System

Abstract

To provide a system.SOLUTION: A system includes: means for determining an inquiry content; means for issuing a ticked based on the inquiry content; means for generating and making a reply to the inquiry content in the case where the inquiry content is a content to which a reply can be made; means for, in the case where a secondary answer is necessary for the inquiry content, making a reply to the effect; means for analyzing the emotion of the user from the inquiry content; and means for adjusting the priority of the ticket based on the analyzed emotion of the 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] Delays and omissions in responding to inquiries during times when there is a shortage of staff, such as on holidays and at night, are becoming an issue. In addition, some inquiries require specialized knowledge, making it difficult to respond to all inquiries quickly and appropriately. [Means for solving the problem]

[0005] We provide a system that automatically determines the content of inquiries, confirms receipt of the inquiry, creates a PFOS ticket, determines what can be responded to, determines the need for a secondary response and automatically responds to it, determines the level of importance, and executes escalation. This makes it possible to respond appropriately even during times when there is a shortage of manpower, and prevents inquiries from being overlooked. [Brief explanation of the drawings]

[0006] [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. 2 is a sequence diagram showing a flow of processing in the data processing system according to the first embodiment of the first form example. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1 of Embodiment 1. [Figure 13] FIG. 10 is a sequence diagram showing a processing flow of a data processing system in a second embodiment of the second form example. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 of Embodiment 2. [Figure 15] FIG. 10 is a sequence diagram showing the flow of processing in a data processing system according to a third embodiment of the third embodiment. [Figure 16]FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 3 of Embodiment 3. [Figure 17] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the first embodiment of the first form example when an emotion engine is combined. [Figure 18] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1 of Form Example 1 when an emotion engine is combined. [Figure 19] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the second embodiment of the second form example when an emotion engine is combined. [Figure 20] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 of Form Example 2 when an emotion engine is combined. [Figure 21] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the third embodiment of the third form example when an emotion engine is combined. [Figure 22] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 3 of Form Example 3 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0007] 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.

[0008] First, the terms used in the following description will be explained.

[0009] 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, the 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), an APU (Accelerated Processing Unit), or a TPU (TENSOR PROCESSING UNIT (registered trademark)).

[0010] 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.

[0011] 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.

[0012] 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.

[0013] 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."

[0014] [First embodiment]

[0015] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0016] 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.

[0017] 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).

[0018] 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.

[0019] 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.

[0020] 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.

[0021] 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.

[0022] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0023] 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.

[0024] 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.

[0025] 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.

[0026] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0027] "Example 1"

[0028] The system of the present invention uses a natural language processing engine as a means for automatically determining the content of the inquiry. This engine analyzes the content of the inquiry and determines an appropriate response based on that content. For example, if the inquiry is about how to operate a product, it searches a database for related information and automatically responds with information about that operation method.

[0029] "Example 2"

[0030] In addition, as a means of automatically raising PFOS tickets, a ticket containing the necessary information is generated based on the inquiry content and its analysis results and sent to the PFOS system. For example, if the inquiry is about a defect in a specific product, information about the product and the defect content is entered in the ticket and it is escalated to the corresponding department.

[0031] "Example 3"

[0032] Furthermore, as a means of automatically responding to inquiries that require secondary support when they are difficult to answer automatically, a message indicating that secondary support is required is generated and sent to the inquirer based on the analysis results of the inquiry content. For example, if the inquiry is about a highly technical problem with a specific product, the inquirer is notified of this and told to wait for a response from an expert.

[0033] "Example 4"

[0034] Finally, an AI engine is used to determine the importance of the inquiry and escalate it as necessary. This engine determines the importance of the inquiry based on the analysis of the inquiry content and other information (such as the inquiry's information, frequency of inquiries, and time of day the inquiry is made), and escalates the inquiry as necessary. For example, if the inquiry is about a serious defect in a large number of products, the engine will determine the importance of the inquiry and immediately escalate it to an expert.

[0035] The processing flow of each embodiment will be described below.

[0036] "Example 1"

[0037] Step 1: The inquiry is entered into the system.

[0038] Step 2: The natural language processing engine analyzes the query, extracting the main points and key points of the query.

[0039] Step 3: Based on the analysis results, an appropriate response is determined. For example, if the inquiry is about how to operate a product, relevant information is searched for in the database and an answer with information on how to operate the product is automatically provided.

[0040] "Example 2"

[0041] Step 1: The inquiry and its analysis results are entered into the system.

[0042] Step 2: A ticket containing the required information is generated and sent to the PFOS system.

[0043] Step 3: If the inquiry is about a specific product issue, enter the product information and the details of the issue in the ticket and escalate it to the appropriate department.

[0044] "Example 3"

[0045] Step 1: The analysis results of the inquiry are entered into the system.

[0046] Step 2: For inquiries that are difficult to answer automatically, a message is generated indicating that a secondary response is required.

[0047] Step 3: The generated message is sent to the inquirer. For example, if the inquiry is about a highly technical issue with a specific product, the inquirer is notified of this and advised to wait for a response from an expert.

[0048] "Example 4"

[0049] Step 1: The analysis results of the inquiry content and other information (e.g., information about the person making the inquiry, frequency of inquiries, time of day of inquiries, etc.) are input into the AI ​​engine.

[0050] Step 2: The AI ​​engine uses this information to determine the importance of the inquiry.

[0051] Step 3: Escalate the inquiry as needed. For example, if the inquiry is about a major defect in a large number of products, the inquiry will be deemed to be of high importance and escalated immediately to an expert.

[0052] Example 1

[0053] Next, a description will be given of Example 1 of Form 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."

[0054] A problem with conventional inquiry response systems is that it takes a lot of time and effort to assess the content of an inquiry and determine the appropriate response. Furthermore, because the analysis of the inquiry content and database search are done manually, it is inefficient and often unable to provide a prompt response to the user. Furthermore, determining the importance and escalation of an inquiry depended on human judgment, which can lead to a lack of consistency in responses.

[0055] 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.

[0056] In this invention, the server includes a means for passing the query content to a natural language processing engine, a means for the natural language processing engine to analyze the query content, and a means for searching a database based on the analysis results, thereby enabling automatic analysis of the query content and determination of an appropriate response.

[0057] The "means for determining the content of the inquiry" is a means for analyzing the content of the inquiry input by the user and understanding the content.

[0058] The "means for responding by confirming receipt of an inquiry" is a means for confirming that an inquiry from a user has been received and notifying the user of that fact.

[0059] A "means for raising a ticket" is a means for automatically creating a ticket for response based on the content of an inquiry.

[0060] The "means for determining whether or not a reply can be made" is a means for determining whether or not a reply can be made automatically within the system to the inquiry content.

[0061] The "means for automatically replying when a secondary response is required" is a means for notifying the user that a secondary response is required when an automatic reply cannot be made to the inquiry content.

[0062] The "means for determining the importance level and escalating as necessary" is a means for determining the importance level of the inquiry content and escalating it to a higher-level responder as necessary.

[0063] The "means for passing the inquiry content to the natural language processing engine" is a means for inputting the inquiry content received from the user into the natural language processing engine.

[0064] "Means for a natural language processing engine to analyze the content of a query" refers to means for analyzing the content of a query using a natural language processing engine and understanding its intent.

[0065] The "means for searching a database based on the analysis results" refers to a means for searching a database for related information based on the analysis results of the natural language processing engine.

[0066] The "means for returning search results to the user" is a means for returning information obtained from the database to the user.

[0067] This invention relates to a system that automatically determines the content of an inquiry and provides an appropriate response. Specifically, this system analyzes the content of the inquiry using a natural language processing engine, searches a database based on the analysis results, and returns appropriate information to the user.

[0068] System configuration

[0069] The system consists of the following main components:

[0070] 1. User terminal: A device used by a user to input inquiries. This includes smartphones, PCs, tablets, etc.

[0071] 2. Server: Receives the query, passes it to the natural language processing engine, searches the database based on the analysis results, and responds to the user.

[0072] 3. Natural language processing engine: Software for analyzing query content, specifically using generative AI models such as BERT and GPT-3 (registered trademark).

[0073] 4. Database: Data storage for product operation instructions and other related information.

[0074] Program processing

[0075] A user inputs a query through a device. For example, "Please tell me how to take a screenshot on my new smartphone." The device sends this query to a server. The server then passes the received query to a natural language processing engine. The natural language processing engine analyzes the query and determines an appropriate response based on the content.

[0076] As a result of the analysis, it is determined that the inquiry is about "how to operate a smartphone." The server then searches for information about "how to operate a smartphone" in a database. The database contains information about how to operate a product, and the server extracts the relevant information. The server then returns the extracted information to the user.

[0077] Specific examples

[0078] As a concrete example, consider the case where a user makes a query such as, "Please tell me how to take a screenshot on my new smartphone." This query is sent from the device to the server. The server analyzes the query using a natural language processing engine (e.g., GPT-3). As a result of the analysis, it is determined that the query is about "how to operate a smartphone."

[0079] Next, the server searches the database for information on "how to operate a smartphone" and extracts information on "how to take a screenshot." For example, it finds information such as "Press the power button and volume down button simultaneously to take a screenshot." The server then returns this information to the user.

[0080] Prompt Sentence Examples

[0081] Below are some example prompts to input to a generative AI model:

[0082] User Asks: "How do I take a screenshot on my new smartphone?"

[0083] Prompt: "A user is asking how to take a screenshot on their new phone. Please provide the appropriate instructions."

[0084] Using this prompt, the generative AI model generates information to provide appropriate operating instructions.

[0085] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0086] Step 1:

[0087] The user enters a query.

[0088] The user enters a query in text form through the device's interface, for example, "How do I take a screenshot on my new smartphone?" The text is then stored in the device's memory.

[0089] Step 2:

[0090] The terminal sends a query to the server.

[0091] The terminal sends the query entered by the user to the server as an HTTP POST request. The query is sent in JSON format. The input is the user's query text, and the output is an HTTP request to the server.

[0092] Step 3:

[0093] The server passes the query to a natural language processing engine.

[0094] The server parses the received JSON data and extracts the query content. The extracted query content is then input to the generative AI model as a prompt. For example, the prompt might be, "A user is asking how to take a screenshot on their new smartphone. Please provide the appropriate instructions." The input is JSON data, and the output is a prompt.

[0095] Step 4:

[0096] A natural language processing engine analyzes the inquiry.

[0097] A natural language processing engine (e.g., GPT-3) analyzes the prompt and understands the intent of the query. As a result of the analysis, it determines that information on "how to take a screenshot" is needed. The input is the prompt, and the output is the analysis result.

[0098] Step 5:

[0099] The server searches the database based on the analysis results.

[0100] The server searches the database for information on "how to take a screenshot" based on the analysis results. It executes a database query to retrieve the relevant information. The input is the analysis results, and the output is the information retrieved from the database.

[0101] Step 6:

[0102] The server returns the search results to the user.

[0103] The server returns the obtained information to the user. For example, it generates a message such as "Press the power button and volume down button simultaneously to take a screenshot" and sends it to the device as an HTTP response. The user can check this information on the device screen. The input is the information obtained from the database, and the output is the response message to the user.

[0104] (Application example 1)

[0105] Next, a description will be given of Application Example 1 of Embodiment 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."

[0106] In conventional inquiry response systems, analyzing the content of inquiries and generating appropriate responses was often done manually, resulting in a long response time. It was also sometimes difficult to properly determine the importance of the inquiry and escalate it as necessary. This led to issues such as a decline in customer satisfaction and an impediment to efficient operations.

[0107] 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.

[0108] In this invention, the server includes means for determining the content of the inquiry, means for responding to confirm receipt of the inquiry, means for issuing an information management ticket, means for determining whether a reply is possible, means for automatically responding if a secondary response is required, means for determining the importance of the inquiry and escalating the inquiry as necessary, means for analyzing the content of the inquiry and generating an appropriate response, means for displaying the generated response to the user, and means for saving the content of the inquiry and the generated response in a database. This enables the content of the inquiry to be analyzed quickly and accurately and responses to be provided, thereby improving customer satisfaction and streamlining operations.

[0109] The "means for determining the content of the inquiry" is a means for analyzing the content of the inquiry from the user, understanding the content, and determining an appropriate response.

[0110] The "means for responding to and confirming receipt of an inquiry" refers to a means for confirming that an inquiry from a user has been received and notifying the user of that fact.

[0111] The "means for issuing an information management ticket" is a means for creating an information management ticket based on the content of an inquiry and for tracking and managing the inquiry.

[0112] The "means for determining whether or not a reply can be made" is a means for automatically determining whether or not a reply can be made to the inquiry content.

[0113] "Means for automatically responding when a secondary response is required" refers to a means for automatically notifying the user that a secondary response is required when an automatic response cannot be made to the inquiry.

[0114] The "means for determining the importance level and escalating as necessary" is a means for determining the importance level of the inquiry content and escalating it to a higher-level responder as necessary.

[0115] The "means for analyzing the content of an inquiry and generating an appropriate response" is a means for analyzing the content of an inquiry using natural language processing technology and generating an appropriate response based on that content.

[0116] The "means for displaying the generated answer to the user" refers to a means for displaying the generated answer to the user.

[0117] The "means for saving the inquiry content and the generated response in a database" refers to a means for saving the inquiry content and the generated response in a database so that they can be referenced later.

[0118] As an embodiment of the present invention, the following system can be constructed.

[0119] System configuration

[0120] 1. Hardware:

[0121] Server: Analyzes inquiries, generates answers, and manages the database.

[0122] Terminal: The device (e.g., smartphone) through which a user enters a query and receives a response.

[0123] 2. Software:

[0124] Natural language processing engine: Used to analyze the query and generate appropriate answers (e.g., OpenAI® API).

[0125] Database: Used to store queries and generated answers (e.g., SQLite).

[0126] Inquiry management system: Confirms receipt of inquiries, raises information management tickets, and manages escalations.

[0127] Processing flow

[0128] 1. Receiving inquiries:

[0129] The user inputs the inquiry details using a terminal and sends them to the server, which receives the inquiry details and responds by confirming receipt.

[0130] 2. Analysis of inquiry content:

[0131] The server uses a natural language processing engine to analyze the inquiry and determines whether a response is possible based on the analysis results.

[0132] 3. Generate answers:

[0133] If a response is possible, the server uses a natural language processing engine to generate an appropriate answer, which is then displayed to the user.

[0134] 4. Data storage:

[0135] The queries and generated responses are stored in a database for future reference.

[0136] 5. Escalation Management:

[0137] If it is difficult to respond or if the inquiry is of high importance, the server will determine the importance of the inquiry and escalate it as necessary.

[0138] Specific examples

[0139] For example, if a user asks, "The item I ordered hasn't arrived yet. What should I do?", the server will process the following:

[0140] 1. The server receives the inquiry and responds by confirming receipt.

[0141] 2. Use a natural language processing engine to analyze the inquiry and determine whether a response is possible.

[0142] 3. If it is determined that a response is possible, the server generates a response saying, "Can you tell me your order number? I'll check." and displays it to the user.

[0143] 4. The query and generated answer are stored in a database.

[0144] 5. If it is difficult to respond or the request is of high importance, the server will determine the importance of the request and escalate it as necessary.

[0145] Prompt Sentence Examples

[0146] An example of a prompt for the generative AI model is as follows:

[0147] User Question: I haven't received my order, what should I do?

[0148] Please provide an appropriate answer.

[0149] In this way, it becomes possible to analyze inquiries quickly and accurately and provide answers, thereby improving customer satisfaction and streamlining operations.

[0150] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0151] Step 1:

[0152] The user uses a terminal to input the inquiry content and sends it to the server. The input is the user's inquiry content, and the output is the inquiry content sent to the server. Specifically, the user enters "The product I ordered hasn't arrived. What should I do?" into the inquiry form on the smartphone app and presses the send button.

[0153] Step 2:

[0154] The server receives the inquiry and responds by confirming receipt. The input is the inquiry sent by the user, and the output is a response message confirming receipt. Specifically, the server generates a message saying "Your inquiry has been received. Please wait a moment." and sends it to the user.

[0155] Step 3:

[0156] The server uses a natural language processing engine to analyze the inquiry. The input is the inquiry sent by the user, and the output is the analysis result. Specifically, the server calls the OpenAI API and analyzes the inquiry, "The item I ordered hasn't arrived yet. What should I do?"

[0157] Step 4:

[0158] The server determines whether a reply is possible based on the analysis results. The input is the analysis results, and the output is the result of determining whether a reply is possible. Specifically, the server evaluates the analysis results and determines whether a reply is possible.

[0159] Step 5:

[0160] If it is determined that a response is possible, the server uses a natural language processing engine to generate an appropriate response. The input is the analysis result, and the output is the generated response. Specifically, the server calls the OpenAI API again and generates the response, "Can you tell me your order number? I'll check."

[0161] Step 6:

[0162] The server displays the generated answer to the user. The input is the generated answer, and the output is the answer displayed to the user. Specifically, the server sends and displays the message "Can you tell me your order number? We will check it for you" to the user's smartphone.

[0163] Step 7:

[0164] The server stores the query and the generated response in a database. The input is the query and the generated response, and the output is the record stored in the database. Specifically, the server stores the query "My order hasn't arrived yet. What should I do?" and the response "Can you tell me your order number? I'll check." in an SQLite database.

[0165] Step 8:

[0166] If it is difficult to respond or if the inquiry is of high importance, the server determines the importance of the inquiry and escalates it as necessary. The input is the analysis results and importance determination criteria, and the output is an instruction for escalation. Specifically, the server uses an AI model to evaluate the importance of the inquiry and escalates it to a higher-level responder as necessary.

[0167] Example 2

[0168] Next, a description will be given of Example 2 of Form 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."

[0169] In conventional inquiry response systems, analyzing the content of inquiries, opening tickets, and deciding whether to escalate were often done manually, which was a time-consuming and labor-intensive process. In addition, the accuracy of analyzing the content of inquiries was low, which sometimes resulted in delays in escalating to the appropriate department. This led to issues such as lower user satisfaction and a decline in the company's response efficiency.

[0170] 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.

[0171] In this invention, the server includes a means for determining the content of the inquiry, a means for responding to the inquiry and confirming receipt, a means for analyzing the content of the inquiry, a means for extracting necessary information, a means for issuing a PFOS ticket, a means for determining whether a reply is possible, a means for automatically responding if a secondary response is required, and a means for determining the importance of the inquiry and escalating the inquiry as necessary. This enables the content of the inquiry to be analyzed quickly and accurately and escalated to the appropriate department.

[0172] The "means for determining the content of an inquiry" is a function for analyzing the content of an inquiry entered by a user and classifying the content into a specific category or topic.

[0173] The "means for responding to and confirming receipt of an inquiry" is a function for confirming that an inquiry from a user has been received and notifying the user of that fact.

[0174] The "means for analyzing the content of the inquiry" is a function that uses natural language processing technology to analyze the content of the inquiry from the user in detail and extract the necessary information.

[0175] The "means for extracting necessary information" is a function that extracts necessary information such as the product name, details of the defect, and the department in charge from the analyzed inquiry content.

[0176] "Means for raising a PFOS ticket" is a function that automatically generates a ticket containing the inquiry details based on the extracted information.

[0177] The "means for determining whether a reply is possible" is a function for determining whether a reply can be made immediately to the content of the generated ticket.

[0178] The "means for automatically replying when a secondary response is required" is a function that automatically notifies the user that a secondary response is required when an immediate reply cannot be made.

[0179] The "means for determining the importance and escalating as necessary" is a function for determining the importance of the inquiry content and escalating it to an appropriate department as necessary.

[0180] MODE FOR CARRYING OUT THE INVENTION

[0181] This invention is a system that automatically analyzes the content of inquiries, generates tickets containing necessary information, and escalates them to the appropriate department. This system consists of three main elements: users, terminals, and servers.

[0182] 1. Enter your inquiry

[0183] A user uses an interface to input the content of their inquiry. This interface can be accessed through a web browser or a dedicated application. For example, a user accesses a dedicated inquiry form in a web browser and inputs, "The screen of my newly purchased smartphone frequently freezes."

[0184] 2. Submitting your inquiry

[0185] The device sends the query entered by the user to the server as JSON format data.

[0186] 3. Analysis of inquiry content

[0187] The server analyzes the received query using natural language processing (NLP) techniques. For this analysis, it uses a generative AI model such as OpenAI's GPT-4 (registered trademark). Specifically, the server inputs the following prompt sentence into the generative AI model:

[0188] Please analyze the following inquiry and extract the necessary information. Enquiry: The screen of my newly purchased smartphone frequently freezes.

[0189] 4. Extracting ticket information

[0190] The server extracts necessary information based on the analysis results from the generative AI model. For example, it extracts the following information from the analysis results:

[0191] Product Name: Smartphone

[0192] Issue details: Frequent screen freezes

[0193] Department in charge: Technical Support Department

[0194] 5. Ticket Creation

[0195] The server automatically generates a PFOS ticket based on the extracted information, which includes the inquiry content, product information, details of the problem, and the department in charge of handling the problem.

[0196] 6. Submit a ticket

[0197] The server sends the generated ticket to the PFOS system, which then escalates the ticket to the technical support department, where a technical support representative reviews the ticket and begins working on it.

[0198] This system enables prompt and accurate analysis of inquiries and escalation to the appropriate department, thereby improving user satisfaction and the company's response efficiency.

[0199] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0200] Step 1:

[0201] The user uses an interface to input the details of their inquiry. For example, the user accesses a dedicated inquiry form in a web browser and enters, "The screen of my newly purchased smartphone frequently freezes." The inputted inquiry details are sent to the system.

[0202] Step 2:

[0203] The terminal sends the query entered by the user to the server. At this time, the terminal sends the query to the server as JSON format data. The input is the user's query, and the output is JSON format data.

[0204] Step 3:

[0205] The server analyzes the received query using natural language processing (NLP) techniques. For example, a generative AI model such as OpenAI's GPT-4 is used for this analysis. Specifically, the server inputs the following prompt sentence into the generative AI model:

[0206] Please analyze the following inquiry and extract the necessary information. Enquiry: The screen of my newly purchased smartphone frequently freezes.

[0207] The input is the query content in JSON format, and the output is the analysis result.

[0208] Step 4:

[0209] The server extracts necessary information based on the analysis results from the generative AI model. For example, it extracts the following information from the analysis results:

[0210] Product Name: Smartphone

[0211] Issue details: Frequent screen freezes

[0212] Department in charge: Technical Support Department

[0213] The input is the analysis result of the generative AI model, and the output is the extracted information.

[0214] Step 5:

[0215] The server automatically generates a PFOS ticket based on the extracted information. The generated ticket includes the inquiry content, product information, details of the defect, the department in charge, etc. The input is the extracted information, and the output is the generated PFOS ticket.

[0216] Step 6:

[0217] The server sends the generated ticket to the PFOS system, which then escalates the ticket to the technical support department. For example, a technical support representative reviews the ticket and initiates a workaround. The input is the generated PFOS ticket, and the output is the escalated ticket.

[0218] (Application example 2)

[0219] Next, a description will be given of Application Example 2 of Form 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."

[0220] In conventional inquiry response systems, the analysis of inquiry content, ticket creation, and escalation processes are often done manually, which is inefficient and increases the likelihood of response delays and errors. Furthermore, manual handling is required for reporting security incidents, making it difficult to respond quickly.

[0221] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for determining the content of the inquiry, means for responding to confirm receipt of the inquiry, means for issuing a PFOS ticket, means for determining whether a response is possible, means for automatically responding if a secondary response is required, means for determining the severity and escalating the incident as necessary, means for receiving a security incident report, means for analyzing the incident content using a generative AI model, and means for generating and sending a ticket based on the analysis results. This automates the analysis and response of the inquiry content and security incident, enabling rapid and accurate responses.

[0222] The "means for determining the content of the inquiry" is a function for analyzing the content of the inquiry from the user and identifying that content.

[0223] The "means for confirming receipt of an inquiry and responding" is a function for confirming that an inquiry from a user has been received and responding to the inquiry.

[0224] "Means to raise a PFOS ticket" is a function that automatically generates a ticket containing the necessary information based on the inquiry content.

[0225] The "means for determining whether a reply is possible" is a function for determining whether a reply can be made immediately to the inquiry.

[0226] The "means for automatically replying when a secondary response is required" is a function for automatically replying when a secondary response is required for the content of an inquiry.

[0227] The "means for determining the importance and escalating as necessary" is a function for determining the importance of the inquiry content and escalating it to a higher-level department as necessary.

[0228] "Means for receiving reports of security incidents" refers to a function for receiving reports of security incidents from users.

[0229] "Means for analyzing incident details using a generative AI model" refers to a function for analyzing the details of a security incident using a generative AI model.

[0230] The "means for generating and sending a ticket based on the analysis result" is a function for generating a ticket based on the analysis result and sending it to the corresponding department.

[0231] The following system configuration will be described as an embodiment of the present invention.

[0232] System Configuration

[0233] Hardware

[0234] Server: Use a server with a powerful processor and sufficient memory. The server analyzes inquiries, generates tickets, and handles escalations.

[0235] Devices: Users use devices such as smartphones and computers to report inquiries and security incidents.

[0236] software

[0237] Generative AI models: Use generative AI models for natural language processing.

[0238] Inquiry management system: Software for receiving inquiries, analyzing them, and generating tickets.

[0239] PFOS system: A system for managing generated tickets and escalating them to the appropriate department.

[0240] Processing flow

[0241] 1. Receiving an Inquiry:

[0242] A user uses the device to report an inquiry or security incident.

[0243] The inquiry is sent from the terminal to the server.

[0244] 2. Analysis of inquiry content:

[0245] The server uses a generative AI model to analyze the query.

[0246] Based on the analysis results, it is determined whether or not a response can be made to the inquiry.

[0247] 3. Ticket generation:

[0248] The server generates a ticket containing the necessary information based on the analysis results.

[0249] A ticket contains information such as the inquiry content, analysis results, and importance.

[0250] 4. Escalation:

[0251] The server determines the importance of the ticket and escalates it if necessary.

[0252] If escalation is required, the ticket will be sent to the appropriate department.

[0253] Specific examples

[0254] Incident Report Example:

[0255] A user reports that "The firewall has detected unauthorized access."

[0256] The server uses a generative AI model to analyze the report and determine its importance.

[0257] Based on the analysis results, a ticket is generated and sent to the appropriate department.

[0258] Example prompt sentence:

[0259] "The firewall has detected unauthorized access."

[0260] "The server is overloaded."

[0261] "A user made an invalid login attempt."

[0262] In this way, the analysis and response to inquiries and security incidents is automated, enabling quick and accurate responses.

[0263] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0264] Step 1:

[0265] A user uses a terminal to report an inquiry or security incident. The user enters the inquiry or incident details into the input form on the terminal and presses the send button. The input data is sent to the server in text format.

[0266] Step 2:

[0267] The server receives the query sent from the device. The received data is in text format. The server preprocesses this data to input it into the generative AI model. Specifically, the server tokenizes the text data and converts it into a format that the model can understand.

[0268] Step 3:

[0269] The server analyzes the query using a generative AI model. Tokenized text data is input into the generative AI model, and the analysis results are obtained. The analysis results include information such as the query category and importance.

[0270] Step 4:

[0271] The server determines whether a reply is possible based on the analysis results. It also determines whether a reply can be made immediately based on the category and importance of the analysis results. If a reply is possible, the server generates an automatic response message and sends it to the user.

[0272] Step 5:

[0273] The server generates a ticket containing the necessary information based on the analysis results. The ticket contains information such as the inquiry content, analysis results, and importance. The generated ticket is saved in JSON format.

[0274] Step 6:

[0275] The server determines the severity of the ticket and escalates it if necessary. If the severity is high, the ticket is automatically sent to the appropriate department. If escalation is not required, the ticket is handled according to the normal response process.

[0276] Step 7:

[0277] The server sends the generated ticket to the PFOS system, which manages the received ticket and escalates it to the appropriate department. If the ticket submission is successful, the server sends a notification to the user.

[0278] Step 8:

[0279] The user receives a notification from the server on their device. The notification states that the inquiry has been successfully accepted and that the matter has been escalated to the appropriate department. The user can check the notification and make additional inquiries as necessary.

[0280] Example 3

[0281] Next, a description will be given of a third embodiment of the third embodiment. 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."

[0282] In conventional inquiry response systems, the analysis of inquiry content and the determination of importance are often done manually, making it difficult to respond quickly and appropriately. Even with automated response systems, it is difficult to respond to complex inquiries or highly technical problems, and there are cases where secondary responses or escalations are not performed appropriately. This has led to problems such as lower user satisfaction and a worsening of response efficiency.

[0283] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[0284] In this invention, the server includes means for determining the content of the inquiry, means for responding to confirm receipt of the inquiry, means for issuing a ticket, means for determining whether a reply is possible, means for automatically responding if a secondary response is necessary, means for determining the importance of the inquiry and escalating as necessary, means for analyzing the content of the inquiry using a generative AI model, means for determining the category and difficulty of the inquiry based on the analysis results of the content of the inquiry, and means for using information such as the inquirer's information, the frequency of the inquiry, and the time period of the inquiry to determine the importance of the inquiry. This enables quick and accurate analysis of the content of the inquiry and appropriate secondary responses and escalation.

[0285] The "means for determining the content of an inquiry" is a function for analyzing the content of an inquiry sent by a user and classifying the content into a specific category or type of problem.

[0286] The "means for responding by confirming receipt of an inquiry" is a function for, when receiving an inquiry from a user, confirming receipt of the inquiry and sending a message confirming receipt to the user.

[0287] The "means of raising a ticket" is a function that assigns a unique identification number to received inquiries, making them traceable within the management system.

[0288] The "means for determining whether or not a response is possible" is a function for analyzing the content of an inquiry and determining whether or not the inquiry can be handled by the automated response system.

[0289] The "means for automatically responding when a secondary response is required" is a function for notifying the user that a secondary response is required for inquiries that are difficult to handle using the automatic response system.

[0290] "Means for determining the level of importance and escalating as necessary" is a function for evaluating the importance of the inquiry and handing over the response to an expert or higher-level support team as necessary.

[0291] "Means for analyzing inquiry content using a generative AI model" refers to a function that uses a generative AI model to analyze inquiry content through natural language processing and understand its content.

[0292] "Means for determining the category and difficulty of an inquiry based on the analysis results of the inquiry content" is a function for determining which category an inquiry belongs to and how difficult it is based on the analysis results of the generative AI model.

[0293] "Means of using information such as the inquirer's information, the frequency of inquiries, and the time period of inquiries to determine the importance of inquiries" refers to a function for comprehensively evaluating the importance of inquiries by taking into account additional information such as the inquirer's past inquiry history, the frequency of inquiries, and the time period of inquiries.

[0294] This invention relates to a system that analyzes the content of inquiries from users and automatically provides appropriate responses. Specifically, this system analyzes the content of inquiries, determines whether a response is possible, and provides secondary support or escalation as necessary. This system is realized using a server, a terminal, and a generative AI model.

[0295] Hardware and software used

[0296] Server: Receives inquiries, analyzes them, determines their importance, determines whether to proceed with a secondary response, and escalates them.

[0297] Device: The device used by the user to send the inquiry, such as a PC or smartphone.

[0298] Generative AI model: Used to analyze inquiry content and determine importance. Uses natural language processing technology.

[0299] Program processing

[0300] 1. User submits a query:

[0301] The user uses the terminal to send an inquiry, which may include a technical problem with the product or other questions.

[0302] 2. The server receives the query:

[0303] The server receives queries sent by users and stores them in a database.

[0304] 3. The server analyzes the request:

[0305] The server uses the generative AI model to analyze the inquiry, determining the category and difficulty of the inquiry as a result of the analysis.

[0306] 4. The server determines whether a secondary action is necessary:

[0307] If an automatic response is difficult, the server generates a message to the user indicating that a secondary response is required. For example, if the inquiry is about a highly technical problem with a specific product, the server notifies the user and tells them to wait for a response from an expert.

[0308] 5. The server determines the importance of the query:

[0309] The server uses a generative AI model to determine the importance of the inquiry based on the analysis of the inquiry content and other information (e.g., information about the person making the inquiry, the frequency of the inquiry, the time of the inquiry, etc.).

[0310] 6. Server escalates as needed:

[0311] For inquiries that are determined to be of high importance, the server immediately escalates them to an expert. For example, if an inquiry is about a serious defect in a large number of products, the server determines that the inquiry is of high importance and escalates it to an expert.

[0312] Specific examples

[0313] User asks: "An error occurred while installing new software. The error message is 'Error 404'."

[0314] Server Action:

[0315] 1. The server receives the query.

[0316] 2. Analyze the query using a generative AI model.

[0317] 3. The error message 'Error 404' is assumed to be a common problem and can be automatically answered.

[0318] 4. Generate an auto-reply message and send it to the user.

[0319] Prompt Sentence Examples

[0320] Analyze the user's inquiry and generate an automatic response message for the error message 'Error 404'.

[0321] This system enables prompt and appropriate responses to user inquiries. By using the generative AI model, the accuracy of analysis of the inquiry content is improved, and appropriate secondary responses and escalation are realized. The flow of the identification process in Example 3 will be described with reference to FIG. 15.

[0322] Step 1:

[0323] A user submits a query

[0324] Input: The user uses the terminal to enter their inquiry and clicks the send button.

[0325] Specific behavior: The user enters "An error occurred while installing new software. The error message is 'Error 404'" into the inquiry form and submits it.

[0326] Output: Query data is sent from the terminal to the server.

[0327] Step 2:

[0328] The server receives the query

[0329] Input: Query data sent from the terminal.

[0330] Specific operation: The server receives the query data and stores it in the database.

[0331] Output: Saved in the database as "User ID: 12345, Enquiry: 'An error occurred while installing new software. The error message was 'Error 404'."

[0332] Step 3:

[0333] The server analyzes the query

[0334] Input: Query data stored in the database.

[0335] Specific operation: The server uses the generative AI model to analyze the query content. The prompt sentence "An error occurred while installing new software. The error message is 'Error 404'" is input to the generative AI model.

[0336] Output: The analysis results in "General error, technical support required."

[0337] Step 4:

[0338] The server determines whether a secondary response is necessary.

[0339] Input: Analysis results from a generative AI model.

[0340] Specific operation: Based on the analysis results, if the server is unable to provide an automatic response, it will generate a message indicating that secondary action is required.

[0341] Output: A message is generated and sent to the user indicating that a secondary action is required.

[0342] Step 5:

[0343] The server determines the importance of the query

[0344] Input: Analysis results from the generative AI model, plus additional information such as enquirer information, enquiry frequency, and enquiry time period.

[0345] Specific operation: The server determines the importance of the query based on this information.

[0346] Output: Information about tickets that are determined to be of high importance.

[0347] Step 6:

[0348] The server will escalate as needed

[0349] Input: Information about the inquiry that was determined to be of high priority.

[0350] Specific Actions: The server immediately sends an escalation message to the expert.

[0351] Output: An escalation message is sent to the expert.

[0352] In this way, the system receives and analyzes inquiries from users, and provides secondary support or escalation as necessary, thereby achieving a prompt and appropriate response.

[0353] (Application example 3)

[0354] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[0355] In conventional inquiry response systems, the analysis of inquiry content and the determination of importance were often done manually, making it difficult to respond quickly and appropriately. Also, when the inquiry involved a highly technical issue, it could take a long time to escalate the inquiry to the appropriate expert, which was a factor in reducing user satisfaction. Furthermore, when it came to security-related inquiries, a quick response was required, and conventional systems often fell short.

[0356] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means. In this invention, the server includes means for determining the content of the inquiry, means for responding to confirm receipt of the inquiry, means for issuing a ticket, means for determining whether a reply is possible, means for automatically responding if a secondary response is necessary, means for determining the level of importance and escalating the inquiry as necessary, means for analyzing the content of the inquiry using natural language processing technology, means for determining the importance of the inquiry using an AI engine, means for generating an automatic response based on the content of the inquiry, and means for escalating the inquiry according to its importance. This enables quick and accurate analysis of the content of the inquiry, and automatic implementation of an appropriate response according to its importance, thereby improving user satisfaction.

[0357] The "means for determining the content of the inquiry" is a function for analyzing the content of the inquiry from the user and identifying that content.

[0358] The "means for confirming receipt of an inquiry and responding" is a function for confirming that an inquiry from a user has been received and responding to the inquiry.

[0359] "Means for raising tickets" is a function that generates tickets for response based on the content of an inquiry.

[0360] The "means for determining whether or not a reply is possible" is a function for determining whether or not an inquiry can be automatically responded to.

[0361] The "means for automatically responding when a secondary response is required" is a function for notifying the user that a secondary response is required when an automatic response is difficult.

[0362] The "means for determining the importance level and escalating as necessary" is a function for determining the importance level of an inquiry and escalating it to an expert as necessary.

[0363] "Means for analyzing the content of inquiries using natural language processing technology" refers to a function for analyzing the content of inquiries from users using natural language processing technology.

[0364] "Means for determining the importance of an inquiry using an AI engine" is a function for determining the importance of an inquiry using an AI engine.

[0365] The "means for generating an automatic response based on the content of an inquiry" is a function for automatically generating a response based on the content of an inquiry.

[0366] The "means for escalating an inquiry according to its importance" is a function for escalating an inquiry to an appropriate expert according to its importance.

[0367] A system for implementing this invention includes means for determining the content of an inquiry, means for responding to confirm receipt of the inquiry, means for creating a ticket, means for determining whether a response is possible, means for automatically responding if a secondary response is required, means for determining the level of importance and escalating as necessary, means for analyzing the content of the inquiry using natural language processing technology, means for determining the importance of the inquiry using an AI engine, means for generating an automatic response based on the content of the inquiry, and means for escalating the inquiry depending on its importance.

[0368] Hardware and software used

[0369] Hardware: Servers, smartphones, robots

[0370] Software: Natural language processing libraries (e.g., spaCy, NLTK), AI engines (e.g., TENSORFLOW (registered trademark), PyTorch), databases (e.g., MySQL (registered trademark))

[0371] Data processing and calculation

[0372] Query Analysis

[0373] The server receives the user's query in text format and uses a natural language processing library (e.g., spaCy, NLTK) to analyze the text and identify the content of the query.

[0374] Auto-reply

[0375] Based on the analysis results, the server automatically generates pre-prepared answers to simple queries, such as "Try restarting your security system. If that doesn't resolve the issue, you'll need to contact a professional."

[0376] Secondary response notification

[0377] For inquiries that are difficult to answer automatically, the server notifies the user that a secondary response is required. For example, it sends a notification saying, "This is a highly technical issue and requires expert assistance. Please wait a moment."

[0378] Severity determination and escalation

[0379] The server uses an AI engine (e.g., TensorFlow, PyTorch) to determine the importance of the inquiry. For example, it takes into account the content of the inquiry, information about the person making the inquiry, the frequency of inquiries, and the time of day. If the inquiry is determined to be of high importance, it immediately escalates it to an expert.

[0380] Examples and prompts

[0381] Specific examples

[0382] Inquiry: "My home security system has been malfunctioning frequently lately. What should I do?"

[0383] Automated response: "Please try rebooting your security system. If that doesn't work, you'll need professional assistance."

[0384] Secondary response notification: "This is a highly technical issue that requires expert assistance. Please wait a moment."

[0385] Severity assessment: "This request is about a critical issue and will be escalated to a specialist immediately."

[0386] Prompt Sentence Examples

[0387] User asks: "My home security system has been malfunctioning frequently lately. What should I do?"

[0388] Analysis result: "Security system malfunction"

[0389] Automated response: "Please try rebooting your security system. If that doesn't work, you'll need professional assistance."

[0390] Secondary response notification: "This is a highly technical issue that requires expert assistance. Please wait a moment."

[0391] Severity assessment: "This request is about a critical issue and will be escalated to a specialist immediately."

[0392] In this way, by specifically showing the form for implementing the invention, it becomes possible to quickly and accurately analyze the content of the inquiry, and appropriate responses are automatically made according to the importance, thereby improving user satisfaction.

[0393] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[0394] Step 1:

[0395] The user sends an inquiry. The user inputs the inquiry in text format using a smartphone or a robot and sends it to the server. The input data is the user's inquiry.

[0396] Step 2:

[0397] The server receives the query. The server receives the text data sent by the user and begins parsing it using a natural language processing library (e.g., spaCy, NLTK). The input data is the user's query, and the output data is the analysis result.

[0398] Step 3:

[0399] The server analyzes the query. The server uses a natural language processing library to analyze the text and identify the query. Specifically, it performs text tokenization, part-of-speech tagging, semantic analysis, etc. The input data is the user's query, and the output data is the analysis result.

[0400] Step 4:

[0401] The server generates an automatic response. Based on the analysis results, the server automatically generates a pre-prepared response to simple inquiries. For example, it generates a response such as, "Try restarting your security system. If that doesn't resolve the issue, you'll need to contact a specialist." The input data is the analysis results, and the output data is the automatic response.

[0402] Step 5:

[0403] The server sends a secondary response notification. For inquiries that are difficult to answer automatically, the server notifies the user that a secondary response is required. For example, it sends a notification saying, "This problem is a highly technical one and requires expert assistance. Please wait a moment." The input data is the analysis result, and the output data is the secondary response notification.

[0404] Step 6:

[0405] The server determines the importance. The server uses an AI engine (e.g., TensorFlow, PyTorch) to determine the importance of the query. For example, it takes into account the content of the query, information about the person making the query, the frequency of the query, and the time of day. The input data are the analysis results and information about the person making the query, and the output data is the importance determination result.

[0406] Step 7:

[0407] The server performs the escalation. If the server determines that the severity is high, it immediately escalates to an expert. For example, an inquiry about a serious defect is notified to an expert. The input data is the severity judgment result, and the output data is the escalation notification.

[0408] In this way, by performing specific operations at each step and performing data processing and calculations based on input data, it becomes possible to respond quickly and appropriately to user inquiries.

[0409] 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.

[0410] "Example 1"

[0411] In one embodiment of the present invention, an emotion engine that recognizes the user's emotions is used as a means for determining the content of the inquiry. This emotion engine can read emotions not only from the text information contained in the inquiry, but also from the user's tone of voice, facial expression, etc. For example, if the user is feeling angry or frustrated, this information is reflected in the analysis of the inquiry content, allowing for a more appropriate response.

[0412] "Example 2"

[0413] The emotion engine can also adjust the analysis results of the inquiry content based on the user's emotions. For example, if the user is angry, the importance of the inquiry can be increased taking into account the user's level of anger. This can improve user satisfaction.

[0414] "Example 3"

[0415] Furthermore, as a means of escalation, AI is used to determine the importance of an inquiry, taking into account the user's emotions. This AI determines the importance of an inquiry not only based on the analysis of the inquiry content, but also taking into account the user's emotions. For example, if the user is very angry, the inquiry will be determined to be of high importance and will be immediately escalated to an expert.

[0416] The processing flow of each embodiment will be described below.

[0417] "Example 1"

[0418] Step 1: A user query is entered into the system.

[0419] Step 2: The emotion engine reads the user's emotions from the text information contained in the inquiry, tone of voice, facial expressions, etc.

[0420] Step 3: The emotional information read by the emotion engine is reflected in the analysis of the inquiry content.

[0421] "Example 2"

[0422] Step 1: A user query is entered into the system.

[0423] Step 2: The emotion engine reads the user's emotions from the text information contained in the inquiry, tone of voice, facial expressions, etc.

[0424] Step 3: Adjust the analysis results of the inquiry content based on the emotional information read by the emotion engine.

[0425] "Example 3"

[0426] Step 1: A user query is entered into the system.

[0427] Step 2: The emotion engine reads the user's emotions from the text information contained in the inquiry, tone of voice, facial expressions, etc.

[0428] Step 3: The AI ​​determines the importance of the inquiry by taking into account the analysis results of the inquiry content and emotional information from the emotion engine.

[0429] Step 4: Based on the AI's findings, the inquiry is escalated to an expert.

[0430] Example 1

[0431] Next, a description will be given of Example 1 of Form 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."

[0432] In conventional inquiry response systems, the judgment of the inquiry content and the decision on the appropriate response are often done manually, which is inefficient. Also, it is difficult to respond in a way that takes the user's feelings into account, making it an issue to improve user satisfaction. Furthermore, there are cases where high-priority inquiries are not escalated appropriately, so a quick response is required.

[0433] 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.

[0434] In this invention, the server includes means for analyzing the content of the inquiry using a natural language processing engine, means for using an emotion engine that recognizes the user's emotions, and means for determining an appropriate response based on the analysis results, thereby enabling automatic analysis of the content of the inquiry and appropriate response that takes the user's emotions into consideration.

[0435] The "means for determining the content of the inquiry" is a means for analyzing the content of the inquiry sent from the user and identifying the content.

[0436] The "means for responding by confirming receipt of an inquiry" is a means for, when receiving an inquiry from a user, confirming receipt of the inquiry and responding to the user by confirming receipt.

[0437] The "means for raising a ticket" is a means for automatically creating a ticket for managing responses based on the content of an inquiry.

[0438] The "means for determining whether or not a reply can be made" is a means for automatically determining whether or not a reply can be made to the received inquiry.

[0439] The "means for automatically replying when a secondary response is required" is a means for automatically notifying the user when the inquiry cannot be resolved by the primary response.

[0440] The "means for determining the importance level and escalating as necessary" is a means for determining the importance level of the inquiry content and escalating it to a higher-level responder as necessary.

[0441] "Means for analyzing the content of an inquiry using a natural language processing engine" refers to means for analyzing the content of an inquiry using natural language processing technology and understanding the content.

[0442] The "means for using an emotion engine to recognize the user's emotion" is a means for using an emotion engine to recognize the user's emotion and analyzing the emotion contained in the inquiry content.

[0443] The "means for determining an appropriate response based on the analysis results" refers to a means for determining the optimal response method based on the analysis results of the natural language processing engine and the emotion engine.

[0444] The "means for returning the action content to the user" is a means for returning the determined action content to the user.

[0445] MODE FOR CARRYING OUT THE INVENTION

[0446] The system of the present invention automatically determines the content of an inquiry and provides an appropriate response. By combining a natural language processing engine and an emotion engine, this system can analyze the content of a user's inquiry and provide a response that takes the user's emotions into consideration.

[0447] Hardware and software used

[0448] The server uses the following hardware and software to implement the system.

[0449] Hardware: A high-performance server machine (e.g., a server equipped with a server processor)

[0450] software:

[0451] Natural language processing engine: A common natural language processing API (e.g., Google® Cloud Natural Language API)

[0452] Emotion engine: General emotion recognition API (e.g., Microsoft® Azure® Emotion API)

[0453] Database: A common database management system (e.g., MySQL)

[0454] System Operation

[0455] The user inputs and sends the inquiry using a terminal (for example, a smartphone or PC). The server receives the inquiry sent from the user and performs the following processes.

[0456] 1. Analysis of inquiry content:

[0457] The server analyzes the query using a natural language processing engine. The analysis includes text tokenization, part-of-speech tagging, sentence structure analysis, etc. For example, if a user asks, "How do I set up my new smartphone?", the natural language processing engine extracts the keyword "how to set up."

[0458] 2. Recognizing user emotions:

[0459] The server uses an emotion engine to recognize the user's emotions. The emotion engine can read emotions not only from the text information contained in the inquiry, but also from the user's tone of voice and facial expression. For example, if a user makes an inquiry in an angry tone, saying, "This product is completely unusable. Please do something about it," the emotion engine will recognize the user's anger.

[0460] 3. Determining the appropriate response:

[0461] The server determines the appropriate response based on the analysis results of the natural language processing engine and emotion engine. For example, if the inquiry is about how to operate a product and the user is angry, the server will respond quickly and courteously.

[0462] 4. Reply with the response:

[0463] The server then replies to the user with the determined response. The reply is sent in text format and displayed on the user's device. For example, the server might reply with detailed instructions such as, "To set up your new smartphone: 1. Open the settings menu. 2. Select Wi-Fi. 3. Enter your password."

[0464] Prompt Sentence Examples

[0465] An example of a prompt sentence to input to the generative AI model is as follows:

[0466] "When a user asks how to set up their new smartphone, how do you respond?"

[0467] "How do you respond when users are angry with your product?"

[0468] Using these prompts, the generative AI model can suggest appropriate responses.

[0469] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0470] Step 1:

[0471] The user submits a query.

[0472] The user uses a terminal (e.g., a smartphone or PC) to enter the inquiry content and press the send button. The entered inquiry content is sent to the server in text format.

[0473] Input: The inquiry typed by the user (e.g., "How do I set up my new smartphone?")

[0474] Output: The query sent to the server

[0475] Step 2:

[0476] The server receives the query.

[0477] The server receives the inquiry sent by the user and stores it in temporary storage.

[0478] Input: Inquiry sent by the user

[0479] Output: Query content saved in temporary storage

[0480] Step 3:

[0481] The server analyzes the query using a natural language processing engine.

[0482] The server sends the saved query content to a natural language processing engine (e.g., a common natural language processing API) for analysis, which includes text tokenization, part-of-speech tagging, and sentence structure analysis.

[0483] Input: Query content saved in temporary storage

[0484] Output: Analysis results (e.g., keyword "setting method")

[0485] Step 4:

[0486] The server recognizes the user's emotions using an emotion engine.

[0487] The server sends the user's tone of voice and facial expression data along with the content of the query to an emotion engine (e.g., a general emotion recognition API) to recognize the user's emotions. The emotion engine determines emotions such as anger, joy, and sadness.

[0488] Input: Enquiry, user tone of voice and facial expression data

[0489] Output: Emotion recognition result (e.g., "anger")

[0490] Step 5:

[0491] The server determines the appropriate response based on the analysis results.

[0492] The server determines the appropriate response based on the analysis results of the natural language processing engine and emotion engine. For example, if the inquiry is about how to use a product and the user is angry, the server will respond quickly and courteously.

[0493] Input: Analysis results of the natural language processing engine, emotion recognition results of the emotion engine

[0494] Output: The action decided (e.g., "How to set up a new smartphone")

[0495] Step 6:

[0496] The server responds by replying to the user.

[0497] The server then returns the determined response to the user in text format and displays it on the user's device.

[0498] Input: Decided action

[0499] Output: The reply sent to the user (e.g., "To set up your new phone: 1. Open the Settings menu. 2. Select Wi-Fi. 3. Enter your password.")

[0500] (Application example 1)

[0501] Next, a description will be given of Application Example 1 of Embodiment 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."

[0502] Conventional inquiry response systems do not take the user's feelings into consideration when responding, which can result in delays in providing an appropriate response to urgent inquiries. Furthermore, natural language processing technology is not fully utilized when analyzing the content of inquiries, making it difficult to provide an accurate response. This leads to issues such as a decrease in user satisfaction and a loss of service quality.

[0503] 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.

[0504] In this invention, the server includes means for determining the content of the inquiry, means for responding by confirming receipt of the inquiry, means for issuing a ticket, means for determining whether a reply is possible, means for automatically responding if a secondary response is required, means for determining the importance of the inquiry and escalating it as necessary, means for recognizing the user's emotions, and means for responding to an emergency based on the emotions, thereby enabling a prompt and appropriate response that takes the user's emotions into consideration.

[0505] The "means for determining the content of an inquiry" is a technique for analyzing an inquiry from a user, understanding its content, and determining an appropriate response.

[0506] The "means for responding to and confirming receipt of an inquiry" is a technique for confirming that an inquiry from a user has been received and notifying the user of that fact.

[0507] "Means for raising tickets" refers to technology that automatically creates tickets to record the details of inquiries and manage the progress of responses.

[0508] The "means for determining whether a reply is possible" is a technique for determining whether the system can automatically reply to the inquiry.

[0509] "Means for automatically responding when a secondary response is required" is a technology that automatically notifies the user when the inquiry is complex and cannot be resolved by a primary response.

[0510] The "means for determining the importance level and escalating as necessary" is a technique for evaluating the importance level of the inquiry content and escalating it to a higher-level responder as necessary.

[0511] "Means for recognizing user emotions" refers to technology that reads emotions from the user's tone of voice, facial expressions, etc., and reflects that information in the analysis.

[0512] The "means for taking emergency measures based on emotions" is a technique for taking an appropriate response immediately when the user's emotions indicate an emergency.

[0513] A system for implementing this invention includes means for determining the content of an inquiry, means for responding by confirming receipt of the inquiry, means for creating a ticket, means for determining whether a reply can be made, means for automatically responding if a secondary response is required, means for determining the importance of the inquiry and escalating the inquiry as necessary, means for recognizing the user's emotions, and means for taking emergency action based on the emotions.

[0514] System configuration

[0515] This system uses the following hardware and software:

[0516] Hardware: Smartphone, microphone

[0517] Software: Python (registered trademark), speech_recognition library, transformers library, emotion_recognition library

[0518] Processing flow

[0519] 1. Acquiring voice input: The user speaks into the smartphone microphone, asking a question such as, "I've been seeing some suspicious people around my house lately. What should I do?"

[0520] 2. Speech Recognition: The smartphone uses the speech_recognition library to convert speech to text.

[0521] 3. Emotion Recognition: The smartphone uses the emotion_recognition library to recognize the user's emotions from their voice. For example, if the user is feeling anxious or scared, it will recognize that emotion.

[0522] 4. Parsing the query: The server uses the natural language processing engine in the transformers library to parse the query and determine the appropriate response.

[0523] 5. Response execution: If the emotion is "anger" or "fear," the server will take emergency action by contacting a security company. Otherwise, it will provide an appropriate response based on the analysis results.

[0524] Specific examples

[0525] When a user inquires, "I've seen some suspicious people around my house recently," the smartphone recognizes the voice and emotion of the inquiry. The server uses a natural language processing engine to analyze the inquiry, and if it determines that the situation is urgent, it immediately contacts the security company. This allows for a prompt and appropriate response that takes the user's emotions into account.

[0526] Prompt Sentence Examples

[0527] User: I've been seeing some suspicious people around my house lately, what should I do?

[0528] This system enables prompt and appropriate responses that take into account the user's feelings, which is expected to improve service quality and user satisfaction.

[0529] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0530] Step 1:

[0531] The user makes a query into the microphone of the smartphone.

[0532] Input: User's voice

[0533] Output: Audio data

[0534] Specific behavior: A user asks a question such as, "I've been seeing some suspicious people around my house lately. What should I do?"

[0535] Step 2:

[0536] The device uses the speech_recognition library to convert speech to text.

[0537] Input: Audio data

[0538] Output: Text data

[0539] Specific operation: The smartphone passes the recorded voice data to the speech_recognition library for speech recognition, which converts the voice into text.

[0540] Step 3:

[0541] The device uses the emotion_recognition library to recognize the user's emotions from their voice.

[0542] Input: Audio data

[0543] Output: Emotion data (e.g., anxiety, fear)

[0544] Specific operation: The smartphone passes the recorded voice data to the emotion_recognition library for emotion recognition. As a result, the user's emotions are recognized as "anxiety" or "fear."

[0545] Step 4:

[0546] The server uses the natural language processing engine in the transformers library to analyze the query and determine the appropriate response.

[0547] Input: Text data, emotion data

[0548] Output: Response (e.g., urgent response required, general response)

[0549] How it works: The server passes the text data to the transformers library, analyzes the query, and considers the emotional data to determine if an urgent response is required.

[0550] Step 5:

[0551] If the server detects the emotion as "anger" or "fear," it will take emergency action by contacting a security company. Otherwise, it will provide an appropriate response based on the analysis results.

[0552] Input: Response content, emotion data

[0553] Output: Perform emergency response or provide general answer

[0554] Specific operation: If the emotion data is "anger" or "fear," the server calls an API to contact the security company. Otherwise, it provides the user with an appropriate answer based on the analysis results.

[0555] Example 2

[0556] Next, a description will be given of Example 2 of Form 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."

[0557] Conventional inquiry response systems process inquiries without considering the user's feelings, which can lead to low user satisfaction. Furthermore, analyzing the inquiry content and generating tickets takes time, making it difficult to respond quickly. Furthermore, the importance of inquiries is not properly determined, and urgent inquiries are sometimes not escalated appropriately.

[0558] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for determining the content of the inquiry, a means for responding to confirm receipt of the inquiry, a means for issuing a ticket, a means for determining whether a reply is possible, a means for automatically responding if a secondary response is required, a means for determining the importance of the inquiry and escalating it as necessary, a means for analyzing the user's emotions, and a means for adjusting the content of the ticket based on the analysis results. This enables a prompt and appropriate response to the inquiry taking the user's emotions into consideration.

[0559] The "means for determining the content of the inquiry" is a means for analyzing the content of the inquiry received from the user and identifying the content.

[0560] The "means for responding by confirming receipt of an inquiry" is a means for, when receiving an inquiry from a user, confirming receipt of the inquiry and responding to the user by confirming receipt.

[0561] The "means for raising a ticket" is a means for automatically generating a ticket including information that requires a response based on the analyzed inquiry content.

[0562] The "means for determining whether or not a reply can be made" is a means for determining whether or not a reply can be made immediately to the received inquiry.

[0563] The "means for automatically responding when a secondary response is required" is a means for automatically notifying the user that a secondary response is required when an immediate response cannot be given to the inquiry.

[0564] The "means for determining the importance and escalating as necessary" is a means for determining the importance of the inquiry content and escalating it to a higher-level department as necessary.

[0565] The "means for analyzing user emotions" is a means for analyzing emotions from the content of a user's inquiry and evaluating the degree of those emotions.

[0566] The "means for adjusting the content of the ticket based on the analysis results" refers to a means for adjusting the content of the generated ticket based on the results of the user's emotion analysis, so that an appropriate response can be made.

[0567] This invention analyzes the content of inquiries, automatically generates tickets containing necessary information, and sends them to the system. It also analyzes user sentiment and can adjust the importance of inquiries based on the results.

[0568] Hardware and software used

[0569] 1. Terminal

[0570] A device that allows a user to input a query, such as a PC, smartphone, or tablet.

[0571] 2. Server

[0572] This is a computer system for analyzing inquiry content, analyzing emotions, generating tickets, and sending tickets. The following software is installed on the server:

[0573] Generative AI model: A model for natural language processing, such as OpenAI's GPT-4.

[0574] Sentiment analysis API: An API for analyzing user emotions. For example, Microsoft Azure's Text Analytics.

[0575] Data processing and calculation

[0576] 1. Receipt of inquiries

[0577] The terminal receives an inquiry from the user and transmits the contents of the inquiry to the server.

[0578] 2. Analysis of inquiry content

[0579] The server uses the generated AI model to analyze the received inquiry, specifically extracting the product name and defect details from the inquiry.

[0580] 3. Emotion analysis

[0581] The server uses the emotion analysis API to analyze the emotion from the user's inquiry, and evaluates the type and degree of the user's emotion as a result of the analysis.

[0582] 4. Ticket Creation

[0583] Based on the analysis results, the server generates a ticket containing the necessary information, such as product information, the details of the defect, and the results of user sentiment analysis.

[0584] 5. Submit a ticket

[0585] The server generates a ticket and sends it to the system for escalation.

[0586] Specific examples

[0587] A user submits a query

[0588] The user enters "Product B's screen is not displayed" through a web form or application and presses the submit button.

[0589] The device receives the query

[0590] The terminal receives the user's input and sends the query to the server.

[0591] The server analyzes the query

[0592] The server uses the generated AI model to analyze the inquiry, "The screen of Product B is not displayed." As a result of the analysis, the product name "Product B" and the problem description "The screen is not displayed" are extracted.

[0593] The server performs sentiment analysis

[0594] The server uses the emotion analysis API to detect the emotion of "dissatisfaction" from the user's inquiry. The emotion analysis result is evaluated as "the user is very dissatisfied."

[0595] The server generates a ticket

[0596] Based on the analysis results, the server generates a ticket that includes the content "Product B's screen is not displayed" and the sentiment analysis result "The user is very dissatisfied."

[0597] The server sends the ticket

[0598] The server generates a ticket and sends it to the system for escalation.

[0599] Prompt Sentence Examples

[0600] User inquiry: "The battery of product C runs out quickly."

[0601] Sentiment analysis results: "Users are frustrated"

[0602] Generate a ticket with the following content: "Product C's battery runs out quickly. Users are frustrated and need immediate attention."

[0603] In this way, the system can analyze the user's inquiry content and emotions, generate an appropriate ticket, and send it to the system, thereby improving user satisfaction.

[0604] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0605] Step 1:

[0606] The user submits a query.

[0607] Input: The user enters the inquiry details through a web form or application. For example, they enter "Product A won't turn on" and press the submit button.

[0608] Output: The query is sent to the terminal.

[0609] Specific operation: The user enters the inquiry content and presses the send button, which sends the inquiry to the terminal.

[0610] Step 2:

[0611] The terminal receives the inquiry.

[0612] Input: The inquiry sent by the user.

[0613] Output: The query is sent to the server.

[0614] Specific operation: The terminal receives the user's input and sends it to the server. For example, the terminal sends the query "Product A won't turn on" to the server.

[0615] Step 3:

[0616] The server analyzes the query.

[0617] Input: The inquiry sent from the device.

[0618] Output: Analyzed inquiry details (product name, defect details, etc.).

[0619] Specific operation: The server uses the generated AI model to analyze the received inquiry. For example, if the server analyzes the inquiry "Product A won't turn on," it will extract the product name "Product A" and the problem description "Won't turn on."

[0620] Step 4:

[0621] The server analyzes the user's emotions using an emotion engine.

[0622] Input: Parsed query content.

[0623] Output: User emotion analysis results (type and degree of emotion).

[0624] Specific operation: The server uses the emotion analysis API to analyze the emotion from the user's inquiry. For example, the server detects the emotion "anger" from the inquiry "Product A won't turn on" and evaluates the degree of that emotion.

[0625] Step 5:

[0626] The server generates a ticket.

[0627] Input: Analyzed inquiry content and user sentiment analysis results.

[0628] Output: Generated ticket (product information, defect details, sentiment analysis results, etc.).

[0629] Specific operation: The server generates a ticket containing the necessary information based on the analysis results. For example, the server generates a ticket containing the content "Product A won't turn on" and the sentiment analysis result "The user is very angry."

[0630] Step 6:

[0631] The server sends the ticket to the system.

[0632] Input: The generated ticket.

[0633] Output: A ticket is submitted to the system and escalated to the appropriate department.

[0634] Specific operation: The ticket generated by the server is sent to the system and escalated to the corresponding department. For example, the ticket generated by the server is sent to the system and the corresponding department responds promptly.

[0635] (Application example 2)

[0636] Next, a description will be given of Application Example 2 of Form 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."

[0637] Conventional inquiry response systems process inquiries without considering the user's feelings, which can lead to lower user satisfaction. They also fail to properly prioritize inquiries, which can lead to important inquiries being put off. This makes it difficult to respond quickly and appropriately.

[0638] 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.

[0639] In this invention, the server includes a means for determining the content of the inquiry, a means for responding to the inquiry and confirming receipt, a means for issuing a PFOS ticket, a means for determining whether a reply is possible, a means for automatically responding if a secondary response is required, a means for determining the importance of the inquiry and escalating it as necessary, a means for analyzing the user's emotions, and a means for adjusting the priority of the inquiry based on the emotion analysis results, thereby enabling a prompt and appropriate response that takes the user's emotions into consideration.

[0640] The "means for determining the content of an inquiry" is a means for analyzing the content of an inquiry from a user and identifying the content.

[0641] The "means for responding by confirming receipt of an inquiry" is a means for confirming that an inquiry from a user has been received and notifying the user of that fact.

[0642] "Means for raising a PFOS ticket" refers to a means for automatically generating a ticket containing the necessary information based on the inquiry content and sending it to the PFOS system.

[0643] The "means for determining whether or not a reply can be made" is a means for determining whether or not an inquiry can be immediately answered.

[0644] The "means for automatically replying when a secondary response is required" is a means for automatically notifying the user when an immediate reply cannot be made to the inquiry.

[0645] The "means for determining the importance and escalating as necessary" is a means for determining the importance of the inquiry content and escalating it to a higher-level department as necessary.

[0646] The "means for analyzing the user's emotions" is a means for analyzing emotions from the content of the user's inquiry and identifying those emotions.

[0647] The "means for adjusting the priority of an inquiry based on the result of emotion analysis" is a means for adjusting the priority of an inquiry based on the result of emotion analysis of a user.

[0648] To implement the present invention, the following system configuration and program are required.

[0649] System Configuration

[0650] The server includes a means for determining the content of the inquiry, a means for responding by confirming receipt of the inquiry, a means for raising a PFOS ticket, a means for determining whether a response is possible, a means for automatically responding if a secondary response is required, a means for determining the importance of the inquiry and escalating it as necessary, a means for analyzing the user's emotions, and a means for adjusting the priority of the inquiry based on the results of the emotion analysis.

[0651] Program processing explanation

[0652] The server first receives the user's inquiry. Next, it uses natural language processing to analyze the inquiry and identify its content. Based on the analysis results, it determines whether a reply is possible, and if so, responds immediately. If a reply is difficult, it automatically notifies the user that a secondary response is required.

[0653] In addition, the server uses emotion analysis software such as EmotionRecognizer to analyze the user's emotions. Based on the emotion analysis results, the server adjusts the priority of the inquiry and escalates it if necessary. For example, if the user is angry, the server sets the priority of the inquiry high and responds quickly.

[0654] Hardware and software used

[0655] Hardware: Servers, user devices (smartphones, etc.)

[0656] Software: Python, TextBlob, EmotionRecognizer, Requests

[0657] Specific examples

[0658] If a user says, "It seems like my account has been compromised recently. I'm very worried," the server receives this inquiry and analyzes it using natural language processing. From the analysis results, it identifies the user's emotion as "anxiety," and sentiment analysis detects a negative trend. Based on this, a high-priority ticket is generated and sent to the security team.

[0659] Prompt Sentence Examples

[0660] Please enter your security query: I believe my account has been compromised recently and I'm very concerned.

[0661] In this way, a quick and appropriate response that takes into account the user's feelings is possible.

[0662] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0663] Step 1:

[0664] The user inputs the inquiry content in text format using a user device (such as a smartphone). The input text is sent to the server.

[0665] Step 2:

[0666] The server receives the query. The server receives the text data sent by the user and saves it for further processing. The input is the user's query, and the output is the saved text data.

[0667] Step 3:

[0668] The server analyzes the query using natural language processing. The server uses a natural language processing library such as TextBlob to analyze the text data of the query. As a result of the analysis, the subject of the query and related keywords are extracted. The input is the saved text data, and the output is the analysis result.

[0669] Step 4:

[0670] The server determines whether a reply is possible for the inquiry. Based on the analysis results, the server determines whether a reply is possible immediately. For example, it checks the FAQ database and provides an immediate reply if the question is known. The input is the analysis results, and the output is the result of the reply possibility judgment.

[0671] Step 5:

[0672] If the server is able to reply, it sends a reply to the user. If the server determines that the reply is possible, it generates an appropriate reply and sends it to the user terminal. The input is the reply possibility determination result, and the output is a reply message to the user.

[0673] Step 6:

[0674] If the server cannot respond, it notifies the user that a secondary response is required. If the server determines that it cannot respond, it automatically notifies the user. The input is the result of the determination, and the output is a notification message to the user.

[0675] Step 7:

[0676] The server analyzes the user's emotions. The server uses emotion analysis software such as EmotionRecognizer to analyze the user's emotions from the text data of the inquiry. The input is the saved text data, and the output is the emotion analysis result.

[0677] Step 8:

[0678] The server adjusts the priority of the query based on the emotion analysis result. The server sets the priority of the query based on the emotion analysis result. For example, if the user is angry, the priority is set high. The input is the emotion analysis result, and the output is the priority setting result.

[0679] Step 9:

[0680] The server raises a PFOS ticket. The server automatically generates a ticket containing the necessary information based on the inquiry content, sentiment analysis results, and priority setting results, and sends it to the PFOS system. The input is the inquiry content, sentiment analysis results, and priority setting results, and the output is the generated ticket.

[0681] Step 10:

[0682] The server determines the importance of the ticket and escalates it as necessary.The server determines the importance of the generated ticket and escalates it to a higher-level department as necessary.The input is the generated ticket, and the output is the result of the escalation.

[0683] Example 3

[0684] Next, a description will be given of a third embodiment of the third embodiment. 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."

[0685] In conventional inquiry response systems, the analysis of inquiry content and the determination of importance were often done manually, which resulted in a lack of speed and accuracy in response. Also, when the inquiry content included a highly technical issue or when a response that took the user's feelings into consideration was required, the appropriate escalation was not performed. This resulted in problems such as a decrease in user satisfaction and delayed response.

[0686] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[0687] In this invention, the server includes means for determining the content of the inquiry, means for responding by confirming receipt of the inquiry, means for saving the content of the inquiry, means for sending the content of the inquiry to an AI engine for analysis, means for determining whether a reply is possible, means for automatically replying if a secondary response is required, and means for determining the importance of the inquiry and escalating the inquiry as necessary. This enables quick and accurate analysis of the content of the inquiry and appropriate escalation.

[0688] The "means for determining the content of the inquiry" is a function for analyzing the content of the inquiry sent from the user and identifying the content.

[0689] The "means for responding by confirming receipt of an inquiry" is a function for, when receiving an inquiry from a user, confirming receipt of the inquiry and responding to the user by confirming receipt.

[0690] The "means for saving the inquiry content" is a function for saving the received inquiry content in a storage device such as a database.

[0691] "Means for sending the inquiry content to an AI engine for analysis" is a function for sending the received inquiry content to an AI engine and requesting analysis.

[0692] The "means for determining whether a response is possible" is a function for determining whether the inquiry can be resolved by an automatic response.

[0693] The "means for automatically responding when a secondary response is required" is a function for automatically notifying the user that a secondary response is required when the inquiry contains a highly technical issue, for example.

[0694] The "means for determining the importance level and escalating as necessary" is a function for determining the importance level of the inquiry content and escalating to an expert as necessary.

[0695] This invention is a system that analyzes the content of inquiries and provides secondary support or escalation as necessary. Specifically, if an inquiry contains a highly technical issue or is judged to be of high importance, a message urging an expert to handle the issue is automatically generated and sent to the inquirer.

[0696] Hardware and software used

[0697] Server: A central processing unit for analyzing query content and generating messages. For example, a virtual server in a cloud service is used.

[0698] AI engine: Analyzes the content of inquiries and determines their importance, for example, using a cloud-based AI platform.

[0699] Database: Stores query details and user information. For example, a relational database is used.

[0700] Specific operation of the system

[0701] 1. User submits a query:

[0702] A user fills out a contact form on their device and hits the submit button. For example, a user types, "After the new software update, my app keeps crashing. How can I fix this?"

[0703] 2. The server receives the query:

[0704] The server receives the inquiry sent by the user. Specifically, it receives the HTTP request and analyzes its contents.

[0705] 3. The server saves the query in a database:

[0706] The server saves the received inquiry in a database. For example, information such as the "inquiry ID," "user ID," "inquiry content," and "received date and time" is saved in a relational database.

[0707] 4. The server sends the query to the AI ​​engine:

[0708] The server sends the query to the AI ​​engine. Specifically, it converts the query into JSON format and sends a POST request to the AI ​​engine's API endpoint.

[0709] 5. The AI ​​engine analyzes the inquiry:

[0710] The AI ​​engine analyzes the inquiry. For example, it uses natural language processing (NLP) technology to extract keywords from the inquiry and determine the technical difficulty and importance. As a result of the analysis, it generates information such as "advanced technical issues" and "user emotions (anger)."

[0711] 6. The server generates a message based on the analysis:

[0712] The server receives the analysis results from the AI ​​engine and generates a message indicating that a secondary response is required if necessary, such as "This inquiry involves a highly technical issue and requires the assistance of an expert."

[0713] 7. The server sends a message to the user:

[0714] The server generates a message and sends it to the user, either by sending it to the user's email address or by displaying it on the inquiry form screen.

[0715] 8. Server escalates based on severity:

[0716] If the server determines that the inquiry is highly important based on the AI ​​engine's judgment, it immediately escalates it to an expert. For example, if the user is very angry, it will determine that the inquiry is highly important and notify the expert. Specifically, it will send a notification email to the expert's email address or display an alert on a dedicated management screen.

[0717] Specific examples

[0718] Question: "After the new software update, the app keeps crashing. How can I fix this?"

[0719] Analysis results of the AI ​​engine: The inquiry was determined to contain a highly technical issue. The user's emotion was determined to be "anger."

[0720] Produces the following message: "This request is highly technical and requires expert assistance. Please wait a moment."

[0721] Escalation: Send an email to the specialist notifying them that "the user is very upset and requires immediate attention."

[0722] Prompt Sentence Examples

[0723] "My app keeps crashing after the new software update. How can I fix this?"

[0724] The purpose of this system is to provide prompt and accurate support by efficiently analyzing the content of the user's inquiry and taking appropriate action as necessary. The flow of the identification process in the third embodiment will be described with reference to FIG.

[0725] Step 1:

[0726] The user submits a query.

[0727] A user fills out an inquiry form on their device and presses the submit button. As input, the user types "My app keeps crashing after a new software update. How can I fix this?". As output, the inquiry is sent to the server.

[0728] Step 2:

[0729] The server receives the query.

[0730] The server receives the query sent by the user. Specifically, it receives an HTTP request and analyzes its contents. As input, it receives an HTTP request containing the query from the user. As output, the query is analyzed and passed to the next process.

[0731] Step 3:

[0732] The server stores the query in a database.

[0733] The server saves the received inquiry content in a database. Specifically, information such as "inquiry ID," "user ID," "inquiry content," and "received date and time" is saved in a relational database. The parsed inquiry content is provided as input. The inquiry content is saved in the database as output.

[0734] Step 4:

[0735] The server sends the query to the AI ​​engine.

[0736] The server sends the query to the AI ​​engine. Specifically, it converts the query into JSON format and sends a POST request to the AI ​​engine's API endpoint. The query stored in the database is provided as input. The query is sent to the AI ​​engine as output.

[0737] Step 5:

[0738] The AI ​​engine analyzes the inquiry content.

[0739] The AI ​​engine analyzes the inquiry. For example, it uses natural language processing (NLP) technology to extract keywords from the inquiry and determine its technical difficulty and importance. The inquiry sent from the server is provided as input. The output is analysis results, such as "highly technical problem" and "user emotion (anger)."

[0740] Step 6:

[0741] The server generates a message based on the analysis results.

[0742] The server receives the analysis results from the AI ​​engine and generates a message indicating that secondary action is required if necessary. For example, it generates a message such as, "This inquiry involves a highly technical issue and requires expert assistance." The analysis results from the AI ​​engine are provided as input. The message to be sent to the user is generated as output.

[0743] Step 7:

[0744] The server sends a message to the user.

[0745] The server generates a message and sends it to the user. Specifically, it sends the message to the user's email address or displays it on the contact form screen. The generated message is provided as input. The message is sent to the user as output.

[0746] Step 8:

[0747] The server will escalate based on the severity.

[0748] If the server determines that the inquiry is highly important based on the AI ​​engine's judgment, it immediately escalates it to an expert. For example, if the user is very angry, it will judge the inquiry to be of high importance and notify the expert. Specifically, it will send a notification email to the expert's email address or display an alert on a dedicated management screen. The importance judgment result from the AI ​​engine is provided as input. The output is a notification to the expert.

[0749] (Application example 3)

[0750] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[0751] Conventional inquiry response systems have difficulty accurately determining the importance of inquiries and escalating them appropriately. Furthermore, they are unable to respond in a way that takes the user's feelings into account, which can lead to a decline in user satisfaction. Furthermore, while quick and appropriate responses are required when reporting security incidents, current systems have difficulty achieving this.

[0752] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[0753] In this invention, the server includes a means for determining the content of the inquiry, a means for responding to the inquiry and confirming receipt, a means for issuing a PFOS ticket, a means for determining whether the content is a responseable item, a means for automatically responding if a secondary response is required, a means for determining the severity and escalating the issue as necessary, a means for analyzing the user's emotions to determine the severity, a means for accepting incident reports, a means for analyzing the report content using natural language processing, a means for determining the severity based on the analysis results, a means for escalating the issue to an expert if the severity is high, and a means for notifying the user if a secondary response is required. This enables the server to accurately determine the severity of the inquiry and appropriately escalate the issue. Furthermore, since the server can respond in a way that takes the user's emotions into consideration, improved user satisfaction can be expected. Furthermore, a prompt and appropriate response to security incident reports is also possible.

[0754] The "means for determining the content of an inquiry" is a function for analyzing the content of an inquiry from a user and identifying that content.

[0755] The "means for confirming receipt of an inquiry and responding" is a function for confirming that an inquiry from a user has been received and responding to the inquiry.

[0756] "Means to raise a PFOS ticket" is a function that generates a ticket to resolve a problem based on the inquiry content.

[0757] The "means for determining whether or not a reply is possible" is a function for determining whether or not an automatic reply can be made to the inquiry content.

[0758] The "means for automatically replying when a secondary response is required" is a function for notifying the user that a secondary response is required when an automatic response cannot be made to the inquiry content.

[0759] The "means for determining the importance level and escalating as necessary" is a function for determining the importance level of the inquiry content and escalating it to a higher-level person in charge as necessary.

[0760] The "means for analyzing the user's emotions and determining the importance" is a function for analyzing the user's emotions and determining the importance of the inquiry based on the results.

[0761] The "means for receiving incident reports" is a function for receiving security incident reports from users.

[0762] The "means for analyzing the report content using natural language processing" is a function for analyzing the report content from the user using natural language processing technology.

[0763] The "means for determining the importance level based on the analysis results" is a function for determining the importance level of the report content based on the results of analysis using natural language processing.

[0764] The "means for escalating to an expert when the importance is high" is a function for escalating to an expert when the importance of the report content is determined to be high.

[0765] The "means for notifying the user when a secondary response is required" is a function for notifying the user when a secondary response is required.

[0766] The system for implementing the present invention includes a plurality of means for determining the content of an inquiry, assessing its importance, and escalating it as necessary. A specific embodiment of this system will be described below.

[0767] Hardware and software used

[0768] Hardware: Servers, smartphones, smart glasses

[0769] Software: Python, TensorFlow, natural language processing libraries (e.g., spaCy), sentiment analysis libraries

[0770] System configuration

[0771] 1. How to determine the content of the inquiry:

[0772] The server receives the user's query and analyzes it using a natural language processing library (spaCy). Through this analysis, the content of the query is identified and classified into the appropriate category.

[0773] 2. How to acknowledge receipt of inquiries:

[0774] The server notifies the user that the query has been received, and the notification is sent to the smartphone or smart glasses as a push notification.

[0775] 3. How to file a PFOS ticket:

[0776] The server generates a ticket for problem resolution based on the analysis results, which is then managed in an internal system and assigned to a responsible person.

[0777] 4. How to determine if a response is possible:

[0778] The server determines whether the query can be answered automatically, and if so, generates an appropriate answer and sends it to the user.

[0779] 5. Automated response when secondary action is required:

[0780] If the server cannot automatically respond to the inquiry, it notifies the user that a secondary response is required.

[0781] 6. How to determine severity and escalate as needed:

[0782] The server determines the importance of the inquiry and escalates it to a higher-level person if necessary. The importance is determined using a TensorFlow model.

[0783] 7. Analyzing user sentiment and determining importance:

[0784] The server analyzes the user's emotions using an emotion analysis library and determines the importance of the inquiry based on the results.

[0785] 8. How to report an incident:

[0786] The server accepts reports of security incidents from users, who report them via their smartphones or smart glasses.

[0787] 9. Methods for analyzing report content using natural language processing:

[0788] The server analyzes the report content using a natural language processing library (spaCy) to identify the content of the incident.

[0789] 10. How to determine importance based on analysis results:

[0790] The server determines the importance of the report content based on the results of analysis using natural language processing.

[0791] 11. How to escalate to an expert if the issue is of high importance:

[0792] If the server determines that the report content is of high importance, it escalates it to an expert.

[0793] 12. How to notify users when a secondary action is required:

[0794] If a secondary response is required, the server notifies the user. The notification is sent as a push notification to a smartphone or smart glasses.

[0795] Specific examples

[0796] For example, if a user reports that "the server may have been hacked," the server will input the following prompt sentence into the generative AI model:

[0797] Example prompt sentence:

[0798] A user reports that "the server may have been hacked." Analyze the report and determine its severity. Consider the user's feelings and escalate the issue as necessary.

[0799] Based on this prompt, the server analyzes the report, determines its importance, and escalates it to an expert if necessary, enabling a prompt and appropriate response.

[0800] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[0801] Step 1:

[0802] A user uses a smartphone or smart glasses to report an inquiry or security incident. The input is text data entered by the user, and the output is the report data sent to the server.

[0803] Step 2:

[0804] The server analyzes the received report data using a natural language processing library (spaCy). The input is the report data sent by the user, and the output is structured data as the analysis result. Specifically, the report content is tokenized and important keywords and phrases are extracted.

[0805] Step 3:

[0806] Based on the analysis results, the server identifies the category of the inquiry and creates a PFOS ticket. The input is the structured data of the analysis results, and the output is the generated ticket data. Specifically, the server selects the appropriate category based on the analysis results and creates a new ticket in the ticket management system.

[0807] Step 4:

[0808] The server determines whether the query can be automatically answered. The input is the structured data of the analysis result, and the output is a flag indicating whether an automatic response is possible. Specifically, the analysis result is evaluated using a rule-based system, and if an automatic response is possible, an appropriate answer is generated.

[0809] Step 5:

[0810] The server notifies the user if a secondary response is required. The input is a flag indicating whether an automatic response is possible, and the output is a notification message to the user. Specifically, if an automatic response is not possible, the server sends a push notification to the user informing them that a secondary response is required.

[0811] Step 6:

[0812] The server determines the importance of the inquiry and escalates it as necessary. The input is the structured data of the analysis results, and the output is the importance evaluation result and escalation instructions. Specifically, the server uses a TensorFlow model to evaluate the importance, and if it is high, it escalates to an expert.

[0813] Step 7:

[0814] The server analyzes the user's emotions using an emotion analysis library and determines the importance of the inquiry based on the results. The input is the report data from the user, and the output is the emotion analysis results and the importance evaluation results. Specifically, the report content is evaluated using the emotion analysis library, and if the emotion is strong, the importance is set high.

[0815] Step 8:

[0816] The server accepts incident reports and analyzes the report contents using natural language processing. The input is the incident report data from the user, and the output is structured data as the analysis result. Specifically, the report contents are tokenized and important keywords and phrases are extracted.

[0817] Step 9:

[0818] The server determines the importance of the incident based on the analysis results, and escalates it to an expert if the importance is high. The input is structured data from the analysis results, and the output is the importance assessment result and escalation instructions. Specifically, the importance of the incident is assessed based on the results of analysis using natural language processing, and if the importance is high, it is escalated to an expert.

[0819] Step 10:

[0820] The server notifies the user when a secondary response is required. The input is the importance evaluation result, and the output is a notification message to the user. Specifically, if the importance is high and a secondary response is required, the server sends a push notification to the user.

[0821] 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.

[0822] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ) 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.

[0823] Another example of generative AI is Gemini (registered trademark) (Internet search engine). <url: https: gemini.google.com ?hl="ja">) are mentioned.

[0824] 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.

[0825] [Second embodiment]

[0826] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0827] 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.

[0828] 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).

[0829] 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.

[0830] 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.

[0831] 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).

[0832] 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.

[0833] 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.

[0834] 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.

[0835] 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.

[0836] 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.

[0837] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0838] "Example 1"

[0839] The system of the present invention uses a natural language processing engine as a means for automatically determining the content of the inquiry. This engine analyzes the content of the inquiry and determines an appropriate response based on that content. For example, if the inquiry is about how to operate a product, it searches a database for related information and automatically responds with information about that operation method.

[0840] "Example 2"

[0841] In addition, as a means of automatically raising PFOS tickets, a ticket containing the necessary information is generated based on the inquiry content and its analysis results and sent to the PFOS system. For example, if the inquiry is about a defect in a specific product, information about the product and the defect content is entered in the ticket and it is escalated to the corresponding department.

[0842] "Example 3"

[0843] Furthermore, as a means of automatically responding to inquiries that require secondary support when they are difficult to answer automatically, a message indicating that secondary support is required is generated and sent to the inquirer based on the analysis results of the inquiry content. For example, if the inquiry is about a highly technical problem with a specific product, the inquirer is notified of this and told to wait for a response from an expert.

[0844] "Example 4"

[0845] Finally, an AI engine is used to determine the importance of the inquiry and escalate it as necessary. This engine determines the importance of the inquiry based on the analysis of the inquiry content and other information (such as the inquiry's information, frequency of inquiries, and time of day the inquiry is made), and escalates the inquiry as necessary. For example, if the inquiry is about a serious defect in a large number of products, the engine will determine the importance of the inquiry and immediately escalate it to an expert.

[0846] The processing flow of each embodiment will be described below.

[0847] "Example 1"

[0848] Step 1: The inquiry is entered into the system.

[0849] Step 2: The natural language processing engine analyzes the query, extracting the main points and key points of the query.

[0850] Step 3: Based on the analysis results, an appropriate response is determined. For example, if the inquiry is about how to operate a product, relevant information is searched for in the database and an answer with information on how to operate the product is automatically provided.

[0851] "Example 2"

[0852] Step 1: The inquiry and its analysis results are entered into the system.

[0853] Step 2: A ticket containing the required information is generated and sent to the PFOS system.

[0854] Step 3: If the inquiry is about a specific product issue, enter the product information and the details of the issue in the ticket and escalate it to the appropriate department.

[0855] "Example 3"

[0856] Step 1: The analysis results of the inquiry are entered into the system.

[0857] Step 2: For inquiries that are difficult to answer automatically, a message is generated indicating that a secondary response is required.

[0858] Step 3: The generated message is sent to the inquirer. For example, if the inquiry is about a highly technical issue with a specific product, the inquirer is notified of this and advised to wait for a response from an expert.

[0859] "Example 4"

[0860] Step 1: The analysis results of the inquiry content and other information (e.g., information about the person making the inquiry, frequency of inquiries, time of day of inquiries, etc.) are input into the AI ​​engine.

[0861] Step 2: The AI ​​engine uses this information to determine the importance of the inquiry.

[0862] Step 3: Escalate the inquiry as needed. For example, if the inquiry is about a major defect in a large number of products, the inquiry will be deemed to be of high importance and escalated immediately to an expert.

[0863] Example 1

[0864] Next, a description will be given of Example 1 of Form 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."

[0865] A problem with conventional inquiry response systems is that it takes a lot of time and effort to assess the content of an inquiry and determine the appropriate response. Furthermore, because the analysis of the inquiry content and database search are done manually, it is inefficient and often unable to provide a prompt response to the user. Furthermore, determining the importance and escalation of an inquiry depended on human judgment, which can lead to a lack of consistency in responses.

[0866] 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.

[0867] In this invention, the server includes a means for passing the query content to a natural language processing engine, a means for the natural language processing engine to analyze the query content, and a means for searching a database based on the analysis results, thereby enabling automatic analysis of the query content and determination of an appropriate response.

[0868] The "means for determining the content of the inquiry" is a means for analyzing the content of the inquiry input by the user and understanding the content.

[0869] The "means for responding by confirming receipt of an inquiry" is a means for confirming that an inquiry from a user has been received and notifying the user of that fact.

[0870] A "means for raising a ticket" is a means for automatically creating a ticket for response based on the content of an inquiry.

[0871] The "means for determining whether or not a reply can be made" is a means for determining whether or not a reply can be made automatically within the system to the inquiry content.

[0872] The "means for automatically replying when a secondary response is required" is a means for notifying the user that a secondary response is required when an automatic reply cannot be made to the inquiry content.

[0873] The "means for determining the importance level and escalating as necessary" is a means for determining the importance level of the inquiry content and escalating it to a higher-level responder as necessary.

[0874] The "means for passing the inquiry content to the natural language processing engine" is a means for inputting the inquiry content received from the user into the natural language processing engine.

[0875] "Means for a natural language processing engine to analyze the content of a query" refers to means for analyzing the content of a query using a natural language processing engine and understanding its intent.

[0876] The "means for searching a database based on the analysis results" refers to a means for searching a database for related information based on the analysis results of the natural language processing engine.

[0877] The "means for returning search results to the user" is a means for returning information obtained from the database to the user.

[0878] This invention relates to a system that automatically determines the content of an inquiry and provides an appropriate response. Specifically, this system analyzes the content of the inquiry using a natural language processing engine, searches a database based on the analysis results, and returns appropriate information to the user.

[0879] System configuration

[0880] The system consists of the following main components:

[0881] 1. User terminal: A device used by a user to input inquiries. This includes smartphones, PCs, tablets, etc.

[0882] 2. Server: Receives the query, passes it to the natural language processing engine, searches the database based on the analysis results, and responds to the user.

[0883] 3. Natural language processing engine: Software for analyzing query content, specifically using generative AI models such as BERT or GPT-3.

[0884] 4. Database: Data storage for product operation instructions and other related information.

[0885] Program processing

[0886] A user inputs a query through a device. For example, "Please tell me how to take a screenshot on my new smartphone." The device sends this query to a server. The server then passes the received query to a natural language processing engine. The natural language processing engine analyzes the query and determines an appropriate response based on the content.

[0887] As a result of the analysis, it is determined that the inquiry is about "how to operate a smartphone." The server then searches for information about "how to operate a smartphone" in a database. The database contains information about how to operate a product, and the server extracts the relevant information. The server then returns the extracted information to the user.

[0888] Specific examples

[0889] As a concrete example, consider the case where a user makes a query such as, "Please tell me how to take a screenshot on my new smartphone." This query is sent from the device to the server. The server analyzes the query using a natural language processing engine (e.g., GPT-3). As a result of the analysis, it is determined that the query is about "how to operate a smartphone."

[0890] Next, the server searches the database for information on "how to operate a smartphone" and extracts information on "how to take a screenshot." For example, it finds information such as "Press the power button and volume down button simultaneously to take a screenshot." The server then returns this information to the user.

[0891] Prompt Sentence Examples

[0892] Below are some example prompts to input to a generative AI model:

[0893] User Asks: "How do I take a screenshot on my new smartphone?"

[0894] Prompt: "A user is asking how to take a screenshot on their new phone. Please provide the appropriate instructions."

[0895] Using this prompt, the generative AI model generates information to provide appropriate operating instructions.

[0896] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0897] Step 1:

[0898] The user enters a query.

[0899] The user enters a query in text form through the device's interface, for example, "How do I take a screenshot on my new smartphone?" The text is then stored in the device's memory.

[0900] Step 2:

[0901] The terminal sends a query to the server.

[0902] The terminal sends the query entered by the user to the server as an HTTP POST request. The query is sent in JSON format. The input is the user's query text, and the output is an HTTP request to the server.

[0903] Step 3:

[0904] The server passes the query to a natural language processing engine.

[0905] The server parses the received JSON data and extracts the query content. The extracted query content is then input to the generative AI model as a prompt. For example, the prompt might be, "A user is asking how to take a screenshot on their new smartphone. Please provide the appropriate instructions." The input is JSON data, and the output is a prompt.

[0906] Step 4:

[0907] A natural language processing engine analyzes the inquiry.

[0908] A natural language processing engine (e.g., GPT-3) analyzes the prompt and understands the intent of the query. As a result of the analysis, it determines that information on "how to take a screenshot" is needed. The input is the prompt, and the output is the analysis result.

[0909] Step 5:

[0910] The server searches the database based on the analysis results.

[0911] The server searches the database for information on "how to take a screenshot" based on the analysis results. It executes a database query to retrieve the relevant information. The input is the analysis results, and the output is the information retrieved from the database.

[0912] Step 6:

[0913] The server returns the search results to the user.

[0914] The server returns the obtained information to the user. For example, it generates a message such as "Press the power button and volume down button simultaneously to take a screenshot" and sends it to the device as an HTTP response. The user can check this information on the device screen. The input is the information obtained from the database, and the output is the response message to the user.

[0915] (Application example 1)

[0916] Next, a description will be given of Application Example 1 of Form 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."

[0917] In conventional inquiry response systems, analyzing the content of inquiries and generating appropriate responses was often done manually, resulting in a long response time. It was also sometimes difficult to properly determine the importance of the inquiry and escalate it as necessary. This led to issues such as a decline in customer satisfaction and an impediment to efficient operations.

[0918] 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.

[0919] In this invention, the server includes means for determining the content of the inquiry, means for responding to confirm receipt of the inquiry, means for issuing an information management ticket, means for determining whether a reply is possible, means for automatically responding if a secondary response is required, means for determining the importance of the inquiry and escalating the inquiry as necessary, means for analyzing the content of the inquiry and generating an appropriate response, means for displaying the generated response to the user, and means for saving the content of the inquiry and the generated response in a database. This enables the content of the inquiry to be analyzed quickly and accurately and responses to be provided, thereby improving customer satisfaction and streamlining operations.

[0920] The "means for determining the content of the inquiry" is a means for analyzing the content of the inquiry from the user, understanding the content, and determining an appropriate response.

[0921] The "means for responding to and confirming receipt of an inquiry" refers to a means for confirming that an inquiry from a user has been received and notifying the user of that fact.

[0922] The "means for issuing an information management ticket" is a means for creating an information management ticket based on the content of an inquiry and for tracking and managing the inquiry.

[0923] The "means for determining whether or not a reply can be made" is a means for automatically determining whether or not a reply can be made to the inquiry content.

[0924] "Means for automatically responding when a secondary response is required" refers to a means for automatically notifying the user that a secondary response is required when an automatic response cannot be made to the inquiry.

[0925] The "means for determining the importance level and escalating as necessary" is a means for determining the importance level of the inquiry content and escalating it to a higher-level responder as necessary.

[0926] The "means for analyzing the content of an inquiry and generating an appropriate response" is a means for analyzing the content of an inquiry using natural language processing technology and generating an appropriate response based on that content.

[0927] The "means for displaying the generated answer to the user" refers to a means for displaying the generated answer to the user.

[0928] The "means for saving the inquiry content and the generated response in a database" refers to a means for saving the inquiry content and the generated response in a database so that they can be referenced later.

[0929] As an embodiment of the present invention, the following system can be constructed.

[0930] System configuration

[0931] 1. Hardware:

[0932] Server: Analyzes inquiries, generates answers, and manages the database.

[0933] Terminal: The device (e.g., smartphone) through which a user enters a query and receives a response.

[0934] 2. Software:

[0935] Natural language processing engine: Used to analyze the query and generate appropriate answers (e.g., OpenAI API).

[0936] Database: Used to store queries and generated answers (e.g., SQLite).

[0937] Inquiry management system: Confirms receipt of inquiries, raises information management tickets, and manages escalations.

[0938] Processing flow

[0939] 1. Receiving inquiries:

[0940] The user inputs the inquiry details using a terminal and sends them to the server, which receives the inquiry details and responds by confirming receipt.

[0941] 2. Analysis of inquiry content:

[0942] The server uses a natural language processing engine to analyze the inquiry and determines whether a response is possible based on the analysis results.

[0943] 3. Generate answers:

[0944] If a response is possible, the server uses a natural language processing engine to generate an appropriate answer, which is then displayed to the user.

[0945] 4. Data storage:

[0946] The queries and generated responses are stored in a database for future reference.

[0947] 5. Escalation Management:

[0948] If it is difficult to respond or if the inquiry is of high importance, the server will determine the importance of the inquiry and escalate it as necessary.

[0949] Specific examples

[0950] For example, if a user asks, "The item I ordered hasn't arrived yet. What should I do?", the server will process the following:

[0951] 1. The server receives the inquiry and responds by confirming receipt.

[0952] 2. Use a natural language processing engine to analyze the inquiry and determine whether a response is possible.

[0953] 3. If it is determined that a response is possible, the server generates a response saying, "Can you tell me your order number? I'll check." and displays it to the user.

[0954] 4. The query and generated answer are stored in a database.

[0955] 5. If it is difficult to respond or the request is of high importance, the server will determine the importance of the request and escalate it as necessary.

[0956] Prompt Sentence Examples

[0957] An example of a prompt for the generative AI model is as follows:

[0958] User Question: I haven't received my order, what should I do?

[0959] Please provide an appropriate answer.

[0960] In this way, it becomes possible to analyze inquiries quickly and accurately and provide answers, thereby improving customer satisfaction and streamlining operations.

[0961] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0962] Step 1:

[0963] The user uses a terminal to input the inquiry content and sends it to the server. The input is the user's inquiry content, and the output is the inquiry content sent to the server. Specifically, the user enters "The product I ordered hasn't arrived. What should I do?" into the inquiry form on the smartphone app and presses the send button.

[0964] Step 2:

[0965] The server receives the inquiry and responds by confirming receipt. The input is the inquiry sent by the user, and the output is a response message confirming receipt. Specifically, the server generates a message saying "Your inquiry has been received. Please wait a moment." and sends it to the user.

[0966] Step 3:

[0967] The server uses a natural language processing engine to analyze the inquiry. The input is the inquiry sent by the user, and the output is the analysis result. Specifically, the server calls the OpenAI API and analyzes the inquiry, "The item I ordered hasn't arrived yet. What should I do?"

[0968] Step 4:

[0969] The server determines whether a reply is possible based on the analysis results. The input is the analysis results, and the output is the result of determining whether a reply is possible. Specifically, the server evaluates the analysis results and determines whether a reply is possible.

[0970] Step 5:

[0971] If it is determined that a response is possible, the server uses a natural language processing engine to generate an appropriate response. The input is the analysis result, and the output is the generated response. Specifically, the server calls the OpenAI API again and generates the response, "Can you tell me your order number? I'll check."

[0972] Step 6:

[0973] The server displays the generated answer to the user. The input is the generated answer, and the output is the answer displayed to the user. Specifically, the server sends and displays the message "Can you tell me your order number? We will check it for you" to the user's smartphone.

[0974] Step 7:

[0975] The server stores the query and the generated response in a database. The input is the query and the generated response, and the output is the record stored in the database. Specifically, the server stores the query "My order hasn't arrived yet. What should I do?" and the response "Can you tell me your order number? I'll check." in an SQLite database.

[0976] Step 8:

[0977] If it is difficult to respond or if the inquiry is of high importance, the server determines the importance of the inquiry and escalates it as necessary. The input is the analysis results and importance determination criteria, and the output is an instruction for escalation. Specifically, the server uses an AI model to evaluate the importance of the inquiry and escalates it to a higher-level responder as necessary.

[0978] Example 2

[0979] Next, a description will be given of Example 2 of Form 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."

[0980] In conventional inquiry response systems, analyzing the content of inquiries, opening tickets, and deciding whether to escalate were often done manually, which was a time-consuming and labor-intensive process. In addition, the accuracy of analyzing the content of inquiries was low, which sometimes resulted in delays in escalating to the appropriate department. This led to issues such as lower user satisfaction and a decline in the company's response efficiency.

[0981] 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.

[0982] In this invention, the server includes a means for determining the content of the inquiry, a means for responding to the inquiry and confirming receipt, a means for analyzing the content of the inquiry, a means for extracting necessary information, a means for issuing a PFOS ticket, a means for determining whether a reply is possible, a means for automatically responding if a secondary response is required, and a means for determining the importance of the inquiry and escalating the inquiry as necessary. This enables the content of the inquiry to be analyzed quickly and accurately and escalated to the appropriate department.

[0983] The "means for determining the content of an inquiry" is a function for analyzing the content of an inquiry entered by a user and classifying the content into a specific category or topic.

[0984] The "means for responding to and confirming receipt of an inquiry" is a function for confirming that an inquiry from a user has been received and notifying the user of that fact.

[0985] The "means for analyzing the content of the inquiry" is a function that uses natural language processing technology to analyze the content of the inquiry from the user in detail and extract the necessary information.

[0986] The "means for extracting necessary information" is a function that extracts necessary information such as the product name, details of the defect, and the department in charge from the analyzed inquiry content.

[0987] "Means for raising a PFOS ticket" is a function that automatically generates a ticket containing the inquiry details based on the extracted information.

[0988] The "means for determining whether a reply is possible" is a function for determining whether a reply can be made immediately to the content of the generated ticket.

[0989] The "means for automatically replying when a secondary response is required" is a function that automatically notifies the user that a secondary response is required when an immediate reply cannot be made.

[0990] The "means for determining the importance and escalating as necessary" is a function for determining the importance of the inquiry content and escalating it to an appropriate department as necessary.

[0991] MODE FOR CARRYING OUT THE INVENTION

[0992] This invention is a system that automatically analyzes the content of inquiries, generates tickets containing necessary information, and escalates them to the appropriate department. This system consists of three main elements: users, terminals, and servers.

[0993] 1. Enter your inquiry

[0994] A user uses an interface to input the content of their inquiry. This interface can be accessed through a web browser or a dedicated application. For example, a user accesses a dedicated inquiry form in a web browser and inputs, "The screen of my newly purchased smartphone frequently freezes."

[0995] 2. Submitting your inquiry

[0996] The device sends the query entered by the user to the server as JSON format data.

[0997] 3. Analysis of inquiry content

[0998] The server analyzes the received query using natural language processing (NLP) techniques. For example, a generative AI model such as OpenAI's GPT-4 is used for this analysis. Specifically, the server inputs the following prompt sentence into the generative AI model:

[0999] Please analyze the following inquiry and extract the necessary information. Enquiry: The screen of my newly purchased smartphone frequently freezes.

[1000] 4. Extracting ticket information

[1001] The server extracts necessary information based on the analysis results from the generative AI model. For example, it extracts the following information from the analysis results:

[1002] Product Name: Smartphone

[1003] Issue details: Frequent screen freezes

[1004] Department in charge: Technical Support Department

[1005] 5. Ticket Creation

[1006] The server automatically generates a PFOS ticket based on the extracted information, which includes the inquiry content, product information, details of the problem, and the department in charge of handling the problem.

[1007] 6. Submit a ticket

[1008] The server sends the generated ticket to the PFOS system, which then escalates the ticket to the technical support department, where a technical support representative reviews the ticket and begins working on it.

[1009] This system enables prompt and accurate analysis of inquiries and escalation to the appropriate department, thereby improving user satisfaction and the company's response efficiency.

[1010] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1011] Step 1:

[1012] The user uses an interface to input the details of their inquiry. For example, the user accesses a dedicated inquiry form in a web browser and enters, "The screen of my newly purchased smartphone frequently freezes." The inputted inquiry details are sent to the system.

[1013] Step 2:

[1014] The terminal sends the query entered by the user to the server. At this time, the terminal sends the query to the server as JSON format data. The input is the user's query, and the output is JSON format data.

[1015] Step 3:

[1016] The server analyzes the received query using natural language processing (NLP) techniques. For example, a generative AI model such as OpenAI's GPT-4 is used for this analysis. Specifically, the server inputs the following prompt sentence into the generative AI model:

[1017] Please analyze the following inquiry and extract the necessary information. Enquiry: The screen of my newly purchased smartphone frequently freezes.

[1018] The input is the query content in JSON format, and the output is the analysis result.

[1019] Step 4:

[1020] The server extracts necessary information based on the analysis results from the generative AI model. For example, it extracts the following information from the analysis results:

[1021] Product Name: Smartphone

[1022] Issue details: Frequent screen freezes

[1023] Department in charge: Technical Support Department

[1024] The input is the analysis result of the generative AI model, and the output is the extracted information.

[1025] Step 5:

[1026] The server automatically generates a PFOS ticket based on the extracted information. The generated ticket includes the inquiry content, product information, details of the defect, the department in charge, etc. The input is the extracted information, and the output is the generated PFOS ticket.

[1027] Step 6:

[1028] The server sends the generated ticket to the PFOS system, which then escalates the ticket to the technical support department. For example, a technical support representative reviews the ticket and initiates a workaround. The input is the generated PFOS ticket, and the output is the escalated ticket.

[1029] (Application example 2)

[1030] Next, a description will be given of Application Example 2 of Form 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."

[1031] In conventional inquiry response systems, the analysis of inquiry content, ticket creation, and escalation processes are often done manually, which is inefficient and increases the likelihood of response delays and errors. Furthermore, manual handling is required for reporting security incidents, making it difficult to respond quickly.

[1032] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for determining the content of the inquiry, means for responding to confirm receipt of the inquiry, means for issuing a PFOS ticket, means for determining whether a response is possible, means for automatically responding if a secondary response is required, means for determining the severity and escalating the incident as necessary, means for receiving a security incident report, means for analyzing the incident content using a generative AI model, and means for generating and sending a ticket based on the analysis results. This automates the analysis and response of the inquiry content and security incident, enabling rapid and accurate responses.

[1033] The "means for determining the content of the inquiry" is a function for analyzing the content of the inquiry from the user and identifying that content.

[1034] The "means for confirming receipt of an inquiry and responding" is a function for confirming that an inquiry from a user has been received and responding to the inquiry.

[1035] "Means to raise a PFOS ticket" is a function that automatically generates a ticket containing the necessary information based on the inquiry content.

[1036] The "means for determining whether a reply is possible" is a function for determining whether a reply can be made immediately to the inquiry.

[1037] The "means for automatically replying when a secondary response is required" is a function for automatically replying when a secondary response is required for the content of an inquiry.

[1038] The "means for determining the importance and escalating as necessary" is a function for determining the importance of the inquiry content and escalating it to a higher-level department as necessary.

[1039] "Means for receiving reports of security incidents" refers to a function for receiving reports of security incidents from users.

[1040] "Means for analyzing incident details using a generative AI model" refers to a function for analyzing the details of a security incident using a generative AI model.

[1041] The "means for generating and sending a ticket based on the analysis result" is a function for generating a ticket based on the analysis result and sending it to the corresponding department.

[1042] The following system configuration will be described as an embodiment of the present invention.

[1043] System Configuration

[1044] Hardware

[1045] Server: Use a server with a powerful processor and sufficient memory. The server analyzes inquiries, generates tickets, and handles escalations.

[1046] Devices: Users use devices such as smartphones and computers to report inquiries and security incidents.

[1047] software

[1048] Generative AI models: Use generative AI models for natural language processing.

[1049] Inquiry management system: Software for receiving inquiries, analyzing them, and generating tickets.

[1050] PFOS system: A system for managing generated tickets and escalating them to the appropriate department.

[1051] Processing flow

[1052] 1. Receiving an Inquiry:

[1053] A user uses the device to report an inquiry or security incident.

[1054] The inquiry is sent from the terminal to the server.

[1055] 2. Analysis of inquiry content:

[1056] The server uses a generative AI model to analyze the query.

[1057] Based on the analysis results, it is determined whether or not a response can be made to the inquiry.

[1058] 3. Ticket generation:

[1059] The server generates a ticket containing the necessary information based on the analysis results.

[1060] A ticket contains information such as the inquiry content, analysis results, and importance.

[1061] 4. Escalation:

[1062] The server determines the importance of the ticket and escalates it if necessary.

[1063] If escalation is required, the ticket will be sent to the appropriate department.

[1064] Specific examples

[1065] Incident Report Example:

[1066] A user reports that "The firewall has detected unauthorized access."

[1067] The server uses a generative AI model to analyze the report and determine its importance.

[1068] Based on the analysis results, a ticket is generated and sent to the appropriate department.

[1069] Example prompt sentence:

[1070] "The firewall has detected unauthorized access."

[1071] "The server is overloaded."

[1072] "A user made an invalid login attempt."

[1073] In this way, the analysis and response to inquiries and security incidents is automated, enabling quick and accurate responses.

[1074] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1075] Step 1:

[1076] A user uses a terminal to report an inquiry or security incident. The user enters the inquiry or incident details into the input form on the terminal and presses the send button. The input data is sent to the server in text format.

[1077] Step 2:

[1078] The server receives the query sent from the device. The received data is in text format. The server preprocesses this data to input it into the generative AI model. Specifically, the server tokenizes the text data and converts it into a format that the model can understand.

[1079] Step 3:

[1080] The server analyzes the query using a generative AI model. Tokenized text data is input into the generative AI model, and the analysis results are obtained. The analysis results include information such as the query category and importance.

[1081] Step 4:

[1082] The server determines whether a reply is possible based on the analysis results. It also determines whether a reply can be made immediately based on the category and importance of the analysis results. If a reply is possible, the server generates an automatic response message and sends it to the user.

[1083] Step 5:

[1084] The server generates a ticket containing the necessary information based on the analysis results. The ticket contains information such as the inquiry content, analysis results, and importance. The generated ticket is saved in JSON format.

[1085] Step 6:

[1086] The server determines the severity of the ticket and escalates it if necessary. If the severity is high, the ticket is automatically sent to the appropriate department. If escalation is not required, the ticket is handled according to the normal response process.

[1087] Step 7:

[1088] The server sends the generated ticket to the PFOS system, which manages the received ticket and escalates it to the appropriate department. If the ticket submission is successful, the server sends a notification to the user.

[1089] Step 8:

[1090] The user receives a notification from the server on their device. The notification states that the inquiry has been successfully accepted and that the matter has been escalated to the appropriate department. The user can check the notification and make additional inquiries as necessary.

[1091] Example 3

[1092] Next, a description will be given of Example 3 of Form Example 3. 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."

[1093] In conventional inquiry response systems, the analysis of inquiry content and the determination of importance are often done manually, making it difficult to respond quickly and appropriately. Even with automated response systems, it is difficult to respond to complex inquiries or highly technical problems, and there are cases where secondary responses or escalations are not performed appropriately. This has led to problems such as lower user satisfaction and a worsening of response efficiency.

[1094] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[1095] In this invention, the server includes means for determining the content of the inquiry, means for responding to confirm receipt of the inquiry, means for issuing a ticket, means for determining whether a reply is possible, means for automatically responding if a secondary response is necessary, means for determining the importance of the inquiry and escalating as necessary, means for analyzing the content of the inquiry using a generative AI model, means for determining the category and difficulty of the inquiry based on the analysis results of the content of the inquiry, and means for using information such as the inquirer's information, the frequency of the inquiry, and the time period of the inquiry to determine the importance of the inquiry. This enables quick and accurate analysis of the content of the inquiry and appropriate secondary responses and escalation.

[1096] The "means for determining the content of an inquiry" is a function for analyzing the content of an inquiry sent by a user and classifying the content into a specific category or type of problem.

[1097] The "means for responding by confirming receipt of an inquiry" is a function for, when receiving an inquiry from a user, confirming receipt of the inquiry and sending a message confirming receipt to the user.

[1098] The "means of raising a ticket" is a function that assigns a unique identification number to received inquiries, making them traceable within the management system.

[1099] The "means for determining whether or not a response is possible" is a function for analyzing the content of an inquiry and determining whether or not the inquiry can be handled by the automated response system.

[1100] The "means for automatically responding when a secondary response is required" is a function for notifying the user that a secondary response is required for inquiries that are difficult to handle using the automatic response system.

[1101] "Means for determining the level of importance and escalating as necessary" is a function for evaluating the importance of the inquiry and handing over the response to an expert or higher-level support team as necessary.

[1102] "Means for analyzing inquiry content using a generative AI model" refers to a function that uses a generative AI model to analyze inquiry content through natural language processing and understand its content.

[1103] "Means for determining the category and difficulty of an inquiry based on the analysis results of the inquiry content" is a function for determining which category an inquiry belongs to and how difficult it is based on the analysis results of the generative AI model.

[1104] "Means of using information such as the inquirer's information, the frequency of inquiries, and the time period of inquiries to determine the importance of inquiries" refers to a function for comprehensively evaluating the importance of inquiries by taking into account additional information such as the inquirer's past inquiry history, the frequency of inquiries, and the time period of inquiries.

[1105] This invention relates to a system that analyzes the content of inquiries from users and automatically provides appropriate responses. Specifically, this system analyzes the content of inquiries, determines whether a response is possible, and provides secondary support or escalation as necessary. This system is realized using a server, a terminal, and a generative AI model.

[1106] Hardware and software used

[1107] Server: Receives inquiries, analyzes them, determines their importance, determines whether to proceed with a secondary response, and escalates them.

[1108] Device: The device used by the user to send the inquiry, such as a PC or smartphone.

[1109] Generative AI model: Used to analyze inquiry content and determine importance. Uses natural language processing technology.

[1110] Program processing

[1111] 1. User submits a query:

[1112] The user uses the terminal to send an inquiry, which may include a technical problem with the product or other questions.

[1113] 2. The server receives the query:

[1114] The server receives queries sent by users and stores them in a database.

[1115] 3. The server analyzes the request:

[1116] The server uses the generative AI model to analyze the inquiry, determining the category and difficulty of the inquiry as a result of the analysis.

[1117] 4. The server determines whether a secondary action is necessary:

[1118] If an automatic response is difficult, the server generates a message to the user indicating that a secondary response is required. For example, if the inquiry is about a highly technical problem with a specific product, the server notifies the user and tells them to wait for a response from an expert.

[1119] 5. The server determines the importance of the query:

[1120] The server uses a generative AI model to determine the importance of the inquiry based on the analysis of the inquiry content and other information (e.g., information about the person making the inquiry, the frequency of the inquiry, the time of the inquiry, etc.).

[1121] 6. Server escalates as needed:

[1122] For inquiries that are determined to be of high importance, the server immediately escalates them to an expert. For example, if an inquiry is about a serious defect in a large number of products, the server determines that the inquiry is of high importance and escalates it to an expert.

[1123] Specific examples

[1124] User asks: "An error occurred while installing new software. The error message is 'Error 404'."

[1125] Server Action:

[1126] 1. The server receives the query.

[1127] 2. Analyze the query using a generative AI model.

[1128] 3. The error message 'Error 404' is assumed to be a common problem and can be automatically answered.

[1129] 4. Generate an auto-reply message and send it to the user.

[1130] Prompt Sentence Examples

[1131] Analyze the user's inquiry and generate an automatic response message for the error message 'Error 404'.

[1132] This system enables prompt and appropriate responses to user inquiries. By using the generative AI model, the accuracy of analysis of the inquiry content is improved, and appropriate secondary responses and escalation are realized. The flow of the identification process in Example 3 will be described with reference to FIG. 15.

[1133] Step 1:

[1134] A user submits a query

[1135] Input: The user uses the terminal to enter their inquiry and clicks the send button.

[1136] Specific behavior: The user enters "An error occurred while installing new software. The error message is 'Error 404'" into the inquiry form and submits it.

[1137] Output: Query data is sent from the terminal to the server.

[1138] Step 2:

[1139] The server receives the query

[1140] Input: Query data sent from the terminal.

[1141] Specific operation: The server receives the query data and stores it in the database.

[1142] Output: Saved in the database as "User ID: 12345, Enquiry: 'An error occurred while installing new software. The error message was 'Error 404'."

[1143] Step 3:

[1144] The server analyzes the query

[1145] Input: Query data stored in the database.

[1146] Specific operation: The server uses the generative AI model to analyze the query content. The prompt sentence "An error occurred while installing new software. The error message is 'Error 404'" is input to the generative AI model.

[1147] Output: The analysis results in "General error, technical support required."

[1148] Step 4:

[1149] The server determines whether a secondary response is necessary.

[1150] Input: Analysis results from a generative AI model.

[1151] Specific operation: Based on the analysis results, if the server is unable to provide an automatic response, it will generate a message indicating that secondary action is required.

[1152] Output: A message is generated and sent to the user indicating that a secondary action is required.

[1153] Step 5:

[1154] The server determines the importance of the query

[1155] Input: Analysis results from the generative AI model, plus additional information such as enquirer information, enquiry frequency, and enquiry time period.

[1156] Specific operation: The server determines the importance of the query based on this information.

[1157] Output: Information about tickets that are determined to be of high importance.

[1158] Step 6:

[1159] The server will escalate as needed

[1160] Input: Information about the inquiry that was determined to be of high priority.

[1161] Specific Actions: The server immediately sends an escalation message to the expert.

[1162] Output: An escalation message is sent to the expert.

[1163] In this way, the system receives and analyzes inquiries from users, and provides secondary support or escalation as necessary, thereby achieving a prompt and appropriate response.

[1164] (Application example 3)

[1165] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[1166] In conventional inquiry response systems, the analysis of inquiry content and the determination of importance were often done manually, making it difficult to respond quickly and appropriately. Also, when the inquiry involved a highly technical issue, it could take a long time to escalate the inquiry to the appropriate expert, which was a factor in reducing user satisfaction. Furthermore, when it came to security-related inquiries, a quick response was required, and conventional systems often fell short.

[1167] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means. In this invention, the server includes means for determining the content of the inquiry, means for responding to confirm receipt of the inquiry, means for issuing a ticket, means for determining whether a reply is possible, means for automatically responding if a secondary response is necessary, means for determining the level of importance and escalating the inquiry as necessary, means for analyzing the content of the inquiry using natural language processing technology, means for determining the importance of the inquiry using an AI engine, means for generating an automatic response based on the content of the inquiry, and means for escalating the inquiry according to its importance. This enables quick and accurate analysis of the content of the inquiry, and automatic implementation of an appropriate response according to its importance, thereby improving user satisfaction.

[1168] The "means for determining the content of the inquiry" is a function for analyzing the content of the inquiry from the user and identifying that content.

[1169] The "means for confirming receipt of an inquiry and responding" is a function for confirming that an inquiry from a user has been received and responding to the inquiry.

[1170] "Means for raising tickets" is a function that generates tickets for response based on the content of an inquiry.

[1171] The "means for determining whether or not a reply is possible" is a function for determining whether or not an inquiry can be automatically responded to.

[1172] The "means for automatically responding when a secondary response is required" is a function for notifying the user that a secondary response is required when an automatic response is difficult.

[1173] The "means for determining the importance level and escalating as necessary" is a function for determining the importance level of an inquiry and escalating it to an expert as necessary.

[1174] "Means for analyzing the content of inquiries using natural language processing technology" refers to a function for analyzing the content of inquiries from users using natural language processing technology.

[1175] "Means for determining the importance of an inquiry using an AI engine" is a function for determining the importance of an inquiry using an AI engine.

[1176] The "means for generating an automatic response based on the content of an inquiry" is a function for automatically generating a response based on the content of an inquiry.

[1177] The "means for escalating an inquiry according to its importance" is a function for escalating an inquiry to an appropriate expert according to its importance.

[1178] A system for implementing this invention includes means for determining the content of an inquiry, means for responding to confirm receipt of the inquiry, means for creating a ticket, means for determining whether a response is possible, means for automatically responding if a secondary response is required, means for determining the level of importance and escalating as necessary, means for analyzing the content of the inquiry using natural language processing technology, means for determining the importance of the inquiry using an AI engine, means for generating an automatic response based on the content of the inquiry, and means for escalating the inquiry depending on its importance.

[1179] Hardware and software used

[1180] Hardware: Servers, smartphones, robots

[1181] Software: Natural language processing libraries (e.g., spaCy, NLTK), AI engines (e.g., TensorFlow, PyTorch), databases (e.g., MySQL)

[1182] Data processing and calculation

[1183] Query Analysis

[1184] The server receives the user's query in text format and uses a natural language processing library (e.g., spaCy, NLTK) to analyze the text and identify the content of the query.

[1185] Auto-reply

[1186] Based on the analysis results, the server automatically generates pre-prepared answers to simple queries, such as "Try restarting your security system. If that doesn't resolve the issue, you'll need to contact a professional."

[1187] Secondary response notification

[1188] For inquiries that are difficult to answer automatically, the server notifies the user that a secondary response is required. For example, it sends a notification saying, "This is a highly technical issue and requires expert assistance. Please wait a moment."

[1189] Severity determination and escalation

[1190] The server uses an AI engine (e.g., TensorFlow, PyTorch) to determine the importance of the inquiry. For example, it takes into account the content of the inquiry, information about the person making the inquiry, the frequency of inquiries, and the time of day. If the inquiry is determined to be of high importance, it immediately escalates it to an expert.

[1191] Examples and prompts

[1192] Specific examples

[1193] Inquiry: "My home security system has been malfunctioning frequently lately. What should I do?"

[1194] Automated response: "Please try rebooting your security system. If that doesn't work, you'll need professional assistance."

[1195] Secondary response notification: "This is a highly technical issue that requires expert assistance. Please wait a moment."

[1196] Severity assessment: "This request is about a critical issue and will be escalated to a specialist immediately."

[1197] Prompt Sentence Examples

[1198] User asks: "My home security system has been malfunctioning frequently lately. What should I do?"

[1199] Analysis result: "Security system malfunction"

[1200] Automated response: "Please try rebooting your security system. If that doesn't work, you'll need professional assistance."

[1201] Secondary response notification: "This is a highly technical issue that requires expert assistance. Please wait a moment."

[1202] Severity assessment: "This request is about a critical issue and will be escalated to a specialist immediately."

[1203] In this way, by specifically showing the form for implementing the invention, it becomes possible to quickly and accurately analyze the content of the inquiry, and appropriate responses are automatically made according to the importance, thereby improving user satisfaction.

[1204] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[1205] Step 1:

[1206] The user sends an inquiry. The user inputs the inquiry in text format using a smartphone or a robot and sends it to the server. The input data is the user's inquiry.

[1207] Step 2:

[1208] The server receives the query. The server receives the text data sent by the user and begins parsing it using a natural language processing library (e.g., spaCy, NLTK). The input data is the user's query, and the output data is the analysis result.

[1209] Step 3:

[1210] The server analyzes the query. The server uses a natural language processing library to analyze the text and identify the query. Specifically, it performs text tokenization, part-of-speech tagging, semantic analysis, etc. The input data is the user's query, and the output data is the analysis result.

[1211] Step 4:

[1212] The server generates an automatic response. Based on the analysis results, the server automatically generates a pre-prepared response to simple inquiries. For example, it generates a response such as, "Try restarting your security system. If that doesn't resolve the issue, you'll need to contact a specialist." The input data is the analysis results, and the output data is the automatic response.

[1213] Step 5:

[1214] The server sends a secondary response notification. For inquiries that are difficult to answer automatically, the server notifies the user that a secondary response is required. For example, it sends a notification saying, "This problem is a highly technical one and requires expert assistance. Please wait a moment." The input data is the analysis result, and the output data is the secondary response notification.

[1215] Step 6:

[1216] The server determines the importance. The server uses an AI engine (e.g., TensorFlow, PyTorch) to determine the importance of the query. For example, it takes into account the content of the query, information about the person making the query, the frequency of the query, and the time of day. The input data are the analysis results and information about the person making the query, and the output data is the importance determination result.

[1217] Step 7:

[1218] The server performs the escalation. If the server determines that the severity is high, it immediately escalates to an expert. For example, an inquiry about a serious defect is notified to an expert. The input data is the severity judgment result, and the output data is the escalation notification.

[1219] In this way, by performing specific operations at each step and performing data processing and calculations based on input data, it becomes possible to respond quickly and appropriately to user inquiries.

[1220] 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.

[1221] "Example 1"

[1222] In one embodiment of the present invention, an emotion engine that recognizes the user's emotions is used as a means for determining the content of the inquiry. This emotion engine can read emotions not only from the text information contained in the inquiry, but also from the user's tone of voice, facial expression, etc. For example, if the user is feeling angry or frustrated, this information is reflected in the analysis of the inquiry content, allowing for a more appropriate response.

[1223] "Example 2"

[1224] The emotion engine can also adjust the analysis results of the inquiry content based on the user's emotions. For example, if the user is angry, the importance of the inquiry can be increased taking into account the user's level of anger. This can improve user satisfaction.

[1225] "Example 3"

[1226] Furthermore, as a means of escalation, AI is used to determine the importance of an inquiry, taking into account the user's emotions. This AI determines the importance of an inquiry not only based on the analysis of the inquiry content, but also taking into account the user's emotions. For example, if the user is very angry, the inquiry will be determined to be of high importance and will be immediately escalated to an expert.

[1227] The processing flow of each embodiment will be described below.

[1228] "Example 1"

[1229] Step 1: A user query is entered into the system.

[1230] Step 2: The emotion engine reads the user's emotions from the text information contained in the inquiry, tone of voice, facial expressions, etc.

[1231] Step 3: The emotional information read by the emotion engine is reflected in the analysis of the inquiry content.

[1232] "Example 2"

[1233] Step 1: A user query is entered into the system.

[1234] Step 2: The emotion engine reads the user's emotions from the text information contained in the inquiry, tone of voice, facial expressions, etc.

[1235] Step 3: Adjust the analysis results of the inquiry content based on the emotional information read by the emotion engine.

[1236] "Example 3"

[1237] Step 1: A user query is entered into the system.

[1238] Step 2: The emotion engine reads the user's emotions from the text information contained in the inquiry, tone of voice, facial expressions, etc.

[1239] Step 3: The AI ​​determines the importance of the inquiry by taking into account the analysis results of the inquiry content and emotional information from the emotion engine.

[1240] Step 4: Based on the AI's findings, the inquiry is escalated to an expert.

[1241] Example 1

[1242] Next, a description will be given of Example 1 of Form 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."

[1243] In conventional inquiry response systems, the judgment of the inquiry content and the decision on the appropriate response are often done manually, which is inefficient. Also, it is difficult to respond in a way that takes the user's feelings into account, making it an issue to improve user satisfaction. Furthermore, there are cases where high-priority inquiries are not escalated appropriately, so a quick response is required.

[1244] 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.

[1245] In this invention, the server includes means for analyzing the content of the inquiry using a natural language processing engine, means for using an emotion engine that recognizes the user's emotions, and means for determining an appropriate response based on the analysis results, thereby enabling automatic analysis of the content of the inquiry and appropriate response that takes the user's emotions into consideration.

[1246] The "means for determining the content of the inquiry" is a means for analyzing the content of the inquiry sent from the user and identifying the content.

[1247] The "means for responding by confirming receipt of an inquiry" is a means for, when receiving an inquiry from a user, confirming receipt of the inquiry and responding to the user by confirming receipt.

[1248] The "means for raising a ticket" is a means for automatically creating a ticket for managing responses based on the content of an inquiry.

[1249] The "means for determining whether or not a reply can be made" is a means for automatically determining whether or not a reply can be made to the received inquiry.

[1250] The "means for automatically replying when a secondary response is required" is a means for automatically notifying the user when the inquiry cannot be resolved by the primary response.

[1251] The "means for determining the importance level and escalating as necessary" is a means for determining the importance level of the inquiry content and escalating it to a higher-level responder as necessary.

[1252] "Means for analyzing the content of an inquiry using a natural language processing engine" refers to means for analyzing the content of an inquiry using natural language processing technology and understanding the content.

[1253] The "means for using an emotion engine to recognize the user's emotion" is a means for using an emotion engine to recognize the user's emotion and analyzing the emotion contained in the inquiry content.

[1254] The "means for determining an appropriate response based on the analysis results" refers to a means for determining the optimal response method based on the analysis results of the natural language processing engine and the emotion engine.

[1255] The "means for returning the action content to the user" is a means for returning the determined action content to the user.

[1256] MODE FOR CARRYING OUT THE INVENTION

[1257] The system of the present invention automatically determines the content of an inquiry and provides an appropriate response. By combining a natural language processing engine and an emotion engine, this system can analyze the content of a user's inquiry and provide a response that takes the user's emotions into consideration.

[1258] Hardware and software used

[1259] The server uses the following hardware and software to implement the system.

[1260] Hardware: A high-performance server machine (e.g., a server equipped with a server processor)

[1261] software:

[1262] Natural language processing engine: A common natural language processing API (e.g., Google Cloud Natural Language API)

[1263] Emotion engine: A common emotion recognition API (e.g., Microsoft Azure Emotion API)

[1264] Database: A common database management system (e.g., MySQL)

[1265] System Operation

[1266] The user inputs and sends the inquiry using a terminal (for example, a smartphone or PC). The server receives the inquiry sent from the user and performs the following processes.

[1267] 1. Analysis of inquiry content:

[1268] The server analyzes the query using a natural language processing engine. The analysis includes text tokenization, part-of-speech tagging, sentence structure analysis, etc. For example, if a user asks, "How do I set up my new smartphone?", the natural language processing engine extracts the keyword "how to set up."

[1269] 2. Recognizing user emotions:

[1270] The server uses an emotion engine to recognize the user's emotions. The emotion engine can read emotions not only from the text information contained in the inquiry, but also from the user's tone of voice and facial expression. For example, if a user makes an inquiry in an angry tone, saying, "This product is completely unusable. Please do something about it," the emotion engine will recognize the user's anger.

[1271] 3. Determining the appropriate response:

[1272] The server determines the appropriate response based on the analysis results of the natural language processing engine and emotion engine. For example, if the inquiry is about how to operate a product and the user is angry, the server will respond quickly and courteously.

[1273] 4. Reply with the response:

[1274] The server then replies to the user with the determined response. The reply is sent in text format and displayed on the user's device. For example, the server might reply with detailed instructions such as, "To set up your new smartphone: 1. Open the settings menu. 2. Select Wi-Fi. 3. Enter your password."

[1275] Prompt Sentence Examples

[1276] An example of a prompt sentence to input to the generative AI model is as follows:

[1277] "When a user asks how to set up their new smartphone, how do you respond?"

[1278] "How do you respond when users are angry with your product?"

[1279] Using these prompts, the generative AI model can suggest appropriate responses.

[1280] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1281] Step 1:

[1282] The user submits a query.

[1283] The user uses a terminal (e.g., a smartphone or PC) to enter the inquiry content and press the send button. The entered inquiry content is sent to the server in text format.

[1284] Input: The inquiry typed by the user (e.g., "How do I set up my new smartphone?")

[1285] Output: The query sent to the server

[1286] Step 2:

[1287] The server receives the query.

[1288] The server receives the inquiry sent by the user and stores it in temporary storage.

[1289] Input: Inquiry sent by the user

[1290] Output: Query content saved in temporary storage

[1291] Step 3:

[1292] The server analyzes the query using a natural language processing engine.

[1293] The server sends the saved query content to a natural language processing engine (e.g., a common natural language processing API) for analysis, which includes text tokenization, part-of-speech tagging, and sentence structure analysis.

[1294] Input: Query content saved in temporary storage

[1295] Output: Analysis results (e.g., keyword "setting method")

[1296] Step 4:

[1297] The server recognizes the user's emotions using an emotion engine.

[1298] The server sends the user's tone of voice and facial expression data along with the content of the query to an emotion engine (e.g., a general emotion recognition API) to recognize the user's emotions. The emotion engine determines emotions such as anger, joy, and sadness.

[1299] Input: Enquiry, user tone of voice and facial expression data

[1300] Output: Emotion recognition result (e.g., "anger")

[1301] Step 5:

[1302] The server determines the appropriate response based on the analysis results.

[1303] The server determines the appropriate response based on the analysis results of the natural language processing engine and emotion engine. For example, if the inquiry is about how to use a product and the user is angry, the server will respond quickly and courteously.

[1304] Input: Analysis results of the natural language processing engine, emotion recognition results of the emotion engine

[1305] Output: The action decided (e.g., "How to set up a new smartphone")

[1306] Step 6:

[1307] The server responds by replying to the user.

[1308] The server then returns the determined response to the user in text format and displays it on the user's device.

[1309] Input: Decided action

[1310] Output: The reply sent to the user (e.g., "To set up your new phone: 1. Open the Settings menu. 2. Select Wi-Fi. 3. Enter your password.")

[1311] (Application example 1)

[1312] Next, a description will be given of Application Example 1 of Form 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."

[1313] Conventional inquiry response systems do not take the user's feelings into consideration when responding, which can result in delays in providing an appropriate response to urgent inquiries. Furthermore, natural language processing technology is not fully utilized when analyzing the content of inquiries, making it difficult to provide an accurate response. This leads to issues such as a decrease in user satisfaction and a loss of service quality.

[1314] 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.

[1315] In this invention, the server includes means for determining the content of the inquiry, means for responding by confirming receipt of the inquiry, means for issuing a ticket, means for determining whether a reply is possible, means for automatically responding if a secondary response is required, means for determining the importance of the inquiry and escalating it as necessary, means for recognizing the user's emotions, and means for responding to an emergency based on the emotions, thereby enabling a prompt and appropriate response that takes the user's emotions into consideration.

[1316] The "means for determining the content of an inquiry" is a technique for analyzing an inquiry from a user, understanding its content, and determining an appropriate response.

[1317] The "means for responding to and confirming receipt of an inquiry" is a technique for confirming that an inquiry from a user has been received and notifying the user of that fact.

[1318] "Means for raising tickets" refers to technology that automatically creates tickets to record the details of inquiries and manage the progress of responses.

[1319] The "means for determining whether a reply is possible" is a technique for determining whether the system can automatically reply to the inquiry.

[1320] "Means for automatically responding when a secondary response is required" is a technology that automatically notifies the user when the inquiry is complex and cannot be resolved by a primary response.

[1321] The "means for determining the importance level and escalating as necessary" is a technique for evaluating the importance level of the inquiry content and escalating it to a higher-level responder as necessary.

[1322] "Means for recognizing user emotions" refers to technology that reads emotions from the user's tone of voice, facial expressions, etc., and reflects that information in the analysis.

[1323] The "means for taking emergency measures based on emotions" is a technique for taking an appropriate response immediately when the user's emotions indicate an emergency.

[1324] A system for implementing this invention includes means for determining the content of an inquiry, means for responding by confirming receipt of the inquiry, means for creating a ticket, means for determining whether a reply can be made, means for automatically responding if a secondary response is required, means for determining the importance of the inquiry and escalating the inquiry as necessary, means for recognizing the user's emotions, and means for taking emergency action based on the emotions.

[1325] System configuration

[1326] This system uses the following hardware and software:

[1327] Hardware: Smartphone, microphone

[1328] Software: Python, speech_recognition library, transformers library, emotion_recognition library

[1329] Processing flow

[1330] 1. Acquiring voice input: The user speaks into the smartphone microphone, asking a question such as, "I've been seeing some suspicious people around my house lately. What should I do?"

[1331] 2. Speech Recognition: The smartphone uses the speech_recognition library to convert speech to text.

[1332] 3. Emotion Recognition: The smartphone uses the emotion_recognition library to recognize the user's emotions from their voice. For example, if the user is feeling anxious or scared, it will recognize that emotion.

[1333] 4. Parsing the query: The server uses the natural language processing engine in the transformers library to parse the query and determine the appropriate response.

[1334] 5. Response execution: If the emotion is "anger" or "fear," the server will take emergency action by contacting a security company. Otherwise, it will provide an appropriate response based on the analysis results.

[1335] Specific examples

[1336] When a user inquires, "I've seen some suspicious people around my house recently," the smartphone recognizes the voice and emotion of the inquiry. The server uses a natural language processing engine to analyze the inquiry, and if it determines that the situation is urgent, it immediately contacts the security company. This allows for a prompt and appropriate response that takes the user's emotions into account.

[1337] Prompt Sentence Examples

[1338] User: I've been seeing some suspicious people around my house lately, what should I do?

[1339] This system enables prompt and appropriate responses that take into account the user's feelings, which is expected to improve service quality and user satisfaction.

[1340] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1341] Step 1:

[1342] The user makes a query into the microphone of the smartphone.

[1343] Input: User's voice

[1344] Output: Audio data

[1345] Specific behavior: A user asks a question such as, "I've been seeing some suspicious people around my house lately. What should I do?"

[1346] Step 2:

[1347] The device uses the speech_recognition library to convert speech to text.

[1348] Input: Audio data

[1349] Output: Text data

[1350] Specific operation: The smartphone passes the recorded voice data to the speech_recognition library for speech recognition, which converts the voice into text.

[1351] Step 3:

[1352] The device uses the emotion_recognition library to recognize the user's emotions from their voice.

[1353] Input: Audio data

[1354] Output: Emotion data (e.g., anxiety, fear)

[1355] Specific operation: The smartphone passes the recorded voice data to the emotion_recognition library for emotion recognition. As a result, the user's emotions are recognized as "anxiety" or "fear."

[1356] Step 4:

[1357] The server uses the natural language processing engine in the transformers library to analyze the query and determine the appropriate response.

[1358] Input: Text data, emotion data

[1359] Output: Response (e.g., urgent response required, general response)

[1360] How it works: The server passes the text data to the transformers library, analyzes the query, and considers the emotional data to determine if an urgent response is required.

[1361] Step 5:

[1362] If the server detects the emotion as "anger" or "fear," it will take emergency action by contacting a security company. Otherwise, it will provide an appropriate response based on the analysis results.

[1363] Input: Response content, emotion data

[1364] Output: Perform emergency response or provide general answer

[1365] Specific operation: If the emotion data is "anger" or "fear," the server calls an API to contact the security company. Otherwise, it provides the user with an appropriate answer based on the analysis results.

[1366] Example 2

[1367] Next, a description will be given of Example 2 of Form 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."

[1368] Conventional inquiry response systems process inquiries without considering the user's feelings, which can lead to low user satisfaction. Furthermore, analyzing the inquiry content and generating tickets takes time, making it difficult to respond quickly. Furthermore, the importance of inquiries is not properly determined, and urgent inquiries are sometimes not escalated appropriately.

[1369] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for determining the content of the inquiry, a means for responding to confirm receipt of the inquiry, a means for issuing a ticket, a means for determining whether a reply is possible, a means for automatically responding if a secondary response is required, a means for determining the importance of the inquiry and escalating it as necessary, a means for analyzing the user's emotions, and a means for adjusting the content of the ticket based on the analysis results. This enables a prompt and appropriate response to the inquiry taking the user's emotions into consideration.

[1370] The "means for determining the content of the inquiry" is a means for analyzing the content of the inquiry received from the user and identifying the content.

[1371] The "means for responding by confirming receipt of an inquiry" is a means for, when receiving an inquiry from a user, confirming receipt of the inquiry and responding to the user by confirming receipt.

[1372] The "means for raising a ticket" is a means for automatically generating a ticket including information that requires a response based on the analyzed inquiry content.

[1373] The "means for determining whether or not a reply can be made" is a means for determining whether or not a reply can be made immediately to the received inquiry.

[1374] The "means for automatically responding when a secondary response is required" is a means for automatically notifying the user that a secondary response is required when an immediate response cannot be given to the inquiry.

[1375] The "means for determining the importance and escalating as necessary" is a means for determining the importance of the inquiry content and escalating it to a higher-level department as necessary.

[1376] The "means for analyzing user emotions" is a means for analyzing emotions from the content of a user's inquiry and evaluating the degree of those emotions.

[1377] The "means for adjusting the content of the ticket based on the analysis results" refers to a means for adjusting the content of the generated ticket based on the results of the user's emotion analysis, so that an appropriate response can be made.

[1378] This invention analyzes the content of inquiries, automatically generates tickets containing necessary information, and sends them to the system. It also analyzes user sentiment and can adjust the importance of inquiries based on the results.

[1379] Hardware and software used

[1380] 1. Terminal

[1381] A device that allows a user to input a query, such as a PC, smartphone, or tablet.

[1382] 2. Server

[1383] This is a computer system for analyzing inquiry content, analyzing emotions, generating tickets, and sending tickets. The following software is installed on the server:

[1384] Generative AI model: A model for natural language processing, such as OpenAI's GPT-4.

[1385] Sentiment analysis API: An API for analyzing user emotions. For example, Microsoft Azure's Text Analytics.

[1386] Data processing and calculation

[1387] 1. Receipt of inquiries

[1388] The terminal receives an inquiry from the user and transmits the contents of the inquiry to the server.

[1389] 2. Analysis of inquiry content

[1390] The server uses the generated AI model to analyze the received inquiry, specifically extracting the product name and defect details from the inquiry.

[1391] 3. Emotion analysis

[1392] The server uses the emotion analysis API to analyze the emotion from the user's inquiry, and evaluates the type and degree of the user's emotion as a result of the analysis.

[1393] 4. Ticket Creation

[1394] Based on the analysis results, the server generates a ticket containing the necessary information, such as product information, the details of the defect, and the results of user sentiment analysis.

[1395] 5. Submit a ticket

[1396] The server generates a ticket and sends it to the system for escalation.

[1397] Specific examples

[1398] A user submits a query

[1399] The user enters "Product B's screen is not displayed" through a web form or application and presses the submit button.

[1400] The device receives the query

[1401] The terminal receives the user's input and sends the query to the server.

[1402] The server analyzes the query

[1403] The server uses the generated AI model to analyze the inquiry, "The screen of Product B is not displayed." As a result of the analysis, the product name "Product B" and the problem description "The screen is not displayed" are extracted.

[1404] The server performs sentiment analysis

[1405] The server uses the emotion analysis API to detect the emotion of "dissatisfaction" from the user's inquiry. The emotion analysis result is evaluated as "the user is very dissatisfied."

[1406] The server generates a ticket

[1407] Based on the analysis results, the server generates a ticket that includes the content "Product B's screen is not displayed" and the sentiment analysis result "The user is very dissatisfied."

[1408] The server sends the ticket

[1409] The server generates a ticket and sends it to the system for escalation.

[1410] Prompt Sentence Examples

[1411] User inquiry: "The battery of product C runs out quickly."

[1412] Sentiment analysis results: "Users are frustrated"

[1413] Generate a ticket with the following content: "Product C's battery runs out quickly. Users are frustrated and need immediate attention."

[1414] In this way, the system can analyze the user's inquiry content and emotions, generate an appropriate ticket, and send it to the system, thereby improving user satisfaction.

[1415] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1416] Step 1:

[1417] The user submits a query.

[1418] Input: The user enters the inquiry details through a web form or application. For example, they enter "Product A won't turn on" and press the submit button.

[1419] Output: The query is sent to the terminal.

[1420] Specific operation: The user enters the inquiry content and presses the send button, which sends the inquiry to the terminal.

[1421] Step 2:

[1422] The terminal receives the inquiry.

[1423] Input: The inquiry sent by the user.

[1424] Output: The query is sent to the server.

[1425] Specific operation: The terminal receives the user's input and sends it to the server. For example, the terminal sends the query "Product A won't turn on" to the server.

[1426] Step 3:

[1427] The server analyzes the query.

[1428] Input: The inquiry sent from the device.

[1429] Output: Analyzed inquiry details (product name, defect details, etc.).

[1430] Specific operation: The server uses the generated AI model to analyze the received inquiry. For example, if the server analyzes the inquiry "Product A won't turn on," it will extract the product name "Product A" and the problem description "Won't turn on."

[1431] Step 4:

[1432] The server analyzes the user's emotions using an emotion engine.

[1433] Input: Parsed query content.

[1434] Output: User emotion analysis results (type and degree of emotion).

[1435] Specific operation: The server uses the emotion analysis API to analyze the emotion from the user's inquiry. For example, the server detects the emotion "anger" from the inquiry "Product A won't turn on" and evaluates the degree of that emotion.

[1436] Step 5:

[1437] The server generates a ticket.

[1438] Input: Analyzed inquiry content and user sentiment analysis results.

[1439] Output: Generated ticket (product information, defect details, sentiment analysis results, etc.).

[1440] Specific operation: The server generates a ticket containing the necessary information based on the analysis results. For example, the server generates a ticket containing the content "Product A won't turn on" and the sentiment analysis result "The user is very angry."

[1441] Step 6:

[1442] The server sends the ticket to the system.

[1443] Input: The generated ticket.

[1444] Output: A ticket is submitted to the system and escalated to the appropriate department.

[1445] Specific operation: The ticket generated by the server is sent to the system and escalated to the corresponding department. For example, the ticket generated by the server is sent to the system and the corresponding department responds promptly.

[1446] (Application example 2)

[1447] Next, a description will be given of Application Example 2 of Form 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."

[1448] Conventional inquiry response systems process inquiries without considering the user's feelings, which can lead to lower user satisfaction. They also fail to properly prioritize inquiries, which can lead to important inquiries being put off. This makes it difficult to respond quickly and appropriately.

[1449] 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.

[1450] In this invention, the server includes a means for determining the content of the inquiry, a means for responding to the inquiry and confirming receipt, a means for issuing a PFOS ticket, a means for determining whether a reply is possible, a means for automatically responding if a secondary response is required, a means for determining the importance of the inquiry and escalating it as necessary, a means for analyzing the user's emotions, and a means for adjusting the priority of the inquiry based on the emotion analysis results, thereby enabling a prompt and appropriate response that takes the user's emotions into consideration.

[1451] The "means for determining the content of an inquiry" is a means for analyzing the content of an inquiry from a user and identifying the content.

[1452] The "means for responding by confirming receipt of an inquiry" is a means for confirming that an inquiry from a user has been received and notifying the user of that fact.

[1453] "Means for raising a PFOS ticket" refers to a means for automatically generating a ticket containing the necessary information based on the inquiry content and sending it to the PFOS system.

[1454] The "means for determining whether or not a reply can be made" is a means for determining whether or not an inquiry can be immediately answered.

[1455] The "means for automatically replying when a secondary response is required" is a means for automatically notifying the user when an immediate reply cannot be made to the inquiry.

[1456] The "means for determining the importance and escalating as necessary" is a means for determining the importance of the inquiry content and escalating it to a higher-level department as necessary.

[1457] The "means for analyzing the user's emotions" is a means for analyzing emotions from the content of the user's inquiry and identifying those emotions.

[1458] The "means for adjusting the priority of an inquiry based on the result of emotion analysis" is a means for adjusting the priority of an inquiry based on the result of emotion analysis of a user.

[1459] To implement the present invention, the following system configuration and program are required.

[1460] System Configuration

[1461] The server includes a means for determining the content of the inquiry, a means for responding by confirming receipt of the inquiry, a means for raising a PFOS ticket, a means for determining whether a response is possible, a means for automatically responding if a secondary response is required, a means for determining the importance of the inquiry and escalating it as necessary, a means for analyzing the user's emotions, and a means for adjusting the priority of the inquiry based on the results of the emotion analysis.

[1462] Program processing explanation

[1463] The server first receives the user's inquiry. Next, it uses natural language processing to analyze the inquiry and identify its content. Based on the analysis results, it determines whether a reply is possible, and if so, responds immediately. If a reply is difficult, it automatically notifies the user that a secondary response is required.

[1464] In addition, the server uses emotion analysis software such as EmotionRecognizer to analyze the user's emotions. Based on the emotion analysis results, the server adjusts the priority of the inquiry and escalates it if necessary. For example, if the user is angry, the server sets the priority of the inquiry high and responds quickly.

[1465] Hardware and software used

[1466] Hardware: Servers, user devices (smartphones, etc.)

[1467] Software: Python, TextBlob, EmotionRecognizer, Requests

[1468] Specific examples

[1469] If a user says, "It seems like my account has been compromised recently. I'm very worried," the server receives this inquiry and analyzes it using natural language processing. From the analysis results, it identifies the user's emotion as "anxiety," and sentiment analysis detects a negative trend. Based on this, a high-priority ticket is generated and sent to the security team.

[1470] Prompt Sentence Examples

[1471] Please enter your security query: I believe my account has been compromised recently and I'm very concerned.

[1472] In this way, a quick and appropriate response that takes into account the user's feelings is possible.

[1473] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1474] Step 1:

[1475] The user inputs the inquiry content in text format using a user device (such as a smartphone). The input text is sent to the server.

[1476] Step 2:

[1477] The server receives the query. The server receives the text data sent by the user and saves it for further processing. The input is the user's query, and the output is the saved text data.

[1478] Step 3:

[1479] The server analyzes the query using natural language processing. The server uses a natural language processing library such as TextBlob to analyze the text data of the query. As a result of the analysis, the subject of the query and related keywords are extracted. The input is the saved text data, and the output is the analysis result.

[1480] Step 4:

[1481] The server determines whether a reply is possible for the inquiry. Based on the analysis results, the server determines whether a reply is possible immediately. For example, it checks the FAQ database and provides an immediate reply if the question is known. The input is the analysis results, and the output is the result of the reply possibility judgment.

[1482] Step 5:

[1483] If the server is able to reply, it sends a reply to the user. If the server determines that the reply is possible, it generates an appropriate reply and sends it to the user terminal. The input is the reply possibility determination result, and the output is a reply message to the user.

[1484] Step 6:

[1485] If the server cannot respond, it notifies the user that a secondary response is required. If the server determines that it cannot respond, it automatically notifies the user. The input is the result of the determination, and the output is a notification message to the user.

[1486] Step 7:

[1487] The server analyzes the user's emotions. The server uses emotion analysis software such as EmotionRecognizer to analyze the user's emotions from the text data of the inquiry. The input is the saved text data, and the output is the emotion analysis result.

[1488] Step 8:

[1489] The server adjusts the priority of the query based on the emotion analysis result. The server sets the priority of the query based on the emotion analysis result. For example, if the user is angry, the priority is set high. The input is the emotion analysis result, and the output is the priority setting result.

[1490] Step 9:

[1491] The server raises a PFOS ticket. The server automatically generates a ticket containing the necessary information based on the inquiry content, sentiment analysis results, and priority setting results, and sends it to the PFOS system. The input is the inquiry content, sentiment analysis results, and priority setting results, and the output is the generated ticket.

[1492] Step 10:

[1493] The server determines the importance of the ticket and escalates it as necessary.The server determines the importance of the generated ticket and escalates it to a higher-level department as necessary.The input is the generated ticket, and the output is the result of the escalation.

[1494] Example 3

[1495] Next, a description will be given of Example 3 of Form Example 3. 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."

[1496] In conventional inquiry response systems, the analysis of inquiry content and the determination of importance were often done manually, which resulted in a lack of speed and accuracy in response. Also, when the inquiry content included a highly technical issue or when a response that took the user's feelings into consideration was required, the appropriate escalation was not performed. This resulted in problems such as a decrease in user satisfaction and delayed response.

[1497] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[1498] In this invention, the server includes means for determining the content of the inquiry, means for responding by confirming receipt of the inquiry, means for saving the content of the inquiry, means for sending the content of the inquiry to an AI engine for analysis, means for determining whether a reply is possible, means for automatically replying if a secondary response is required, and means for determining the importance of the inquiry and escalating the inquiry as necessary. This enables quick and accurate analysis of the content of the inquiry and appropriate escalation.

[1499] The "means for determining the content of the inquiry" is a function for analyzing the content of the inquiry sent from the user and identifying the content.

[1500] The "means for responding by confirming receipt of an inquiry" is a function for, when receiving an inquiry from a user, confirming receipt of the inquiry and responding to the user by confirming receipt.

[1501] The "means for saving the inquiry content" is a function for saving the received inquiry content in a storage device such as a database.

[1502] "Means for sending the inquiry content to an AI engine for analysis" is a function for sending the received inquiry content to an AI engine and requesting analysis.

[1503] The "means for determining whether a response is possible" is a function for determining whether the inquiry can be resolved by an automatic response.

[1504] The "means for automatically responding when a secondary response is required" is a function for automatically notifying the user that a secondary response is required when the inquiry contains a highly technical issue, for example.

[1505] The "means for determining the importance level and escalating as necessary" is a function for determining the importance level of the inquiry content and escalating to an expert as necessary.

[1506] This invention is a system that analyzes the content of inquiries and provides secondary support or escalation as necessary. Specifically, if an inquiry contains a highly technical issue or is judged to be of high importance, a message urging an expert to handle the issue is automatically generated and sent to the inquirer.

[1507] Hardware and software used

[1508] Server: A central processing unit for analyzing query content and generating messages. For example, a virtual server in a cloud service is used.

[1509] AI engine: Analyzes the content of inquiries and determines their importance, for example, using a cloud-based AI platform.

[1510] Database: Stores query details and user information. For example, a relational database is used.

[1511] Specific operation of the system

[1512] 1. User submits a query:

[1513] A user fills out a contact form on their device and hits the submit button. For example, a user types, "After the new software update, my app keeps crashing. How can I fix this?"

[1514] 2. The server receives the query:

[1515] The server receives the inquiry sent by the user. Specifically, it receives the HTTP request and analyzes its contents.

[1516] 3. The server saves the query in a database:

[1517] The server saves the received inquiry in a database. For example, information such as the "inquiry ID," "user ID," "inquiry content," and "received date and time" is saved in a relational database.

[1518] 4. The server sends the query to the AI ​​engine:

[1519] The server sends the query to the AI ​​engine. Specifically, it converts the query into JSON format and sends a POST request to the AI ​​engine's API endpoint.

[1520] 5. The AI ​​engine analyzes the inquiry:

[1521] The AI ​​engine analyzes the inquiry. For example, it uses natural language processing (NLP) technology to extract keywords from the inquiry and determine the technical difficulty and importance. As a result of the analysis, it generates information such as "advanced technical issues" and "user emotions (anger)."

[1522] 6. The server generates a message based on the analysis:

[1523] The server receives the analysis results from the AI ​​engine and generates a message indicating that a secondary response is required if necessary, such as "This inquiry involves a highly technical issue and requires the assistance of an expert."

[1524] 7. The server sends a message to the user:

[1525] The server generates a message and sends it to the user, either by sending it to the user's email address or by displaying it on the inquiry form screen.

[1526] 8. Server escalates based on severity:

[1527] If the server determines that the inquiry is highly important based on the AI ​​engine's judgment, it immediately escalates it to an expert. For example, if the user is very angry, it will determine that the inquiry is highly important and notify the expert. Specifically, it will send a notification email to the expert's email address or display an alert on a dedicated management screen.

[1528] Specific examples

[1529] Question: "After the new software update, the app keeps crashing. How can I fix this?"

[1530] Analysis results of the AI ​​engine: The inquiry was determined to contain a highly technical issue. The user's emotion was determined to be "anger."

[1531] Produces the following message: "This request is highly technical and requires expert assistance. Please wait a moment."

[1532] Escalation: Send an email to the specialist notifying them that "the user is very upset and requires immediate attention."

[1533] Prompt Sentence Examples

[1534] "My app keeps crashing after the new software update. How can I fix this?"

[1535] The purpose of this system is to provide prompt and accurate support by efficiently analyzing the content of the user's inquiry and taking appropriate action as necessary. The flow of the identification process in the third embodiment will be described with reference to FIG.

[1536] Step 1:

[1537] The user submits a query.

[1538] A user fills out an inquiry form on their device and presses the submit button. As input, the user types "My app keeps crashing after a new software update. How can I fix this?". As output, the inquiry is sent to the server.

[1539] Step 2:

[1540] The server receives the query.

[1541] The server receives the query sent by the user. Specifically, it receives an HTTP request and analyzes its contents. As input, it receives an HTTP request containing the query from the user. As output, the query is analyzed and passed to the next process.

[1542] Step 3:

[1543] The server stores the query in a database.

[1544] The server saves the received inquiry content in a database. Specifically, information such as "inquiry ID," "user ID," "inquiry content," and "received date and time" is saved in a relational database. The parsed inquiry content is provided as input. The inquiry content is saved in the database as output.

[1545] Step 4:

[1546] The server sends the query to the AI ​​engine.

[1547] The server sends the query to the AI ​​engine. Specifically, it converts the query into JSON format and sends a POST request to the AI ​​engine's API endpoint. The query stored in the database is provided as input. The query is sent to the AI ​​engine as output.

[1548] Step 5:

[1549] The AI ​​engine analyzes the inquiry content.

[1550] The AI ​​engine analyzes the inquiry. For example, it uses natural language processing (NLP) technology to extract keywords from the inquiry and determine its technical difficulty and importance. The inquiry sent from the server is provided as input. The output is analysis results, such as "highly technical problem" and "user emotion (anger)."

[1551] Step 6:

[1552] The server generates a message based on the analysis results.

[1553] The server receives the analysis results from the AI ​​engine and generates a message indicating that secondary action is required if necessary. For example, it generates a message such as, "This inquiry involves a highly technical issue and requires expert assistance." The analysis results from the AI ​​engine are provided as input. The message to be sent to the user is generated as output.

[1554] Step 7:

[1555] The server sends a message to the user.

[1556] The server generates a message and sends it to the user. Specifically, it sends the message to the user's email address or displays it on the contact form screen. The generated message is provided as input. The message is sent to the user as output.

[1557] Step 8:

[1558] The server will escalate based on the severity.

[1559] If the server determines that the inquiry is highly important based on the AI ​​engine's judgment, it immediately escalates it to an expert. For example, if the user is very angry, it will judge the inquiry to be of high importance and notify the expert. Specifically, it will send a notification email to the expert's email address or display an alert on a dedicated management screen. The importance judgment result from the AI ​​engine is provided as input. The output is a notification to the expert.

[1560] (Application example 3)

[1561] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[1562] Conventional inquiry response systems have difficulty accurately determining the importance of inquiries and escalating them appropriately. Furthermore, they are unable to respond in a way that takes the user's feelings into account, which can lead to a decline in user satisfaction. Furthermore, while quick and appropriate responses are required when reporting security incidents, current systems have difficulty achieving this.

[1563] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[1564] In this invention, the server includes a means for determining the content of the inquiry, a means for responding to the inquiry and confirming receipt, a means for issuing a PFOS ticket, a means for determining whether the content is a responseable item, a means for automatically responding if a secondary response is required, a means for determining the severity and escalating the issue as necessary, a means for analyzing the user's emotions to determine the severity, a means for accepting incident reports, a means for analyzing the report content using natural language processing, a means for determining the severity based on the analysis results, a means for escalating the issue to an expert if the severity is high, and a means for notifying the user if a secondary response is required. This enables the server to accurately determine the severity of the inquiry and appropriately escalate the issue. Furthermore, since the server can respond in a way that takes the user's emotions into consideration, improved user satisfaction can be expected. Furthermore, a prompt and appropriate response to security incident reports is also possible.

[1565] The "means for determining the content of an inquiry" is a function for analyzing the content of an inquiry from a user and identifying that content.

[1566] The "means for confirming receipt of an inquiry and responding" is a function for confirming that an inquiry from a user has been received and responding to the inquiry.

[1567] "Means to raise a PFOS ticket" is a function that generates a ticket to resolve a problem based on the inquiry content.

[1568] The "means for determining whether or not a reply is possible" is a function for determining whether or not an automatic reply can be made to the inquiry content.

[1569] The "means for automatically replying when a secondary response is required" is a function for notifying the user that a secondary response is required when an automatic response cannot be made to the inquiry content.

[1570] The "means for determining the importance level and escalating as necessary" is a function for determining the importance level of the inquiry content and escalating it to a higher-level person in charge as necessary.

[1571] The "means for analyzing the user's emotions and determining the importance" is a function for analyzing the user's emotions and determining the importance of the inquiry based on the results.

[1572] The "means for receiving incident reports" is a function for receiving security incident reports from users.

[1573] The "means for analyzing the report content using natural language processing" is a function for analyzing the report content from the user using natural language processing technology.

[1574] The "means for determining the importance level based on the analysis results" is a function for determining the importance level of the report content based on the results of analysis using natural language processing.

[1575] The "means for escalating to an expert when the importance is high" is a function for escalating to an expert when the importance of the report content is determined to be high.

[1576] The "means for notifying the user when a secondary response is required" is a function for notifying the user when a secondary response is required.

[1577] The system for implementing the present invention includes a plurality of means for determining the content of an inquiry, assessing its importance, and escalating it as necessary. A specific embodiment of this system will be described below.

[1578] Hardware and software used

[1579] Hardware: Servers, smartphones, smart glasses

[1580] Software: Python, TensorFlow, natural language processing libraries (e.g., spaCy), sentiment analysis libraries

[1581] System configuration

[1582] 1. How to determine the content of the inquiry:

[1583] The server receives the user's query and analyzes it using a natural language processing library (spaCy). Through this analysis, the content of the query is identified and classified into the appropriate category.

[1584] 2. How to acknowledge receipt of inquiries:

[1585] The server notifies the user that the query has been received, and the notification is sent to the smartphone or smart glasses as a push notification.

[1586] 3. How to file a PFOS ticket:

[1587] The server generates a ticket for problem resolution based on the analysis results, which is then managed in an internal system and assigned to a responsible person.

[1588] 4. How to determine if a response is possible:

[1589] The server determines whether the query can be answered automatically, and if so, generates an appropriate answer and sends it to the user.

[1590] 5. Automated response when secondary action is required:

[1591] If the server cannot automatically respond to the inquiry, it notifies the user that a secondary response is required.

[1592] 6. How to determine severity and escalate as needed:

[1593] The server determines the importance of the inquiry and escalates it to a higher-level person if necessary. The importance is determined using a TensorFlow model.

[1594] 7. Analyzing user sentiment and determining importance:

[1595] The server analyzes the user's emotions using an emotion analysis library and determines the importance of the inquiry based on the results.

[1596] 8. How to report an incident:

[1597] The server accepts reports of security incidents from users, who report them via their smartphones or smart glasses.

[1598] 9. Methods for analyzing report content using natural language processing:

[1599] The server analyzes the report content using a natural language processing library (spaCy) to identify the content of the incident.

[1600] 10. How to determine importance based on analysis results:

[1601] The server determines the importance of the report content based on the results of analysis using natural language processing.

[1602] 11. How to escalate to an expert if the issue is of high importance:

[1603] If the server determines that the report content is of high importance, it escalates it to an expert.

[1604] 12. How to notify users when a secondary action is required:

[1605] If a secondary response is required, the server notifies the user. The notification is sent as a push notification to a smartphone or smart glasses.

[1606] Specific examples

[1607] For example, if a user reports that "the server may have been hacked," the server will input the following prompt sentence into the generative AI model:

[1608] Example prompt sentence:

[1609] A user reports that "the server may have been hacked." Analyze the report and determine its severity. Consider the user's feelings and escalate the issue as necessary.

[1610] Based on this prompt, the server analyzes the report, determines its importance, and escalates it to an expert if necessary, enabling a prompt and appropriate response.

[1611] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[1612] Step 1:

[1613] A user uses a smartphone or smart glasses to report an inquiry or security incident. The input is text data entered by the user, and the output is the report data sent to the server.

[1614] Step 2:

[1615] The server analyzes the received report data using a natural language processing library (spaCy). The input is the report data sent by the user, and the output is structured data as the analysis result. Specifically, the report content is tokenized and important keywords and phrases are extracted.

[1616] Step 3:

[1617] Based on the analysis results, the server identifies the category of the inquiry and creates a PFOS ticket. The input is the structured data of the analysis results, and the output is the generated ticket data. Specifically, the server selects the appropriate category based on the analysis results and creates a new ticket in the ticket management system.

[1618] Step 4:

[1619] The server determines whether the query can be automatically answered. The input is the structured data of the analysis result, and the output is a flag indicating whether an automatic response is possible. Specifically, the analysis result is evaluated using a rule-based system, and if an automatic response is possible, an appropriate answer is generated.

[1620] Step 5:

[1621] The server notifies the user if a secondary response is required. The input is a flag indicating whether an automatic response is possible, and the output is a notification message to the user. Specifically, if an automatic response is not possible, the server sends a push notification to the user informing them that a secondary response is required.

[1622] Step 6:

[1623] The server determines the importance of the inquiry and escalates it as necessary. The input is the structured data of the analysis results, and the output is the importance evaluation result and escalation instructions. Specifically, the server uses a TensorFlow model to evaluate the importance, and if it is high, it escalates to an expert.

[1624] Step 7:

[1625] The server analyzes the user's emotions using an emotion analysis library and determines the importance of the inquiry based on the results. The input is the report data from the user, and the output is the emotion analysis results and the importance evaluation results. Specifically, the report content is evaluated using the emotion analysis library, and if the emotion is strong, the importance is set high.

[1626] Step 8:

[1627] The server accepts incident reports and analyzes the report contents using natural language processing. The input is the incident report data from the user, and the output is structured data as the analysis result. Specifically, the report contents are tokenized and important keywords and phrases are extracted.

[1628] Step 9:

[1629] The server determines the importance of the incident based on the analysis results, and escalates it to an expert if the importance is high. The input is structured data from the analysis results, and the output is the importance assessment result and escalation instructions. Specifically, the importance of the incident is assessed based on the results of analysis using natural language processing, and if the importance is high, it is escalated to an expert.

[1630] Step 10:

[1631] The server notifies the user when a secondary response is required. The input is the importance evaluation result, and the output is a notification message to the user. Specifically, if the importance is high and a secondary response is required, the server sends a push notification to the user.

[1632] 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.

[1633] 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> ) 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.

[1634] Another example of generative AI is Gemini (internet search engine). <url: https: gemini.google.com ?hl="ja">) are mentioned.

[1635] 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.

[1636] [Third embodiment]

[1637] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1638] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[1639] 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).

[1640] 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.

[1641] 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.

[1642] 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).

[1643] 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.

[1644] 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.

[1645] 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.

[1646] 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.

[1647] 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.

[1648] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[1649] "Example 1"

[1650] The system of the present invention uses a natural language processing engine as a means for automatically determining the content of the inquiry. This engine analyzes the content of the inquiry and determines an appropriate response based on that content. For example, if the inquiry is about how to operate a product, it searches a database for related information and automatically responds with information about that operation method.

[1651] "Example 2"

[1652] In addition, as a means of automatically raising PFOS tickets, a ticket containing the necessary information is generated based on the inquiry content and its analysis results and sent to the PFOS system. For example, if the inquiry is about a defect in a specific product, information about the product and the defect content is entered in the ticket and it is escalated to the corresponding department.

[1653] "Example 3"

[1654] Furthermore, as a means of automatically responding to inquiries that require secondary support when they are difficult to answer automatically, a message indicating that secondary support is required is generated and sent to the inquirer based on the analysis results of the inquiry content. For example, if the inquiry is about a highly technical problem with a specific product, the inquirer is notified of this and told to wait for a response from an expert.

[1655] "Example 4"

[1656] Finally, an AI engine is used to determine the importance of the inquiry and escalate it as necessary. This engine determines the importance of the inquiry based on the analysis of the inquiry content and other information (such as the inquiry's information, frequency of inquiries, and time of day the inquiry is made), and escalates the inquiry as necessary. For example, if the inquiry is about a serious defect in a large number of products, the engine will determine the importance of the inquiry and immediately escalate it to an expert.

[1657] The processing flow of each embodiment will be described below.

[1658] "Example 1"

[1659] Step 1: The inquiry is entered into the system.

[1660] Step 2: The natural language processing engine analyzes the query, extracting the main points and key points of the query.

[1661] Step 3: Based on the analysis results, an appropriate response is determined. For example, if the inquiry is about how to operate a product, relevant information is searched for in the database and an answer with information on how to operate the product is automatically provided.

[1662] "Example 2"

[1663] Step 1: The inquiry and its analysis results are entered into the system.

[1664] Step 2: A ticket containing the required information is generated and sent to the PFOS system.

[1665] Step 3: If the inquiry is about a specific product issue, enter the product information and the details of the issue in the ticket and escalate it to the appropriate department.

[1666] "Example 3"

[1667] Step 1: The analysis results of the inquiry are entered into the system.

[1668] Step 2: For inquiries that are difficult to answer automatically, a message is generated indicating that a secondary response is required.

[1669] Step 3: The generated message is sent to the inquirer. For example, if the inquiry is about a highly technical issue with a specific product, the inquirer is notified of this and advised to wait for a response from an expert.

[1670] "Example 4"

[1671] Step 1: The analysis results of the inquiry content and other information (e.g., information about the person making the inquiry, frequency of inquiries, time of day of inquiries, etc.) are input into the AI ​​engine.

[1672] Step 2: The AI ​​engine uses this information to determine the importance of the inquiry.

[1673] Step 3: Escalate the inquiry as needed. For example, if the inquiry is about a major defect in a large number of products, the inquiry will be deemed to be of high importance and escalated immediately to an expert.

[1674] Example 1

[1675] Next, a description will be given of Example 1 of Form 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."

[1676] A problem with conventional inquiry response systems is that it takes a lot of time and effort to assess the content of an inquiry and determine the appropriate response. Furthermore, because the analysis of the inquiry content and database search are done manually, it is inefficient and often unable to provide a prompt response to the user. Furthermore, determining the importance and escalation of an inquiry depended on human judgment, which can lead to a lack of consistency in responses.

[1677] 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.

[1678] In this invention, the server includes a means for passing the query content to a natural language processing engine, a means for the natural language processing engine to analyze the query content, and a means for searching a database based on the analysis results, thereby enabling automatic analysis of the query content and determination of an appropriate response.

[1679] The "means for determining the content of the inquiry" is a means for analyzing the content of the inquiry input by the user and understanding the content.

[1680] The "means for responding by confirming receipt of an inquiry" is a means for confirming that an inquiry from a user has been received and notifying the user of that fact.

[1681] A "means for raising a ticket" is a means for automatically creating a ticket for response based on the content of an inquiry.

[1682] The "means for determining whether or not a reply can be made" is a means for determining whether or not a reply can be made automatically within the system to the inquiry content.

[1683] The "means for automatically replying when a secondary response is required" is a means for notifying the user that a secondary response is required when an automatic reply cannot be made to the inquiry content.

[1684] The "means for determining the importance level and escalating as necessary" is a means for determining the importance level of the inquiry content and escalating it to a higher-level responder as necessary.

[1685] The "means for passing the inquiry content to the natural language processing engine" is a means for inputting the inquiry content received from the user into the natural language processing engine.

[1686] "Means for a natural language processing engine to analyze the content of a query" refers to means for analyzing the content of a query using a natural language processing engine and understanding its intent.

[1687] The "means for searching a database based on the analysis results" refers to a means for searching a database for related information based on the analysis results of the natural language processing engine.

[1688] The "means for returning search results to the user" is a means for returning information obtained from the database to the user.

[1689] This invention relates to a system that automatically determines the content of an inquiry and provides an appropriate response. Specifically, this system analyzes the content of the inquiry using a natural language processing engine, searches a database based on the analysis results, and returns appropriate information to the user.

[1690] System configuration

[1691] The system consists of the following main components:

[1692] 1. User terminal: A device used by a user to input inquiries. This includes smartphones, PCs, tablets, etc.

[1693] 2. Server: Receives the query, passes it to the natural language processing engine, searches the database based on the analysis results, and responds to the user.

[1694] 3. Natural language processing engine: Software for analyzing query content, specifically using generative AI models such as BERT or GPT-3.

[1695] 4. Database: Data storage for product operation instructions and other related information.

[1696] Program processing

[1697] A user inputs a query through a device. For example, "Please tell me how to take a screenshot on my new smartphone." The device sends this query to a server. The server then passes the received query to a natural language processing engine. The natural language processing engine analyzes the query and determines an appropriate response based on the content.

[1698] As a result of the analysis, it is determined that the inquiry is about "how to operate a smartphone." The server then searches for information about "how to operate a smartphone" in a database. The database contains information about how to operate a product, and the server extracts the relevant information. The server then returns the extracted information to the user.

[1699] Specific examples

[1700] As a concrete example, consider the case where a user makes a query such as, "Please tell me how to take a screenshot on my new smartphone." This query is sent from the device to the server. The server analyzes the query using a natural language processing engine (e.g., GPT-3). As a result of the analysis, it is determined that the query is about "how to operate a smartphone."

[1701] Next, the server searches the database for information on "how to operate a smartphone" and extracts information on "how to take a screenshot." For example, it finds information such as "Press the power button and volume down button simultaneously to take a screenshot." The server then returns this information to the user.

[1702] Prompt Sentence Examples

[1703] Below are some example prompts to input to a generative AI model:

[1704] User Asks: "How do I take a screenshot on my new smartphone?"

[1705] Prompt: "A user is asking how to take a screenshot on their new phone. Please provide the appropriate instructions."

[1706] Using this prompt, the generative AI model generates information to provide appropriate operating instructions.

[1707] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1708] Step 1:

[1709] The user enters a query.

[1710] The user enters a query in text form through the device's interface, for example, "How do I take a screenshot on my new smartphone?" The text is then stored in the device's memory.

[1711] Step 2:

[1712] The terminal sends a query to the server.

[1713] The terminal sends the query entered by the user to the server as an HTTP POST request. The query is sent in JSON format. The input is the user's query text, and the output is an HTTP request to the server.

[1714] Step 3:

[1715] The server passes the query to a natural language processing engine.

[1716] The server parses the received JSON data and extracts the query content. The extracted query content is then input to the generative AI model as a prompt. For example, the prompt might be, "A user is asking how to take a screenshot on their new smartphone. Please provide the appropriate instructions." The input is JSON data, and the output is a prompt.

[1717] Step 4:

[1718] A natural language processing engine analyzes the inquiry.

[1719] A natural language processing engine (e.g., GPT-3) analyzes the prompt and understands the intent of the query. As a result of the analysis, it determines that information on "how to take a screenshot" is needed. The input is the prompt, and the output is the analysis result.

[1720] Step 5:

[1721] The server searches the database based on the analysis results.

[1722] The server searches the database for information on "how to take a screenshot" based on the analysis results. It executes a database query to retrieve the relevant information. The input is the analysis results, and the output is the information retrieved from the database.

[1723] Step 6:

[1724] The server returns the search results to the user.

[1725] The server returns the obtained information to the user. For example, it generates a message such as "Press the power button and volume down button simultaneously to take a screenshot" and sends it to the device as an HTTP response. The user can check this information on the device screen. The input is the information obtained from the database, and the output is the response message to the user.

[1726] (Application example 1)

[1727] Next, a description will be given of Application Example 1 of Form 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."

[1728] In conventional inquiry response systems, analyzing the content of inquiries and generating appropriate responses was often done manually, resulting in a long response time. It was also sometimes difficult to properly determine the importance of the inquiry and escalate it as necessary. This led to issues such as a decline in customer satisfaction and an impediment to efficient operations.

[1729] 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.

[1730] In this invention, the server includes means for determining the content of the inquiry, means for responding to confirm receipt of the inquiry, means for issuing an information management ticket, means for determining whether a reply is possible, means for automatically responding if a secondary response is required, means for determining the importance of the inquiry and escalating the inquiry as necessary, means for analyzing the content of the inquiry and generating an appropriate response, means for displaying the generated response to the user, and means for saving the content of the inquiry and the generated response in a database. This enables the content of the inquiry to be analyzed quickly and accurately and responses to be provided, thereby improving customer satisfaction and streamlining operations.

[1731] The "means for determining the content of the inquiry" is a means for analyzing the content of the inquiry from the user, understanding the content, and determining an appropriate response.

[1732] The "means for responding to and confirming receipt of an inquiry" refers to a means for confirming that an inquiry from a user has been received and notifying the user of that fact.

[1733] The "means for issuing an information management ticket" is a means for creating an information management ticket based on the content of an inquiry and for tracking and managing the inquiry.

[1734] The "means for determining whether or not a reply can be made" is a means for automatically determining whether or not a reply can be made to the inquiry content.

[1735] "Means for automatically responding when a secondary response is required" refers to a means for automatically notifying the user that a secondary response is required when an automatic response cannot be made to the inquiry.

[1736] The "means for determining the importance level and escalating as necessary" is a means for determining the importance level of the inquiry content and escalating it to a higher-level responder as necessary.

[1737] The "means for analyzing the content of an inquiry and generating an appropriate response" is a means for analyzing the content of an inquiry using natural language processing technology and generating an appropriate response based on that content.

[1738] The "means for displaying the generated answer to the user" refers to a means for displaying the generated answer to the user.

[1739] The "means for saving the inquiry content and the generated response in a database" refers to a means for saving the inquiry content and the generated response in a database so that they can be referenced later.

[1740] As an embodiment of the present invention, the following system can be constructed.

[1741] System configuration

[1742] 1. Hardware:

[1743] Server: Analyzes inquiries, generates answers, and manages the database.

[1744] Terminal: The device (e.g., smartphone) through which a user enters a query and receives a response.

[1745] 2. Software:

[1746] Natural language processing engine: Used to analyze the query and generate appropriate answers (e.g., OpenAI API).

[1747] Database: Used to store queries and generated answers (e.g., SQLite).

[1748] Inquiry management system: Confirms receipt of inquiries, raises information management tickets, and manages escalations.

[1749] Processing flow

[1750] 1. Receiving inquiries:

[1751] The user inputs the inquiry details using a terminal and sends them to the server, which receives the inquiry details and responds by confirming receipt.

[1752] 2. Analysis of inquiry content:

[1753] The server uses a natural language processing engine to analyze the inquiry and determines whether a response is possible based on the analysis results.

[1754] 3. Generate answers:

[1755] If a response is possible, the server uses a natural language processing engine to generate an appropriate answer, which is then displayed to the user.

[1756] 4. Data storage:

[1757] The queries and generated responses are stored in a database for future reference.

[1758] 5. Escalation Management:

[1759] If it is difficult to respond or if the inquiry is of high importance, the server will determine the importance of the inquiry and escalate it as necessary.

[1760] Specific examples

[1761] For example, if a user asks, "The item I ordered hasn't arrived yet. What should I do?", the server will process the following:

[1762] 1. The server receives the inquiry and responds by confirming receipt.

[1763] 2. Use a natural language processing engine to analyze the inquiry and determine whether a response is possible.

[1764] 3. If it is determined that a response is possible, the server generates a response saying, "Can you tell me your order number? I'll check." and displays it to the user.

[1765] 4. The query and generated answer are stored in a database.

[1766] 5. If it is difficult to respond or the request is of high importance, the server will determine the importance of the request and escalate it as necessary.

[1767] Prompt Sentence Examples

[1768] An example of a prompt for the generative AI model is as follows:

[1769] User Question: I haven't received my order, what should I do?

[1770] Please provide an appropriate answer.

[1771] In this way, it becomes possible to analyze inquiries quickly and accurately and provide answers, thereby improving customer satisfaction and streamlining operations.

[1772] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1773] Step 1:

[1774] The user uses a terminal to input the inquiry content and sends it to the server. The input is the user's inquiry content, and the output is the inquiry content sent to the server. Specifically, the user enters "The product I ordered hasn't arrived. What should I do?" into the inquiry form on the smartphone app and presses the send button.

[1775] Step 2:

[1776] The server receives the inquiry and responds by confirming receipt. The input is the inquiry sent by the user, and the output is a response message confirming receipt. Specifically, the server generates a message saying "Your inquiry has been received. Please wait a moment." and sends it to the user.

[1777] Step 3:

[1778] The server uses a natural language processing engine to analyze the inquiry. The input is the inquiry sent by the user, and the output is the analysis result. Specifically, the server calls the OpenAI API and analyzes the inquiry, "The item I ordered hasn't arrived yet. What should I do?"

[1779] Step 4:

[1780] The server determines whether a reply is possible based on the analysis results. The input is the analysis results, and the output is the result of determining whether a reply is possible. Specifically, the server evaluates the analysis results and determines whether a reply is possible.

[1781] Step 5:

[1782] If it is determined that a response is possible, the server uses a natural language processing engine to generate an appropriate response. The input is the analysis result, and the output is the generated response. Specifically, the server calls the OpenAI API again and generates the response, "Can you tell me your order number? I'll check."

[1783] Step 6:

[1784] The server displays the generated answer to the user. The input is the generated answer, and the output is the answer displayed to the user. Specifically, the server sends and displays the message "Can you tell me your order number? We will check it for you" to the user's smartphone.

[1785] Step 7:

[1786] The server stores the query and the generated response in a database. The input is the query and the generated response, and the output is the record stored in the database. Specifically, the server stores the query "My order hasn't arrived yet. What should I do?" and the response "Can you tell me your order number? I'll check." in an SQLite database.

[1787] Step 8:

[1788] If it is difficult to respond or if the inquiry is of high importance, the server determines the importance of the inquiry and escalates it as necessary. The input is the analysis results and importance determination criteria, and the output is an instruction for escalation. Specifically, the server uses an AI model to evaluate the importance of the inquiry and escalates it to a higher-level responder as necessary.

[1789] Example 2

[1790] Next, a description will be given of Example 2 of Form 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."

[1791] In conventional inquiry response systems, analyzing the content of inquiries, opening tickets, and deciding whether to escalate were often done manually, which was a time-consuming and labor-intensive process. In addition, the accuracy of analyzing the content of inquiries was low, which sometimes resulted in delays in escalating to the appropriate department. This led to issues such as lower user satisfaction and a decline in the company's response efficiency.

[1792] 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.

[1793] In this invention, the server includes a means for determining the content of the inquiry, a means for responding to the inquiry and confirming receipt, a means for analyzing the content of the inquiry, a means for extracting necessary information, a means for issuing a PFOS ticket, a means for determining whether a reply is possible, a means for automatically responding if a secondary response is required, and a means for determining the importance of the inquiry and escalating the inquiry as necessary. This enables the content of the inquiry to be analyzed quickly and accurately and escalated to the appropriate department.

[1794] The "means for determining the content of an inquiry" is a function for analyzing the content of an inquiry entered by a user and classifying the content into a specific category or topic.

[1795] The "means for responding to and confirming receipt of an inquiry" is a function for confirming that an inquiry from a user has been received and notifying the user of that fact.

[1796] The "means for analyzing the content of the inquiry" is a function that uses natural language processing technology to analyze the content of the inquiry from the user in detail and extract the necessary information.

[1797] The "means for extracting necessary information" is a function that extracts necessary information such as the product name, details of the defect, and the department in charge from the analyzed inquiry content.

[1798] "Means for raising a PFOS ticket" is a function that automatically generates a ticket containing the inquiry details based on the extracted information.

[1799] The "means for determining whether a reply is possible" is a function for determining whether a reply can be made immediately to the content of the generated ticket.

[1800] The "means for automatically replying when a secondary response is required" is a function that automatically notifies the user that a secondary response is required when an immediate reply cannot be made.

[1801] The "means for determining the importance and escalating as necessary" is a function for determining the importance of the inquiry content and escalating it to an appropriate department as necessary.

[1802] MODE FOR CARRYING OUT THE INVENTION

[1803] This invention is a system that automatically analyzes the content of inquiries, generates tickets containing necessary information, and escalates them to the appropriate department. This system consists of three main elements: users, terminals, and servers.

[1804] 1. Enter your inquiry

[1805] A user uses an interface to input the content of their inquiry. This interface can be accessed through a web browser or a dedicated application. For example, a user accesses a dedicated inquiry form in a web browser and inputs, "The screen of my newly purchased smartphone frequently freezes."

[1806] 2. Submitting your inquiry

[1807] The device sends the query entered by the user to the server as JSON format data.

[1808] 3. Analysis of inquiry content

[1809] The server analyzes the received query using natural language processing (NLP) techniques. For example, a generative AI model such as OpenAI's GPT-4 is used for this analysis. Specifically, the server inputs the following prompt sentence into the generative AI model:

[1810] Please analyze the following inquiry and extract the necessary information. Enquiry: The screen of my newly purchased smartphone frequently freezes.

[1811] 4. Extracting ticket information

[1812] The server extracts necessary information based on the analysis results from the generative AI model. For example, it extracts the following information from the analysis results:

[1813] Product Name: Smartphone

[1814] Issue details: Frequent screen freezes

[1815] Department in charge: Technical Support Department

[1816] 5. Ticket Creation

[1817] The server automatically generates a PFOS ticket based on the extracted information, which includes the inquiry content, product information, details of the problem, and the department in charge of handling the problem.

[1818] 6. Submit a ticket

[1819] The server sends the generated ticket to the PFOS system, which then escalates the ticket to the technical support department, where a technical support representative reviews the ticket and begins working on it.

[1820] This system enables prompt and accurate analysis of inquiries and escalation to the appropriate department, thereby improving user satisfaction and the company's response efficiency.

[1821] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1822] Step 1:

[1823] The user uses an interface to input the details of their inquiry. For example, the user accesses a dedicated inquiry form in a web browser and enters, "The screen of my newly purchased smartphone frequently freezes." The inputted inquiry details are sent to the system.

[1824] Step 2:

[1825] The terminal sends the query entered by the user to the server. At this time, the terminal sends the query to the server as JSON format data. The input is the user's query, and the output is JSON format data.

[1826] Step 3:

[1827] The server analyzes the received query using natural language processing (NLP) techniques. For example, a generative AI model such as OpenAI's GPT-4 is used for this analysis. Specifically, the server inputs the following prompt sentence into the generative AI model:

[1828] Please analyze the following inquiry and extract the necessary information. Enquiry: The screen of my newly purchased smartphone frequently freezes.

[1829] The input is the query content in JSON format, and the output is the analysis result.

[1830] Step 4:

[1831] The server extracts necessary information based on the analysis results from the generative AI model. For example, it extracts the following information from the analysis results:

[1832] Product Name: Smartphone

[1833] Issue details: Frequent screen freezes

[1834] Department in charge: Technical Support Department

[1835] The input is the analysis result of the generative AI model, and the output is the extracted information.

[1836] Step 5:

[1837] The server automatically generates a PFOS ticket based on the extracted information. The generated ticket includes the inquiry content, product information, details of the defect, the department in charge, etc. The input is the extracted information, and the output is the generated PFOS ticket.

[1838] Step 6:

[1839] The server sends the generated ticket to the PFOS system, which then escalates the ticket to the technical support department. For example, a technical support representative reviews the ticket and initiates a workaround. The input is the generated PFOS ticket, and the output is the escalated ticket.

[1840] (Application example 2)

[1841] Next, a description will be given of Application Example 2 of Form 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."

[1842] In conventional inquiry response systems, the analysis of inquiry content, ticket creation, and escalation processes are often done manually, which is inefficient and increases the likelihood of response delays and errors. Furthermore, manual handling is required for reporting security incidents, making it difficult to respond quickly.

[1843] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for determining the content of the inquiry, means for responding to confirm receipt of the inquiry, means for issuing a PFOS ticket, means for determining whether a response is possible, means for automatically responding if a secondary response is required, means for determining the severity and escalating the incident as necessary, means for receiving a security incident report, means for analyzing the incident content using a generative AI model, and means for generating and sending a ticket based on the analysis results. This automates the analysis and response of the inquiry content and security incident, enabling rapid and accurate responses.

[1844] The "means for determining the content of the inquiry" is a function for analyzing the content of the inquiry from the user and identifying that content.

[1845] The "means for confirming receipt of an inquiry and responding" is a function for confirming that an inquiry from a user has been received and responding to the inquiry.

[1846] "Means to raise a PFOS ticket" is a function that automatically generates a ticket containing the necessary information based on the inquiry content.

[1847] The "means for determining whether a reply is possible" is a function for determining whether a reply can be made immediately to the inquiry.

[1848] The "means for automatically replying when a secondary response is required" is a function for automatically replying when a secondary response is required for the content of an inquiry.

[1849] The "means for determining the importance and escalating as necessary" is a function for determining the importance of the inquiry content and escalating it to a higher-level department as necessary.

[1850] "Means for receiving reports of security incidents" refers to a function for receiving reports of security incidents from users.

[1851] "Means for analyzing incident details using a generative AI model" refers to a function for analyzing the details of a security incident using a generative AI model.

[1852] The "means for generating and sending a ticket based on the analysis result" is a function for generating a ticket based on the analysis result and sending it to the corresponding department.

[1853] The following system configuration will be described as an embodiment of the present invention.

[1854] System Configuration

[1855] Hardware

[1856] Server: Use a server with a powerful processor and sufficient memory. The server analyzes inquiries, generates tickets, and handles escalations.

[1857] Devices: Users use devices such as smartphones and computers to report inquiries and security incidents.

[1858] software

[1859] Generative AI models: Use generative AI models for natural language processing.

[1860] Inquiry management system: Software for receiving inquiries, analyzing them, and generating tickets.

[1861] PFOS system: A system for managing generated tickets and escalating them to the appropriate department.

[1862] Processing flow

[1863] 1. Receiving an Inquiry:

[1864] A user uses the device to report an inquiry or security incident.

[1865] The inquiry is sent from the terminal to the server.

[1866] 2. Analysis of inquiry content:

[1867] The server uses a generative AI model to analyze the query.

[1868] Based on the analysis results, it is determined whether or not a response can be made to the inquiry.

[1869] 3. Ticket generation:

[1870] The server generates a ticket containing the necessary information based on the analysis results.

[1871] A ticket contains information such as the inquiry content, analysis results, and importance.

[1872] 4. Escalation:

[1873] The server determines the importance of the ticket and escalates it if necessary.

[1874] If escalation is required, the ticket will be sent to the appropriate department.

[1875] Specific examples

[1876] Incident Report Example:

[1877] A user reports that "The firewall has detected unauthorized access."

[1878] The server uses a generative AI model to analyze the report and determine its importance.

[1879] Based on the analysis results, a ticket is generated and sent to the appropriate department.

[1880] Example prompt sentence:

[1881] "The firewall has detected unauthorized access."

[1882] "The server is overloaded."

[1883] "A user made an invalid login attempt."

[1884] In this way, the analysis and response to inquiries and security incidents is automated, enabling quick and accurate responses.

[1885] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1886] Step 1:

[1887] A user uses a terminal to report an inquiry or security incident. The user enters the inquiry or incident details into the input form on the terminal and presses the send button. The input data is sent to the server in text format.

[1888] Step 2:

[1889] The server receives the query sent from the device. The received data is in text format. The server preprocesses this data to input it into the generative AI model. Specifically, the server tokenizes the text data and converts it into a format that the model can understand.

[1890] Step 3:

[1891] The server analyzes the query using a generative AI model. Tokenized text data is input into the generative AI model, and the analysis results are obtained. The analysis results include information such as the query category and importance.

[1892] Step 4:

[1893] The server determines whether a reply is possible based on the analysis results. It also determines whether a reply can be made immediately based on the category and importance of the analysis results. If a reply is possible, the server generates an automatic response message and sends it to the user.

[1894] Step 5:

[1895] The server generates a ticket containing the necessary information based on the analysis results. The ticket contains information such as the inquiry content, analysis results, and importance. The generated ticket is saved in JSON format.

[1896] Step 6:

[1897] The server determines the severity of the ticket and escalates it if necessary. If the severity is high, the ticket is automatically sent to the appropriate department. If escalation is not required, the ticket is handled according to the normal response process.

[1898] Step 7:

[1899] The server sends the generated ticket to the PFOS system,...

Claims

1. A means for determining the content of the inquiry; A means for issuing a ticket based on the content of the inquiry; means for generating and replying to the inquiry content if the inquiry content is a replyable content; A means for responding to a request for a secondary response to the inquiry; A means for analyzing user emotions from the content of the inquiry; A means for adjusting the priority of the ticket based on the analyzed user's sentiment; A system including:

2. a means for escalating the inquiry depending on the importance of the inquiry, The system of claim 1 .

3. The adjusting means adjusts the priority using a generative AI model based on a prompt sentence corresponding to the user's emotion. The system of claim 1 .

Images