on the extracted keywords and selected past inquiries.

[1300] 6. Transmission to the User

[1301] The server sends the generated automatic response message to the user's terminal.

[1302] The user checks the sent response message on the terminal.

[1303] 7. Redirect to the appropriate department

[1304] The server re-analyzes the query content and identifies an appropriate department if one has not been set.

[1305] The server forwards the query to the specified department if redirection is necessary.

[1306] A terminal (in the appropriate department) receives the inquiry and provides the information the user requests.

[1307] Specific examples

[1308] Example 1: Inquiry about network trouble

[1309] A user enters a query into the portal, such as "What should I do if my network is unstable?" and presses the send button. The server receives this query and stores it in a database. The server then formats the query based on keywords such as "network," "unstable," and "how to deal with it," and analyzes it using natural language processing technology. It then searches the database for similar past queries and identifies the most similar case. Generative AI is used to generate an automated response message (e.g., "If your network is unstable, please follow the steps below...") and sends it to the user. When the user receives the response message, it becomes easier for them to solve the problem themselves.

[1310] Example 2: Redirection due to misidentification of inquiry content

[1311] If a user sends a query saying "My PC won't start," but the server initially analyzes it and tries to treat it as a network problem, the server will reanalyze it and determine that the query is actually a hardware problem. Therefore, the query will be automatically redirected to the IT support department. The terminal (the IT support department's terminal) will receive this query and provide an appropriate solution.

[1312] In this way, the present invention can dramatically improve the efficiency of responding to inquiries and reduce the man-hours of both users and responders.

[1313] The processing flow will be explained below.

[1314] Step 1:

[1315] The user accesses a dedicated inquiry portal and enters the question into the inquiry form. The user also fills in the necessary information in the input fields and presses the send button to send the inquiry.

[1316] Step 2:

[1317] The server receives the inquiry sent by the user, and the received data is stored in a database for record keeping.

[1318] Step 3:

[1319] The server formats the received queries into a unified format, converting raw data into a defined format for easier subsequent processing.

[1320] Step 4:

[1321] The server analyzes the query using natural language processing technology, specifically extracting key keywords and phrases and obtaining key information needed to understand the query.

[1322] Step 5:

[1323] The server searches the database for similar past queries, identifies the most similar past queries based on the extracted keywords, and calculates a relevance score.

[1324] Step 6:

[1325] The server selects the query with the highest similarity from the search results and prepares to generate a response message based on the selected past query data.

[1326] Step 7:

[1327] The server uses a generation AI to generate an automated response message, creating a specific and appropriate response message based on selected past inquiry data and extracted keywords.

[1328] Step 8:

[1329] The server then sends the generated automated response message to the user's terminal, allowing the user to receive an immediate response.

[1330] Step 9:

[1331] The user checks the automatic response message on the terminal, which allows the user to quickly find a way to resolve the problem.

[1332] Step 10:

[1333] The server re-parses the query and identifies the appropriate department if one is not already set, so that the query can be re-evaluated and handled by the correct department.

[1334] Step 11:

[1335] The server then forwards the query to the appropriate department, and the query is automatically redirected to the correct department.

[1336] Step 12:

[1337] The terminal (the terminal in the appropriate department) receives the inquiry and provides the information the user is looking for. The department staff responds, allowing the user to receive professional support.

[1338] Through these steps, inquiries from users are handled efficiently and quickly. By automating the entire inquiry response process, the burden on staff is reduced and users can have their problems resolved quickly.

[1339] Example 1

[1340] 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."

[1341] Conventional inquiry response systems analyzed the content of inquiries individually and responded manually, resulting in issues with efficiency and productivity. Furthermore, they did not redirect inquiries to the appropriate department or generate automated response messages, increasing the likelihood of incorrect responses. Furthermore, responding to inquiries required time and effort, which led to lower user satisfaction. This resulted in increased response workloads across the company and a decline in overall performance.

[1342] 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.

[1343] In this invention, the server includes: means for receiving inquiry content; means for formatting the received inquiry content into a unified format; means for analyzing the inquiry content and extracting key keywords; means for searching for similar past inquiries and identifying the most similar inquiry; means for generating an automatic response message using a generative AI model; means for sending the generated response message to the user; means for providing a redirect function to an appropriate department; means for reanalyzing the inquiry content and identifying an appropriate department; means for forwarding the inquiry to the identified appropriate department; means for providing a user interface for accessing the inquiry portal; means for scoring answer candidates and selecting the most appropriate answer; means for analyzing the inquiry content using natural language processing technology; and means for generating a prompt sentence based on the analysis result and inputting it into the generative AI model. This automates the inquiry content analysis and response process, enabling prompt and accurate responses. Furthermore, adding a redirect function to the appropriate department improves user satisfaction and reduces response time.

[1344] "Inquiry content" is text data of questions, requests, and inquiries sent by users through the system.

[1345] The "receiving means" refers to the technical means for inputting the contents of the inquiry sent by the user into the server.

[1346] The "means for formatting into a unified format" refers to a technical means for converting the received query content into a format that is easy to process for subsequent analysis.

[1347] The "analysis means" is a natural language processing technique for extracting key keywords from the content of the inquiry.

[1348] A "keyword" is a word or phrase that contains important information in the query.

[1349] The "means for searching for similar past queries" refers to a technical means for finding similar past queries from a database.

[1350] A "generative AI model" is a system that uses artificial intelligence technology to automatically generate new response messages.

[1351] A "means for generating an auto-response message" is a technical means for automatically creating a response to a user's inquiry using a generative AI model.

[1352] "Means for sending" refers to the technical means for delivering the generated auto-reply message to the user.

[1353] The "redirect function to the appropriate department" is a function that transfers an inquiry to the appropriate department based on the content of the inquiry.

[1354] "Reanalysis measures" are technical measures that perform additional analysis when the results of the initial analysis are unclear.

[1355] "Means of identification" refers to technical means for determining the appropriate department to respond based on reanalysis.

[1356] "Forwarding means" refers to the technical means by which the inquiry is redirected to a specified department.

[1357] "User Interface" means the means by which a user accesses the inquiry portal.

[1358] "Scoring means" refers to a technical means for evaluating detected answer candidates and selecting the most appropriate answer.

[1359] "Natural language processing technology" is an artificial intelligence technology for understanding, analyzing, and processing human language.

[1360] A "prompt sentence" is an instruction sentence that is input into a generative AI model to generate a response.

[1361] The present invention is a system that automates responses to user inquiries, thereby improving the efficiency and productivity of inquiry responses. This system functions through communication between a server, terminals, and users.

[1362] Hardware and software used

[1363] The server is a high-performance computer system, and uses a database management system (DBMS) such as MySQL or PostgreSQL. Natural language processing engines such as TensorFlow and Watson are used for server-side natural language processing. Furthermore, OpenAI's GPT-3 and other generative AI models are used.

[1364] A terminal is a device where a user inputs a query and receives a response message. This can be a PC, tablet, smartphone, or other device. The user interface is implemented as a web application or a mobile application.

[1365] Processing Overview

[1366] 1. Receiving an Inquiry

[1367] Users access a dedicated inquiry portal and enter their questions in the inquiry form. For example, they might enter, "Please tell me what to do if the network is unstable."

[1368] When the user presses the send button, the query content is sent to the server as an HTTP request.

[1369] The server stores the received inquiry in a database.

[1370] 2. Formatting into a unified format

[1371] The server formats the received query content into a specific format (e.g., JSON format), which makes subsequent processing easier.

[1372] 3. Content Analysis

[1373] The server analyzes the formatted query using a natural language processing engine (e.g., TensorFlow) and extracts key keywords, such as "network," "unstable," and "how to respond."

[1374] 4. Search for similar queries from the past

[1375] The server searches the database to identify similar past queries, and the search results are scored to select the most similar queries.

[1376] 5. Generate Auto-Response Messages

[1377] The server uses a generative AI model (e.g., GPT-3) to generate an automated response message based on the extracted keywords and selected past inquiries. The prompt text is "Please generate an automated response message in response to the user's inquiry, 'Please tell me what to do if the network is unstable.'"

[1378] 6. Transmission to the User

[1379] The server then sends the generated auto-reply message to the user's device via email or a notification API (e.g., Firebase Cloud Messaging).

[1380] The user checks the response message sent and uses it to solve the problem.

[1381] 7. Redirect to the appropriate department

[1382] The server re-analyzes the query and identifies the appropriate department.

[1383] The server redirects the inquiry to the specified department, for example, contacting the IT support department via email or an internal notification system.

[1384] The terminal (the terminal of the appropriate department) receives this inquiry and initiates specific actions.

[1385] Specific examples

[1386] Example 1: Inquiry about network trouble

[1387] When a user enters "What to do if the network is unstable" into the inquiry portal and submits it, the server receives the inquiry and stores it in a database. Next, the inquiry is formatted based on keywords such as "network," "unstable," and "how to deal with it," and analyzed using natural language processing technology. After that, similar past inquiries are searched for in the database to identify the most similar case. Using generative AI, an automatic response message is generated that reads, "If the network is unstable, please follow the steps below..." and sent to the user. When the user receives the response message, it becomes easier for them to solve the problem themselves.

[1388] Example 2: Redirection due to misidentification of inquiry content

[1389] If a user sends a query saying "My PC won't start," but the server initially analyzes it and treats it as a network problem, the server reanalyzes it and determines that the query is actually a hardware problem. Therefore, the query is automatically redirected to the IT support department. The terminal (the IT support department's terminal) receives this query and can provide an appropriate solution.

[1390] In this way, the present invention can dramatically improve the efficiency of responding to inquiries and reduce the man-hours of both users and responders.

[1391] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1392] Step 1:

[1393] The user accesses the inquiry portal, enters the question in the inquiry form, and presses the send button.

[1394] Input: The inquiry entered by the user (e.g., "Please tell me what to do if the network is unstable").

[1395] Output: The query stored in the database.

[1396] Specific operation: The server stores the query content received as an HTTP request in a database. In this case, a database management system such as MySQL or PostgreSQL is used.

[1397] Step 2:

[1398] The server formats the received query content into a specified format (e.g., JSON format).

[1399] Input: The raw query stored in the database.

[1400] Output: Pretty-printed JSON representation of the query.

[1401] What happens: The server converts the query to JSON format, removing line breaks and special characters. This formatting makes it easier to parse later.

[1402] Step 3:

[1403] The server analyzes the formatted query using a natural language processing (NLP) engine and extracts key keywords.

[1404] Input: Query content in JSON format.

[1405] Output: Extracted key keywords (e.g., "network," "unstable," "how to respond").

[1406] Specific operation: The server uses a natural language processing engine such as TensorFlow or Watson to analyze the query content and identify important keywords.

[1407] Step 4:

[1408] The server searches a database to identify similar past queries.

[1409] Input: Extracted main keywords.

[1410] Output: The most similar previous queries.

[1411] Specific operation: The server generates an SQL query based on the extracted keywords and searches the database. The search results are scored using a search engine such as Elasticsearch, and the most similar results are selected.

[1412] Step 5:

[1413] The server generates the automated response message using the generative AI model.

[1414] Input: The most similar previous query, extracted keywords, and prompt (e.g., "Please generate an auto-response message in response to the user's query 'What should I do if my network is unstable?'").

[1415] Output: Auto-reply message (e.g. "If your network is unstable, please try the following steps...").

[1416] Specific operation: The server inputs a prompt sentence into a generative AI model such as OpenAI's GPT-3 to generate an automated response message.

[1417] Step 6:

[1418] The server sends the generated automatic response message to the user's terminal.

[1419] Input: Auto-reply message.

[1420] Output: The response message that is displayed on the user's terminal.

[1421] Specific operation: The server sends the generated message to the user using a notification API such as email or Firebase Cloud Messaging. The user then checks the message on their device.

[1422] Step 7:

[1423] The server re-analyzes the query and identifies the appropriate department.

[1424] Input: Query content, reanalysis instructions.

[1425] Output: The appropriate department identified.

[1426] Specific operation: The server uses the natural language processing engine again to re-analyze the inquiry content and identify the appropriate department.

[1427] Step 8:

[1428] The server redirects the query to the specified department.

[1429] Input: Identified appropriate department, inquiry details.

[1430] Output: Query redirected to the appropriate department's terminal.

[1431] Specific operation: The server forwards the inquiry to the specified department using an internal notification system or email. The appropriate department's terminal receives the inquiry and begins to respond.

[1432] (Application example 1)

[1433] 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."

[1434] Current inquiry response systems have difficulty responding quickly and accurately to a wide variety of inquiries from users. They also face the problem of increasing the workload of responders, reducing the efficiency of customer service. Furthermore, it is difficult to provide consistent responses to similar inquiries, making it difficult to improve customer satisfaction.

[1435] 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.

[1436] In this invention, the server includes means for receiving the inquiry content, means for formatting the received inquiry content into a unified format, means for analyzing the inquiry content and extracting key keywords, means for searching past similar inquiries and identifying the most similar inquiry, means for generating a prompt sentence using a generative AI model and generating an automatic response message, means for transmitting the generated response message to an information processing device, and means for reanalyzing the inquiry content, identifying an appropriate department if an appropriate department has not been set, and transferring the inquiry to the appropriate department. This dramatically improves the efficiency of inquiry response and reduces the man-hours of both users and responders.

[1437] The "means for receiving the inquiry content" is a function for transmitting the inquiry content submitted by the user to the server via the network, and for the server to receive the content.

[1438] The "means for converting received inquiry content into a unified format" is a function for converting inquiry content sent in various formats into a consistent unified format.

[1439] "Means for analyzing the content of inquiries and extracting key keywords" refers to a function that utilizes natural language processing technology to identify and extract key keywords from the content of inquiries.

[1440] The "means for searching past similar queries and identifying the most similar query" is a function for comparing a query received from a database in the past with the most similar query and identifying the most similar query.

[1441] "Means for generating prompt sentences using a generative AI model and generating automatic response messages" refers to a function that uses a pre-trained generative AI model to generate prompt sentences required for a response and creates automatic response messages based on those sentences.

[1442] The "means for transmitting the generated response message to the information processing device" is a function for transferring the generated response message to the user's information processing device (for example, a smartphone or a computer) and displaying it.

[1443] "Means for reanalyzing the content of the inquiry, identifying an appropriate department if an appropriate department has not been set, and transferring the inquiry to the relevant department" is a function that reanalyzes the content of the inquiry, and if an appropriate department has not been set, identifies the most appropriate department and transfers the inquiry to that department.

[1444] MODE FOR CARRYING OUT THE INVENTION

[1445] System Configuration

[1446] The system implemented in this invention is broadly composed of the following main means.

[1447] 1. Means of receiving inquiries

[1448] 2. A means of formatting received inquiries into a unified format

[1449] 3. A method for analyzing inquiries and extracting key keywords

[1450] 4. A means to search for similar queries from the past and identify the most similar queries

[1451] 5. A means of generating prompts and auto-response messages using generative AI models

[1452] 6. Means for transmitting the generated response message to the information processing device

[1453] 7. A means to re-analyze the content of the inquiry, identify the appropriate department if one is not set, and transfer the inquiry to the appropriate department.

[1454] Program processing

[1455] Hardware and software usage

[1456] server:

[1457] Hardware: High-performance server

[1458] Software: Database management systems (e.g., MySQL), search engines (e.g., Elasticsearch), natural language processing libraries (e.g., Spacy, BERT), generative AI models (e.g., ChatGPT, GPT-3)

[1459] User device:

[1460] Hardware: Smartphones, tablets, computers

[1461] Software: Mobile application, browser-based interface

[1462] Data processing and calculation flow

[1463] 1. Receiving an Inquiry

[1464] Users access the inquiry form using a smartphone or PC, enter their question, and submit it. The entered inquiry is sent to a server via a network and stored in a database.

[1465] 2. Formatting into a unified format

[1466] The server converts the received query content into a unified format, using a text formatting library (e.g., Python's NLTK) for this conversion process.

[1467] 3. Content Analysis

[1468] The server analyzes the query using natural language processing techniques (e.g., Spacy, BERT) to extract key keywords. This analysis identifies the core of the query.

[1469] 4. Search for similar queries from the past

[1470] The server compares the queries against previous queries in a database to find the most similar queries, using a search engine such as Elasticsearch to score the similarity and identify highly similar queries.

[1471] 5. Generate Auto-Response Messages

[1472] The server uses a generative AI model (e.g., ChatGPT, GPT-3) to generate an automated response message based on the prompt, which includes extracted keywords and information about similar past inquiries.

[1473] 6. Transmission to the User

[1474] The server generates a response message and sends it to the user's device. A real-time notification service (e.g., Firebase) is used to deliver the message to the user quickly and reliably.

[1475] 7. Redirect to the appropriate department

[1476] The server re-analyzes the inquiry and, if an appropriate department is not set, identifies the appropriate department and transfers the inquiry. This allows inquiries that require specialized assistance to be handled appropriately.

[1477] Specific examples

[1478] Example 1: Delivery enquiry

[1479] A user submits an inquiry saying, "The item I ordered hasn't arrived." This inquiry is received by the server, and key keywords such as "order," "item," and "not arrived" are extracted. Similar past inquiries are searched for, and the generative AI model generates a response message saying, "We will confirm your order number and let you know the delivery status. Delivery usually takes 5-7 days." This message is then promptly sent to the user's device.

[1480] Prompt Sentence Examples

[1481] "Please generate an auto-reply message for the following inquiry: 'My order hasn't arrived'"

[1482] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1483] Step 1:

[1484] The user accesses the smart mail-order support application, enters their question into the inquiry form, and presses the send button. At this time, the user may enter something like, "The item I ordered hasn't arrived." The entered data is saved in JSON format within the application and sent to the server via the network. The server receives the inquiry and simultaneously saves the information in a database. The input is the user's question, and the output is saved in the database.

[1485] Step 2:

[1486] The server formats the received raw query content into a specified unified format. This process uses a text formatting library (e.g., Python's NLTK). Specifically, it parses JSON-formatted data, removes unnecessary whitespace and special characters, and formats the query content. The input is the received raw query content, and the output is the data formatted in a unified format.

[1487] Step 3:

[1488] The server analyzes the formatted query using natural language processing technology (e.g., Spacy, BERT) and extracts key keywords. This results in keywords such as "order," "product," and "not delivered." The input is data formatted in a unified format, and the output is the key keywords.

[1489] Step 4:

[1490] The server uses a search engine (e.g., Elasticsearch) to search a database for similar past queries and identify the most similar queries. It scores the search results and selects queries with high similarity. The input is the main keywords, and the output is the past query data with high similarity.

[1491] Step 5:

[1492] The server uses a generative AI model (e.g., ChatGPT, GPT-3) to generate a prompt and generate an auto-response message. This prompt includes extracted keywords and information on the most similar past inquiries. For example, it could be in the format "Please generate an auto-response message in response to the following inquiry: 'The item I ordered has not arrived.'" The input is similar inquiry data, and the output is the generated response message.

[1493] Step 6:

[1494] The server sends the generated automatic response message to the user's device using a real-time notification service (e.g., Firebase). The user can check this response message within the application. The input is the generated response message, and the output is the notification and display on the user's device.

[1495] Step 7:

[1496] The server re-analyzes the query and identifies the appropriate department if one is not already set. If redirection is required, the server forwards the query to the appropriate department. The input is the user's query, and the output is a redirect to the identified appropriate department.

[1497] 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.

[1498] This invention is a system that automates responses to inquiries from users, aiming to improve the efficiency and productivity of inquiry responses. This system receives inquiry content, formats it into a unified format, analyzes it to extract key keywords, searches for similar past inquiries, and uses generative AI to generate an automatic response message that is sent to the user. It can also reanalyze the inquiry content to identify the appropriate department and redirect the inquiry to the identified department. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it also has the function of improving the quality of inquiry responses.

[1499] Program processing overview

[1500] 1. Receiving an Inquiry

[1501] Users access a dedicated inquiry portal, enter their questions into the inquiry form, and then press the send button to send the inquiry to the server.

[1502] The server checks the received inquiry and stores it in a database.

[1503] 2. Formatting into a unified format

[1504] The server converts the received raw query content into a unified format that can be easily parsed and processed.

[1505] 3. Content Analysis

[1506] The server uses NLP technology to analyze the query content, which has been formatted into a unified format, and extracts key keywords and phrases.

[1507] 4. Search for similar queries from the past

[1508] The server searches a database for similar queries from the past to identify the most similar queries, and the search results are evaluated based on a relevance score.

[1509] 5. Generate Auto-Response Messages

[1510] The server uses a generation AI to create an automated response message based on the extracted keywords and identified similar queries.

[1511] 6. Emotional Recognition and Regulation

[1512] The server uses an emotion engine to recognize emotions from the user's query content and expressions, and adjusts the tone and content of the generated response message according to the user's emotions.

[1513] For example, if the user is expressing anger or frustration, the response message will have a more polite and empathetic tone.

[1514] 7. Transmission to the User

[1515] The server sends the generated automatic response message to the user's terminal, so that the user can receive the response quickly.

[1516] 8. Redirect to the appropriate department

[1517] The server re-analyzes the query and, if the appropriate department is not set, forwards the query to the correct department.

[1518] The terminal (the terminal of the appropriate department) responds to the received inquiry and provides the user with an appropriate answer.

[1519] Specific examples

[1520] Example 1: Inquiry about network trouble

[1521] The user enters "Please tell me what to do if the network is unstable" into an inquiry form and submits it. The server receives the inquiry and saves it in a database. The server then formats the inquiry into a unified format and uses NLP technology to extract keywords such as "network," "unstable," and "how to deal with it." The server then searches the database for similar past inquiries and identifies the most similar one. It then uses generative AI to generate an automated response message, which the server then sends to the user. The server then uses an emotion engine to recognize the user's emotions. For example, if the user expresses frustration, the response message is adjusted to show empathy, such as "We apologize for the inconvenience. If the network is unstable, please follow the steps below..."

[1522] Example 2: Redirection due to misidentification of inquiry content

[1523] A user enters "My PC won't start" into an inquiry form and submits it. The server receives the inquiry and saves it in a database. The server analyzes the inquiry and extracts keywords such as "PC" and "won't start." An initial analysis identifies it as a network problem, but reanalysis identifies it as a hardware problem. In this case, the server automatically redirects the inquiry to the IT support department. A terminal (a terminal in the IT support department) receives the inquiry and provides an appropriate solution.

[1524] In this way, the present invention dramatically improves the efficiency of inquiries and reduces the man-hours required by both users and responders. The introduction of an emotion engine can provide an even higher level of user experience.

[1525] The processing flow will be explained below.

[1526] Step 1:

[1527] Users access a dedicated inquiry portal, enter their questions in the inquiry form, and then press the send button to send the inquiry.

[1528] Step 2:

[1529] The server receives the inquiry sent by the user, and the received inquiry is immediately saved in the database.

[1530] Step 3:

[1531] The server formats the raw query content it receives into a uniform format, ensuring data consistency and facilitating subsequent processing.

[1532] Step 4:

[1533] The server uses natural language processing technology to analyze the query, which has been formatted in a unified format, and extracts key keywords and phrases.

[1534] Step 5:

[1535] The server searches the database for similar queries from the past, and identifies the most similar query based on the extracted keywords.

[1536] Step 6:

[1537] The server evaluates the results of similar queries based on relevance scores and selects the most appropriate answers.

[1538] Step 7:

[1539] The server uses AI to generate an automated response message based on the selected answer, utilizing extracted keywords and past inquiry data.

[1540] Step 8:

[1541] The server uses an emotion engine to recognize the user's emotions and determines their emotions from the content of the user's query and the way they input it.

[1542] Step 9:

[1543] The server adjusts the tone and content of the automated response message depending on the emotion recognized by the emotion engine. For example, if the user expresses anger or frustration, the response message will be more polite and empathetic.

[1544] Step 10:

[1545] The server then sends the generated automated response message to the user's terminal, allowing the user to quickly obtain a solution.

[1546] Step 11:

[1547] The user checks the automatic response message on the terminal, which allows the user to immediately obtain a solution to the problem.

[1548] Step 12:

[1549] The server re-analyzes the inquiry content and identifies an appropriate department if one has not been set.

[1550] Step 13:

[1551] The server forwards the query to the appropriate identified department, and the query is automatically redirected to ensure it reaches the correct department.

[1552] Step 14:

[1553] The terminal (in the appropriate department) receives the inquiry and provides the information the user is looking for, allowing for a prompt and professional response.

[1554] By following the above steps, the system of the present invention can efficiently process user inquiries and provide appropriate responses. The introduction of the emotion engine will provide a higher quality user experience.

[1555] Example 2

[1556] 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."

[1557] In conventional inquiry response systems, user inquiries are often handled manually, which results in long response times and reduced user satisfaction. In addition, inquiries are sometimes not accurately routed to the appropriate department based on their content, reducing response efficiency. Furthermore, a uniform response is returned without considering the user's feelings, which also results in a poor user experience.

[1558] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1559] In this invention, the server includes means for receiving inquiry content, means for formatting the received inquiry content into a unified format, means for analyzing the inquiry content and extracting key keywords, means for searching past similar inquiries and identifying the most similar inquiry, means for generating an automatic response message using a generation AI, means for recognizing a user's emotion, means for adjusting the content of the automatic response message according to the recognized emotion, means for sending the generated response message to the user, means for reanalyzing the inquiry content and identifying an appropriate department, means for transferring the inquiry to the identified appropriate department, and means for providing a user interface for accessing the inquiry portal. This dramatically improves the efficiency of inquiry response, improves user satisfaction, enables accurate department transfers, and enables flexible responses according to the user's emotion.

[1560] "Inquiry content" refers to information that a user inputs and transmits to the system through the inquiry portal.

[1561] "Receiving" refers to the process in which the server acquires the inquiry content sent by the user.

[1562] A "uniform format" is a standardized data format that is formatted to facilitate analysis and processing.

[1563] "Analysis" is the process of extracting and understanding meaning and keywords from the content of a query.

[1564] "Keywords" refer to the main concepts or phrases in the query content.

[1565] A "similar query" refers to a query sent by a user in the past that is similar to the current query.

[1566] "Generative AI" refers to a system that uses artificial intelligence to generate automated response messages.

[1567] An "automatic response message" is a response created by a generation AI and automatically sent to the user.

[1568] "Emotion recognition" is a technology that recognizes the emotional state of a user from the content of their inquiry.

[1569] "Adjustment" is the process of changing the tone and content of a response message based on the results of emotion recognition.

[1570] "Sending" is the process by which the server delivers the generated automated response message to the user.

[1571] "Reanalysis" is the process of reviewing the results of an initial analysis to improve accuracy.

[1572] The "appropriate department" refers to the department that can respond to the inquiry most quickly and accurately based on the content of the inquiry.

[1573] "Forwarding" is the process by which the server routes the query to the appropriate department.

[1574] "User Interface" means the interface through which a user accesses the system and inputs and submits inquiries.

[1575] The present invention is a system for automating responses to user inquiries, and aims to improve the efficiency and productivity of inquiries. This system includes the following processes.

[1576] Receiving inquiries

[1577] The user accesses a dedicated inquiry portal, enters their question in the inquiry form, and submits it. For example, they might enter, "I forgot my password. Please tell me how to reset it." The server retrieves the received inquiry and stores it in a MySQL database or similar.

[1578] Formatting to a unified format

[1579] The server formats the received query content into a unified format, for example, by removing unnecessary line breaks and spaces from the received text and converting it to JSON format.

[1580] Content Analysis

[1581] The server uses a natural language processing (NLP) library (e.g., spaCy or NLTK) to parse the query, specifically by performing morphological analysis to extract key keywords and phrases such as "password," "forgot," and "reset."

[1582] Search for similar past inquiries

[1583] The server generates a search query and searches a database for similar queries, calculating a similarity score using techniques such as TF-IDF and Word2Vec, identifying the most relevant query, and retrieving its content and answer.

[1584] Generate auto-response messages

[1585] The server creates a generation prompt for the generative AI model (for example, OpenAI's GPT-3). The prompt text includes the extracted keywords and the content of similar queries. For example, a prompt text such as "A user has asked how to reset their forgotten password. Please follow the steps below to reset it..." is input to the generative AI. The generated automated response message is then obtained.

[1586] Emotion Recognition and Regulation

[1587] The server uses an emotion engine (for example, Microsoft Azure's Text Analytics API) to recognize emotions from the user's inquiry. For example, if the user expresses dissatisfaction or anger, the server adjusts the tone and content of the generated response message. For example, it might make the message more polite, such as, "We apologize for the inconvenience. Please follow the steps below to reset your password."

[1588] Send to user

[1589] The server sends the generated automatic response message to the user's terminal, so that the user can receive the response quickly.

[1590] Redirecting to the appropriate department

[1591] The server re-analyzes the inquiry, and if the appropriate department is not set, it automatically redirects the inquiry to the appropriate department. For example, if an inquiry that was initially misidentified as a network problem is re-analyzed to be a hardware problem, the server redirects the inquiry to the IT support department. The terminal (for example, a terminal in the IT support department) responds appropriately to the received inquiry and provides the user with a solution.

[1592] The above is an embodiment of the present invention, and this system dramatically improves the efficiency of inquiries and reduces the man-hours of both users and responders. The introduction of an emotion engine can provide an even higher level of user experience.

[1593] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1594] Step 1: Receiving an inquiry

[1595] A user accesses a dedicated inquiry portal, enters a question in the inquiry form, and submits it. For example, the user might enter, "I forgot my password, please tell me how to reset it." The server receives the HTTP request sent from the portal and extracts its contents. The input is the user's inquiry text, and the output is the received inquiry text.

[1596] Step 2: Formatting into a unified format

[1597] The server formats the received query content into a unified format. Specifically, it removes unnecessary line breaks and spaces from the received text and converts it into JSON format. The input is the received query text, and the output is the formatted JSON data.

[1598] Step 3: Content Analysis

[1599] The server uses a natural language processing (NLP) library (e.g., spaCy or NLTK) to parse the query formatted in a unified format. Specifically, it performs morphological analysis to extract key keywords and phrases such as "password," "forgot," and "reset." The input is formatted JSON data, and the output is a list of extracted keywords.

[1600] Step 4: Search for similar past queries

[1601] The server generates a search query and searches a database for similar queries, calculating a similarity score using methods such as TF-IDF or Word2Vec. The input is a list of extracted keywords, and the output is related queries and their answers.

[1602] Step 5: Generate an auto-response message

[1603] The server creates a generation prompt for the generative AI model (for example, OpenAI's GPT-3). The prompt text includes the extracted keywords and the content of similar queries. For example, a prompt text such as "A user has asked how to reset their forgotten password. Please follow the steps below to reset it..." is input to the generative AI. The generative AI generates a response message, and the server receives that message. The input is the prompt text, and the output is the generated automated response message.

[1604] Step 6: Emotional awareness and regulation

[1605] The server uses an emotion engine (for example, Microsoft Azure's Text Analytics API) to recognize emotions from the user's inquiry. Specifically, it performs emotion analysis to identify the emotion expressed by the user (for example, anger, frustration, or joy). If the user expresses frustration or anger, it adjusts the tone and content of the generated response message. The input is the generated automated response message and the user's emotion data, and the output is the adjusted response message.

[1606] Step 7: Send to users

[1607] The server sends the generated automatic response message to the user's terminal. Specifically, it returns the generated message to the user's terminal as an HTTP response. The input is the adjusted response message, and the output is the response message sent to the user.

[1608] Step 8: Redirect to the appropriate department

[1609] The server re-analyzes the inquiry content, and if the appropriate department is not set, it automatically identifies the department and redirects the user. Specifically, it performs NLP analysis again to extract keywords to identify the appropriate department. It then sends an email or notification to that department. The input is the inquiry content and the re-analysis results, and the output is the redirected notification. The terminal (for example, a terminal in the IT support department) responds appropriately to the received inquiry and provides the user with a solution.

[1610] (Application example 2)

[1611] 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."

[1612] In modern brick-and-mortar stores, it is important to respond to customer inquiries quickly and appropriately, but this requires a lot of manpower and time, which is a challenge. Furthermore, it is also necessary to improve customer satisfaction by appropriately recognizing customer emotions and providing appropriate responses. However, a system for doing this efficiently has not yet been fully developed.

[1613] 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.

[1614] In this invention, the server includes means for receiving the inquiry content, means for formatting the received inquiry content into a unified format, means for analyzing the inquiry content and extracting key keywords, means for searching for similar past inquiries and identifying the most similar inquiry, means for generating an automatic response message using a generative AI, means for recognizing the user's emotions using a sentiment analysis engine and adjusting the response message, means for sending the generated response message to the user, means for reanalyzing the inquiry content and identifying an appropriate department, means for forwarding the inquiry to the identified appropriate department, means for providing a user interface for accessing the inquiry portal, means for quickly and accurately interacting with users in physical stores using terminals, means for using a generative AI model to generate a response message based on the customer's emotions, and means for generating the automatic response message using prompt sentences, thereby enabling efficiency and productivity improvement in inquiry response.

[1615] The "means for receiving the inquiry content" is an interface for the system to take in the inquiry content input by the user.

[1616] The "means for converting the received query content into a unified format" is a process for converting the received query content into a format that is easy to parse and process.

[1617] "Means for analyzing the content of the query and extracting key keywords" refers to a means for extracting important information from the query using natural language processing technology.

[1618] The "means for searching past similar queries and identifying the most similar query" is a technique for finding an example that is closest to the current query from past queries in a database.

[1619] "Means for generating automated response messages using generative AI" refers to a process that uses AI technology to generate appropriate response messages in response to inquiries.

[1620] "Means for recognizing a user's emotions using an emotion analysis engine and adjusting a response message" is a technology for detecting emotions from a user's input text and adjusting the tone and content of a response message.

[1621] The "means for transmitting the generated response message to the user" refers to a communication means for transmitting the generated response message to the user's terminal.

[1622] The "means for re-analyzing the inquiry content and identifying the appropriate department" is a process for re-analyzing the inquiry content and automatically identifying the appropriate department in charge.

[1623] The "means for transferring the inquiry to the specified appropriate department" is a technique for redirecting the inquiry content to the specified department.

[1624] "Means for providing a user interface for accessing the inquiry portal" refers to a method for providing an interface for a user to submit an inquiry to the system.

[1625] "Means for quickly and accurately communicating with users in a physical store using a terminal" refers to a method for efficiently communicating with customers via a terminal located in a physical store.

[1626] "Means for using a generative AI model to generate a response message based on customer sentiment" refers to an AI model for generating a response that takes into account customer sentiment information.

[1627] The "means for generating an automatic response message using a prompt sentence" is a method for generating an appropriate automatic response message based on a specified prompt sentence.

[1628] In order to carry out the present invention, it is important to build a system that combines a server and a terminal when dealing with customers in a brick-and-mortar store.

[1629] Server processing overview

[1630] The server efficiently processes customer inquiries using the following methods:

[1631] 1. Receiving and storing your inquiry:

[1632] The user enters the inquiry using a terminal installed in the physical store and sends it to the server. The server receives the inquiry and stores it in a database. This step requires a network connection and database software (e.g., MySQL).

[1633] 2. Formatting to Unified Format:

[1634] The server formats the received query content into a unified format for easy parsing. This is done using a script (e.g., a Python script) that performs text preprocessing.

[1635] 3. Extracting Key Keywords:

[1636] The server analyzes the query content using NLP techniques (e.g., NLTK library) and extracts key keywords, which are then used in the next process.

[1637] 4. Search for similar queries from the past:

[1638] The server searches the database for similar queries and identifies the most similar queries. This step utilizes a search algorithm (e.g., Elasticsearch).

[1639] 5. Generate auto-response messages:

[1640] The server generates an automated response message using a generative AI (e.g., GPT-2 model) based on the extracted keywords and identified similar queries, calling the model using an API (e.g., Hugging Face Transformers).

[1641] 6. Emotion recognition and regulation:

[1642] The server uses an emotion analysis engine (e.g., SentimentIntensityAnalyzer) to recognize the user's emotions and adjust the tone of the response message: if the user is expressing frustration, the response message will have a more polite and empathetic tone.

[1643] 7. Sending a response message:

[1644] The server then sends the generated automatic response message back to the user's terminal, allowing the user to receive a prompt reply.

[1645] 8. Redirect to the appropriate department:

[1646] The server re-parses the query and, if the appropriate department is not set, forwards the query to the correct department, along with the information to whom it should be assigned.

[1647] Use of concrete examples and prompts

[1648] For example, if a user types "I can't find the product in the store. Help me!" into a terminal in a physical store, the server receives the content, formats it into a unified format, and extracts keywords such as "product," "can't find," and "help." Sentiment analysis determines that the user's emotion is frustration. Based on this information, the following prompt sentence is provided to the generation AI:

[1649] Prompt: "Generate sentences containing the following keywords: product, not found, help. Sentiment: negative."

[1650] The response message generated using this prompt is "We apologize for the inconvenience. If you cannot find the product, please speak to a store staff member or check this link." The server then sends this message to the user.

[1651] As described above, the present invention is a system that can improve the efficiency and productivity of responding to inquiries and increase customer satisfaction in physical stores.

[1652] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1653] Step 1:

[1654] The user inputs the inquiry details using a terminal installed in the physical store, and then sends the input inquiry details to the server by pressing the send button.

[1655] Input: User's query (e.g. "I can't find the item in your store. Please help!")

[1656] Output: Received query data

[1657] Step 2:

[1658] The server formats the received query into a unified format, preprocessing the text to convert the query to lower case and removing special characters and unnecessary spaces.

[1659] Input: Received inquiry data

[1660] Output: Query data formatted in a unified format

[1661] Step 3:

[1662] The server uses NLP techniques to parse the query and extract key keywords, then uses the NLTK library to tokenize it and extract important nouns and verbs.

[1663] Input: Query data formatted in a unified format

[1664] Output: Extracted keywords (e.g. "product", "cannot find", "help")

[1665] Step 4:

[1666] The server searches the database for similar queries from the past, identifies the most similar queries, calculates a similarity score based on the extracted keywords using Elasticsearch, and identifies the query with the highest score.

[1667] Input: Extracted keywords

[1668] Output: Most similar previous queries

[1669] Step 5:

[1670] The server uses generative AI to generate an automated response message based on the most similar past queries, generating prompt sentences and feeding them into a GPT-2 model to generate an appropriate response message.

[1671] Input: Most similar previous query, prompt (e.g., "Generate a sentence containing the following keywords: product, not found, help. Sentiment: negative.")

[1672] Output: Generated automated response message (e.g., "We apologize for the inconvenience. If you cannot find the product, please contact a store staff member or check this link.")

[1673] Step 6:

[1674] The server uses a sentiment analysis engine to recognize the user's emotions and adjusts the response message. It calculates an emotion score using the SentimentIntensityAnalyzer and changes the tone and content of the response message as needed.

[1675] Input: Generated auto-response message, user inquiry

[1676] Output: Reconciled response message

[1677] Step 7:

[1678] The server then sends the generated response message back to the user's terminal, using a communication protocol to transfer the message so that the user can receive a response quickly.

[1679] Input: Tailored response message

[1680] Output: The response message displayed on the user's terminal.

[1681] Step 8:

[1682] The server reparses the query and redirects the query to the correct department if the appropriate department is not set. Based on the results of the reparse, the server redirects the query.

[1683] Input: Tailored response message, inquiry content

[1684] Output: Enquiry data redirected to the appropriate department

[1685] 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.

[1686] 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.

[1687] 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 robot 414.

[1688] 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.

[1689] 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.

[1690] 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.

[1691] 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).

[1692] 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.

[1693] 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."

[1694] 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.

[1695] 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).

[1696] 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.

[1697] 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.

[1698] 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.

[1699] 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.

[1700] 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.

[1701] 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.

[1702] 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.

[1703] 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.

[1704] 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.

[1705] 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.

[1706] The following is further disclosed regarding the above embodiment.

[1707] (Claim 1)

[1708] A means for receiving an inquiry;

[1709] means for formatting received inquiries into a unified format;

[1710] A means for analyzing the content of the inquiry and extracting key keywords;

[1711] a means for searching past similar queries and identifying the most similar queries;

[1712] a means for generating auto-response messages using generative AI;

[1713] The system includes means for transmitting the generated response message to the user.

[1714] (Claim 2)

[1715] A means of reanalyzing the inquiry content and identifying the appropriate department;

[1716] 10. The system of claim 1, further comprising means for forwarding the inquiry to the identified appropriate department.

[1717] (Claim 3)

[1718] 10. The system of claim 1, further comprising means for providing a user interface for accessing the inquiry portal.

[1719] "Example 1"

[1720] (Claim 1)

[1721] A means for receiving an inquiry;

[1722] means for formatting received inquiries into a unified format;

[1723] A means for analyzing the content of the inquiry and extracting key keywords;

[1724] a means for searching past similar queries and identifying the most similar queries;

[1725] a means for generating an auto-response message using a generative AI model;

[1726] means for transmitting the generated response message to the user;

[1727] A means to provide redirection to the appropriate department;

[1728] A means of reanalyzing the inquiry and identifying the appropriate department;

[1729] A means of forwarding inquiries to the appropriate identified department;

[1730] means for providing a user interface for accessing the inquiry portal;

[1731] a means for scoring the possible answers and selecting the most appropriate answer;

[1732] A means for analyzing the content of the inquiry using natural language processing technology;

[1733] A system that includes a means for generating prompt sentences based on the analysis results and inputting them into a generative AI model.

[1734] (Claim 2)

[1735] a means for reanalyzing the query to identify the appropriate department;

[1736] 10. The system of claim 1, further comprising means for redirecting the query based on the results of the reanalysis.

[1737] (Claim 3)

[1738] 10. The system of claim 1, further comprising means for providing a user interface for accessing the inquiry portal.

[1739] "Application Example 1"

[1740] (Claim 1)

[1741] A means for receiving an inquiry;

[1742] means for formatting received inquiries into a unified format;

[1743] A means for analyzing the content of the inquiry and extracting key keywords;

[1744] a means for searching past similar queries and identifying the most similar queries;

[1745] a means for generating prompt sentences using a generative AI model to generate an auto-response message;

[1746] means for transmitting the generated response message to the information processing device;

[1747] It is a means to reanalyze the content of an inquiry, identify the appropriate department if one is not set, and transfer the inquiry to the appropriate department.

[1748] A system including:

[1749] (Claim 2)

[1750] 2. The system according to claim 1, further comprising means for scoring search results of similar past queries based on the query content, and selecting those with a high degree of similarity.

[1751] (Claim 3)

[1752] 10. The system of claim 1, further comprising: means for providing an input / output interface for accessing the inquiry management portal.

[1753] "Example 2: Combining Emotion Engines"

[1754] (Claim 1)

[1755] A means for receiving an inquiry;

[1756] means for formatting received inquiries into a unified format;

[1757] A means for analyzing the content of the inquiry and extracting key keywords;

[1758] a means for searching past similar queries and identifying the most similar queries;

[1759] a means for generating auto-response messages using generative AI;

[1760] means for recognizing a user's emotion;

[1761] means for tailoring the content of the automated response message in response to the recognized emotion;

[1762] The system includes means for transmitting the generated response message to the user.

[1763] (Claim 2)

[1764] A means of reanalyzing the inquiry content and identifying the appropriate department;

[1765] Further includes a means for forwarding inquiries to the appropriate identified department.

[1766] 10. The system of claim 1.

[1767] (Claim 3)

[1768] and means for providing a user interface for accessing the inquiry portal.

[1769] 10. The system of claim 1.

[1770] "Application example 2 when combining emotion engines"

[1771] (Claim 1)

[1772] A means for receiving an inquiry;

[1773] means for formatting received inquiries into a unified format;

[1774] A means for analyzing the content of the inquiry and extracting key keywords;

[1775] a means for searching past similar queries and identifying the most similar queries;

[1776] a means for generating auto-response messages using generative AI;

[1777] a means for recognizing a user's emotion using a sentiment analysis engine and adjusting a response message;

[1778] The system includes means for transmitting the generated response message to the user.

[1779] (Claim 2)

[1780] A means of reanalyzing the inquiry and identifying the appropriate department;

[1781] 10. The system of claim 1, further comprising means for forwarding the inquiry to the identified appropriate department.

[1782] (Claim 3)

[1783] means for providing a user interface for accessing the inquiry portal;

[1784] A means for quickly and accurately communicating with a user in a physical store using a terminal;

[1785] a means for using a generative AI model to generate a response message based on customer sentiment;

[1786] 10. The system of claim 1, further comprising means for generating an auto-reply message using the prompt sentence. [Explanation of symbols]

[1787] 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. A means for receiving an inquiry; means for formatting received inquiries into a unified format; A means for analyzing the content of the inquiry and extracting key keywords; a means for searching past similar queries and identifying the most similar queries; a means for generating auto-response messages using generative AI; The system includes means for transmitting the generated response message to the user.

2. A means of reanalyzing the inquiry content and identifying the appropriate department; 10. The system of claim 1, further comprising means for forwarding the inquiry to the identified appropriate department.

3. The system of claim 1 further comprising means for providing a user interface for accessing the inquiry portal.

Images