System

Abstract

A system is provided.SOLUTION: A system comprising: means for receiving a query; means for parsing the received query; means for generating an answer based on the parsed query; means for transmitting the generated answer; means for storing the query and the generated answer in a database; and means for analyzing the stored data.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] In today's business environment, it is necessary to respond to inquiries from many customers and business partners, and the same inquiries are often repeated. This often results in wasted time and effort. Furthermore, if the content of inquiries is not recorded or analyzed properly, it is difficult to provide useful information that will lead to improved customer satisfaction. There is a need for a system that can solve these problems, improve the efficiency of inquiry responses, and provide useful information by utilizing accumulated data. [Means for solving the problem]

[0005] The present invention solves the above-mentioned problems by providing a system including a means for receiving an inquiry, a means for analyzing the received inquiry, a means for generating a response based on the analyzed inquiry, a means for transmitting the generated response, a means for storing the inquiry and the generated response in a database, and a means for analyzing the stored data. This system automates and efficiently handles inquiries. Furthermore, accurate responses are generated through analysis using natural language processing (NLP). Furthermore, by analyzing the stored data, it becomes possible to provide more useful information to customers in the future.

[0006] The "means for receiving an inquiry" is a function or mechanism that allows the system to receive the contents of an inquiry from a user.

[0007] "Means for analyzing the content of the received inquiry" refers to the process or function for interpreting the content of the inquiry and understanding its meaning and intent.

[0008] The "means for generating an answer based on the analyzed inquiry content" is a function for automatically generating an appropriate answer that corresponds to the meaning and intent of the analyzed inquiry.

[0009] The "means for transmitting the generated answer" refers to a communication means or function for communicating the generated answer to the user.

[0010] The "means for saving the inquiry content and the generated response to a database" is a function for recording the inquiry content and the response to the inquiry and storing it in a database in a format that can be referenced in the future.

[0011] "Means for analyzing stored data" refers to the processes or functions used to analyze inquiries and responses stored in a database and extract trends and patterns. [Brief explanation of the drawings]

[0012] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0013] 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.

[0014] First, the terms used in the following description will be explained.

[0015] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0016] 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.

[0017] 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.

[0018] 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.

[0019] 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."

[0020] [First embodiment]

[0021] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0022] 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.

[0023] 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).

[0024] 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.

[0025] 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.

[0026] 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.

[0027] 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.

[0028] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0029] 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.

[0030] 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.

[0031] 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.

[0032] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0033] The system of the present invention responds quickly and efficiently by automatically receiving and analyzing inquiries from users and generating and returning appropriate answers. This system is primarily composed of a server, terminals, and users. The following describes the processing of the system program and a specific example.

[0034] System Configuration

[0035] This system consists of the following main components:

[0036] 1. Means of receiving inquiries

[0037] 2. Means of analyzing the content of received inquiries

[0038] 3. A means of generating answers based on the parsed query content

[0039] 4. A means of sending the generated answer

[0040] 5. A means of storing queries and generated responses in a database

[0041] 6. Means of analyzing stored data

[0042] Program processing

[0043] How to receive inquiries

[0044] The user inputs a query using their own terminal and presses the send button, which sends the query to the server.

[0045] The terminal sends the entered inquiry to the server via API.

[0046] A means of analyzing the content of received inquiries

[0047] The server temporarily stores the inquiry received from the terminal in a database.

[0048] The server passes the received query content to a natural language processing (NLP) model to analyze the content.

[0049] A means of generating answers based on the parsed query content

[0050] The server determines the intent of the query based on the analysis results of the NLP model.

[0051] For example, if a user inquires, "Please tell me the stock status of the product," the server determines that the user's intent is "to check stock."

[0052] A means of sending the generated answer

[0053] The server retrieves the relevant information from the database based on the determined intent.

[0054] The server generates a response based on the acquired information and sends it to the user's terminal.

[0055] The terminal displays the received response to the user.

[0056] A means of storing queries and generated responses in a database

[0057] The server permanently stores the final answer together with the inquiry details in a database.

[0058] A means of analyzing stored data

[0059] The server periodically analyzes the accumulated inquiry and response data to extract inquiry trends and patterns.

[0060] Based on the analysis results, we will build a system that predicts future inquiries and provides relevant information in advance.

[0061] Specific examples

[0062] Product inquiries

[0063] User: The user enters "Is product A in stock?" and presses the send button.

[0064] Terminal: The terminal sends this query to the server.

[0065] Server: The server temporarily stores the query content in a database and passes it to the natural language processing model.

[0066] Server: The NLP model determines the intent to "check stock," and the server retrieves stock information for product A from the database.

[0067] Server: Generates a response saying "Currently, there are 15 units of product A in stock" and sends it to the user's device.

[0068] Terminal: The terminal displays this answer to the user.

[0069] Server: The final query and answer are stored in a database for later data analysis.

[0070] Delivery inquiries

[0071] User: The user enters "Please tell me the delivery status of product B" and submits.

[0072] Terminal: The terminal sends this query to the server.

[0073] Server: The server temporarily stores the query content in a database and passes it to the natural language processing model.

[0074] Server: The NLP model determines the intent to "check delivery status," and the server retrieves the delivery status of Product B from the delivery system database.

[0075] Server: Generates a response stating, "Item B is currently being delivered. The estimated delivery date is next Tuesday," and sends it to the user's device.

[0076] Terminal: The terminal displays this answer to the user.

[0077] Server: The final query and answer are stored in a database for later data analysis.

[0078] This allows the system to automate inquiries efficiently and accurately, and the accumulated data can be used to improve customer service in the future.

[0079] The processing flow will be explained below.

[0080] Step 1:

[0081] User: Enter the inquiry details and click the send button.

[0082] For example, enter "Please let me know the product availability."

[0083] Step 2:

[0084] Terminal: Sends the entered inquiry to the server.

[0085] Specifically, the inquiry data is sent via the API using an HTTP POST request.

[0086] Step 3:

[0087] Server: Temporarily stores the received inquiry in a database.

[0088] Specifically, the query content is recorded in the query table using an SQL insert statement.

[0089] INSERT INTO inquiries (user_id, inquiry_text) VALUES ('1234', 'What is the product availability?')

[0090] Step 4:

[0091] Server: Passes the received query content to a natural language processing (NLP) model for analysis.

[0092] Specifically, the query is passed to the NLP model API and the analysis results are received in JSON format.

[0093] Example: "Product availability" is parsed as "Intent: Check inventory."

[0094] Step 5:

[0095] Server: Determines the intent of the query based on the analysis results of the NLP model.

[0096] Based on the analysis results, the corresponding processing is selected.

[0097] Example: Determine the intent "check inventory."

[0098] Step 6:

[0099] Server: Based on the intent of the query, retrieves the necessary information from the database.

[0100] Specifically, an SQL query is executed to obtain the stock quantity of the relevant product from the inventory table.

[0101] SELECT stock_quantity FROM products WHERE product_name = 'A product'

[0102] Step 7:

[0103] Server: Generates an answer based on the information obtained.

[0104] For example, construct an answer such as "Currently, there are 15 units of product A in stock."

[0105] Step 8:

[0106] Server: Sends the generated answer to the user's device.

[0107] Specifically, the answer is sent as an HTTP response.

[0108] Step 9:

[0109] Terminal: Displays the received answer to the user.

[0110] For example, the chat screen in your browser or app might display "Currently, there are 15 units of product A in stock."

[0111] Step 10:

[0112] Server: The query is permanently stored in a database along with the final answer.

[0113] Use SQL insert or update statements to store queries and their answers.

[0114] UPDATE inquiries SET response = 'Currently, there are 15 units of product A in stock' WHERE inquiry_id = '5678'

[0115] Step 11:

[0116] Server: Periodically analyzes the accumulated data to identify trends and patterns in inquiries.

[0117] Specifically, data analysis tools such as Python and R are used to analyze past inquiries in terms of distribution and frequency.

[0118] The above is the specific flow of processing from the inquiry content to returning a response to the user. This system improves the efficiency of inquiry responses and contributes to improving customer satisfaction.

[0119] Example 1

[0120] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0121] Conventional inquiry response systems take time to analyze the content of inquiries and generate responses, making it difficult to respond to user inquiries quickly. Furthermore, there was no established method for effectively utilizing accumulated data, making it difficult to predict future inquiries or provide appropriate information. This resulted in a lack of improvement in the user experience and a decline in the quality of a company's services.

[0122] 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.

[0123] In this invention, the server includes means for receiving an inquiry, means for analyzing the received inquiry, means for generating a response based on the analyzed inquiry, means for transmitting the generated response, means for storing the inquiry and the generated response in an information storage medium, and means for analyzing the stored data. This makes it possible to quickly and accurately analyze the inquiry and generate and return an appropriate response. Furthermore, by effectively utilizing the accumulated data, it is possible to predict future inquiries and provide useful information, thereby significantly improving the user experience.

[0124] The "means for receiving an inquiry" is a device or software that has the function of transmitting an inquiry from a user to a server as digital data.

[0125] "Means for analyzing the content of the received inquiry" refers to a device or software that has the function of analyzing the received digital data using methods such as natural language processing and understanding its intent and content.

[0126] The "means for generating a response based on the analyzed inquiry content" is a device or software that has the function of automatically generating an appropriate response based on the analysis results.

[0127] The "means for transmitting the generated answer" is a device or software that has the function of transmitting the generated answer to the user's terminal.

[0128] The "means for storing the inquiry content and the generated response in an information storage medium" refers to a device or software that has the function of recording the inquiry content and the response in digital form and storing it permanently.

[0129] "Means for analyzing stored data" refers to devices or software that have the ability to analyze accumulated data using statistical or machine learning techniques and extract trends and patterns.

[0130] "Natural language processing" is a technology that enables computers to understand and analyze human language, and involves processing text data using language models and algorithms.

[0131] A "terminal" is a device that allows a user to input and send an inquiry, and includes a computer, smartphone, tablet, etc.

[0132] An "information storage medium" is a device or medium for storing digital data for a long period of time, and includes hard disk drives (HDDs) and solid-state drives (SSDs).

[0133] The present invention relates to a system for improving the efficiency of an inquiry response process. The system is composed of a user, a terminal, and a server, and has the function of automating the reception, analysis, response generation, and response transmission of an inquiry. The program processing of the system and a specific embodiment are described below.

[0134] System Configuration

[0135] The system consists of the following main components:

[0136] 1. Means of receiving inquiries

[0137] 2. Means of analyzing the content of received inquiries

[0138] 3. A means of generating answers based on the parsed query content

[0139] 4. A means of sending the generated answer

[0140] 5. Means for storing the inquiry content and the generated response in an information storage medium

[0141] 6. Means of analyzing stored data

[0142] Receiving inquiries

[0143] The user enters an inquiry into their device and presses the send button. For example, they may enter "Please tell me the product's stock status."

[0144] The device sends the input query to the server via API, structuring the query in JSON format and sending it to the server using an HTTP POST request.

[0145] Saving inquiry details

[0146] The server temporarily stores the query received from the terminal in an information storage medium. Specifically, the server inserts the received query into the "ReceivedQueries" table.

[0147] Analysis of inquiry content

[0148] The server passes the received query content to a natural language processing (NLP) model to analyze the content, for example using TextBlob, spaCy, or a custom NLP model (e.g., the BERT model).

[0149] The server determines the intent of the query based on the output of the NLP model. For example, if a user asks, "Please tell me the stock status of a product," the NLP model determines the intent as "check stock."

[0150] Generate answers

[0151] The server retrieves the relevant information from the information storage medium based on the determined intention. For example, it selects the inventory information of the relevant product from the "Inventory" table.

[0152] The server generates an answer based on the information it has obtained, for example, constructing a sentence such as "There are currently 15 units of this product in stock." It may also generate sentences using templates.

[0153] Submit your answer

[0154] The server sends the generated answer to the user's device in JSON format as an HTTP response.

[0155] The device displays the received response to the user, for example, in a browser or a mobile app UI component.

[0156] Data storage

[0157] The server permanently stores the query and the generated response in an information storage medium. Specifically, the columns to be inserted into the "QueriesAndResponses" table include "user_id", "query", "response", and "timestamp".

[0158] Analyzing the data

[0159] The server periodically analyzes the accumulated query and response data, for example, using Python's Pandas library or SQL queries.

[0160] The server will use the analysis results to predict future inquiries and proactively provide relevant information. For example, machine learning models can be used to predict inquiry trends, improve the automated response system, and update FAQs.

[0161] Example prompt

[0162] An example of a prompt sentence could be "What is the system's procedure when a user asks, 'Is the product in stock?'"

[0163] This enables the system to respond to user inquiries quickly and accurately, enabling efficient inquiry management and future service improvements through the use of data.

[0164] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0165] Step 1:

[0166] The user inputs an inquiry into the terminal and presses the send button. At this time, the user inputs an inquiry such as "Please tell me the inventory status of the product." The input data is the inquiry content (e.g., "Do you have product A in stock?").

[0167] Step 2:

[0168] The device structures the query and sends it to the server via API. Specifically, it converts the query into JSON format and sends it using an HTTP POST request. The input is the user's query, and the output is JSON format data.

[0169] Step 3:

[0170] The server temporarily saves the query received from the terminal in a database. Specifically, it inserts it into the "ReceivedQueries" table. The input is the query in JSON format, and the output is the data saved in the database.

[0171] Step 4:

[0172] The server passes the received query content to a natural language processing (NLP) model to analyze the content. For example, it uses TextBlob, spaCy, or a custom NLP model (e.g., the BERT model). The input is the stored query data, and the output is the analysis result (e.g., the intent "check inventory").

[0173] Step 5:

[0174] The server determines the intent of the query based on the analysis results of the NLP model. The input is the analysis results from the NLP model, and the output is the determined intent (e.g., "check inventory").

[0175] Step 6:

[0176] The server retrieves the relevant information from the database based on the determined intent. For example, SELECT the inventory information of a specific product from the "Inventory" table. The input is the determined intent, and the output is the retrieved inventory information (e.g., "15 units in stock").

[0177] Step 7:

[0178] The server generates a response based on the acquired information. For example, it generates a sentence such as "Currently, there are 15 units of product A in stock." The input is the acquired inventory information, and the output is the generated response sentence.

[0179] Step 8:

[0180] The server sends the generated answer to the terminal in JSON format as an HTTP response. The input is the generated answer text, and the output is the JSON format response.

[0181] Step 9:

[0182] The device displays the received answer to the user, for example, in a UI component of a browser or mobile app. The input is a JSON-formatted response, and the output is the answer text that is displayed to the user.

[0183] Step 10:

[0184] The server stores the query and generated response in a database permanently. Specifically, it inserts the query and generated response into the "QueriesAndResponses" table. The input is the query and generated response, and the output is the record stored in the database.

[0185] Step 11:

[0186] The server periodically analyzes the accumulated data using Python's Pandas library and SQL queries. The input is the accumulated inquiry and response data, and the output is the analysis results and trend analysis.

[0187] Step 12:

[0188] The server uses the analysis results to predict future inquiries and builds a system that proactively provides relevant information. For example, a machine learning model can be used to predict inquiry trends, improve the automated response system, and update FAQs. The input is the analysis results, and the output is the predictive model and an improved system.

[0189] (Application example 1)

[0190] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0191] Conventional inquiry response systems often required manual response, resulting in the problem of long response times. It was also difficult to provide an appropriate response immediately based on the inquiry, which could lead to a decline in customer satisfaction. Furthermore, because data was not centrally managed, it was difficult to later analyze inquiries and their responses.

[0192] 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.

[0193] In this invention, the server includes means for using a natural language processing model to identify the intent of the inquiry, means for retrieving information from a database based on the intent of the inquiry, and means for returning the generated answer in JSON format. This enables the provision of an immediate and appropriate answer to the inquiry, and furthermore, centralized management and analysis of data enables the efficiency of subsequent inquiry responses and improvement of customer satisfaction.

[0194] The "means for receiving an inquiry" is a system that has the function of receiving an inquiry from a user via a digital device and transmitting the content of the inquiry to a server.

[0195] The "means for analyzing the received inquiry content" is a system that understands the inquiry content and analyzes it to determine its meaning and intent.

[0196] The "means for generating a response based on the analyzed inquiry content" is a system that has the function of automatically generating an appropriate response based on the analysis results.

[0197] The "means for transmitting the generated answer" is a system having a function for returning and displaying the generated answer to the user.

[0198] The "means for saving the inquiry content and the generated response in a database" is a system that has the function of recording and saving the inquiry content and the response to the inquiry in a database.

[0199] The "means for analyzing stored data" is a system that analyzes the history of inquiries and responses stored in a database and predicts and optimizes future inquiries.

[0200] The "means of using a natural language processing model to identify the intent of a query" is a system that has the function of using natural language processing technology to identify the intent of a query from its content.

[0201] The "means for retrieving information from a database based on the intent of a query" is a system that has the function of searching and retrieving related information from a database in accordance with the specified intent.

[0202] "Means for returning the generated answer in JSON format" refers to a system that has the function of returning the generated answer to the user as JSON format data.

[0203] This invention relates to a system that automatically receives and analyzes inquiries from users, and generates and transmits answers. In this system, the roles of the server, terminal, and user are clarified, and specific implementation means are described below.

[0204] System Configuration

[0205] This system consists of the following main components:

[0206] 1. Means of receiving inquiries

[0207] 2. A means of analyzing the content of the received inquiry (using a natural language processing model)

[0208] 3. A means of generating answers based on the parsed query content

[0209] 4. How to send the generated answer (returned in JSON format)

[0210] 5. A means of storing queries and generated responses in a database

[0211] 6. Means of analyzing stored data

[0212] Hardware and software used

[0213] Server: High-performance cloud server (e.g. AWS, Google Cloud)

[0214] Device: The user's smartphone, tablet, or PC

[0215] Database: SQLite or MySQL

[0216] Natural language processing models: SpaCy, Transformers library (e.g., Hugging Face transformers)

[0217] Framework: Python, Flask (Web application framework)

[0218] Program processing overview

[0219] The server receives queries sent from the user's device via an API and temporarily stores them in a database. It then passes the queries to a natural language processing model (e.g., SpaCy or Transformers) to analyze them and identify their intent. It then retrieves relevant information from the database based on the analysis results and generates an appropriate answer. This answer is sent to the user's device in JSON format. The query and the generated answer are also stored in a database for use in subsequent data analysis.

[0220] Specific examples

[0221] For product inquiries:

[0222] A user makes a query through a smartphone application, asking, "Please tell me the stock of product A." This query is sent from the device to the server, which temporarily stores the query in a database. Using a natural language processing model, the intent "to check stock" is identified and the stock information of product A is retrieved from the database. Finally, the server generates a response saying, "There are currently 15 units of product A in stock," and sends it back to the user's device in JSON format.

[0223] Example prompt sentence:

[0224] "Please let me know the stock of product A."

[0225] For shipping inquiries:

[0226] A user makes a query from their PC asking, "Please tell me the delivery status of product B." This query is sent from the device to the server, which temporarily stores the query in a database. Using a natural language processing model, the intent "check delivery status" is identified and the delivery information for product B is retrieved from the database. Finally, the server generates a response stating, "Product B is currently being delivered. The expected delivery date is next Tuesday," and sends it back to the user's device in JSON format.

[0227] Example prompt sentence:

[0228] "Please let me know the delivery status of product B."

[0229] This will enable immediate and appropriate responses to inquiries, and through centralized data management and analysis, will improve customer satisfaction and make future inquiries more efficient.

[0230] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0231] Step 1:

[0232] The user enters an inquiry from their device (smartphone, tablet, or PC) and presses the send button. At this time, the entered inquiry content is sent to the server as an API request.

[0233] Input: The inquiry entered by the user (e.g., "Please tell me the inventory of product A.")

[0234] Output: Query data in API request format

[0235] Step 2:

[0236] The terminal sends the received inquiry to the server via API and temporarily stores it in a database.

[0237] Input: Inquiry data in API request format

[0238] Output: A database containing the query results

[0239] Specific operation: The server receives a request from a terminal and records the contents in a specific table in the database.

[0240] Step 3:

[0241] The server passes the query content stored in the database to a natural language processing (NLP) model to analyze the content.

[0242] Input: Inquiry details saved in the database

[0243] Output: Analysis results from the natural language processing model (intent and extracted entities)

[0244] Specific operation: The server passes the query content to a natural language processing model (e.g., SpaCy or Hugging Face's transformers model), determines the intent as "check inventory," etc., and extracts entities such as product names.

[0245] Step 4:

[0246] The server determines the intent of the query based on the analysis results of the NLP model and retrieves the relevant information from the database.

[0247] Input: Analysis results from natural language processing model

[0248] Output: Result of retrieving related information (e.g., inventory quantity and delivery status)

[0249] Specific behavior: The server generates and executes a query to the database based on the identified intent, and retrieves the relevant data (e.g., product inventory and delivery status).

[0250] Step 5:

[0251] The server generates a response based on the acquired information and sends it to the user's terminal.

[0252] Input: Related information (e.g., inventory quantity and delivery status)

[0253] Output: Generated answer (e.g. "There are 15 units of product A in stock.")

[0254] Specific operation: The server uses the acquired data to create a corresponding answer using a template or plain text generation, and returns it to the user's device in JSON format.

[0255] Step 6:

[0256] The terminal displays the received response to the user.

[0257] Input: Response data sent from the server

[0258] Output: User-visible answers

[0259] Specific operation: The terminal analyzes the received JSON formatted response data and displays it in a format that is easy for the user to understand via the user interface.

[0260] Step 7:

[0261] The server stores the final query and response in a database for later data analysis.

[0262] Input: Final query and generated answer

[0263] Output: A record of the query and answer stored in a database

[0264] Specific operation: The server records pairs of queries and their answers in a database and uses them as data for future trend analysis and system improvement.

[0265] Example prompt sentence:

[0266] "Please let me know the stock of product A."

[0267] 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.

[0268] This invention provides a system that automatically receives and analyzes user inquiries, generates appropriate responses, and returns them with an emotion engine that recognizes the user's emotions, enabling even more sophisticated responses. This system is primarily composed of a server, terminals, and users. Below are the system's program processing and specific examples.

[0269] System Configuration

[0270] This system consists of the following main components:

[0271] 1. Means of receiving inquiries

[0272] 2. Means of analyzing the content of received inquiries

[0273] 3. A means of generating answers based on the parsed query content

[0274] 4. A means of sending the generated answer

[0275] 5. A means of storing queries and generated responses in a database

[0276] 6. Means of analyzing stored data

[0277] 7. Emotion engine that recognizes user emotions

[0278] Program processing

[0279] How to receive inquiries

[0280] The user inputs a query using their own terminal and presses the send button, which sends the query to the server.

[0281] The terminal sends the entered inquiry to the server via API.

[0282] A means of analyzing the content of received inquiries

[0283] The server temporarily stores the inquiry received from the terminal in a database.

[0284] The server passes the received query content to a natural language processing (NLP) model to analyze the content.

[0285] Emotion engine that recognizes user emotions

[0286] The server passes the data, along with the query content, to an emotion engine for analyzing the user's emotions.

[0287] The emotion engine analyzes emotions from the user's text and returns the results to the server, identifying emotions such as "anger," "joy," and "sadness."

[0288] A means of generating answers based on the parsed query content

[0289] The server determines the intent of the query and the user's emotions based on the analysis results of the NLP model and the emotion engine.

[0290] For example, if a user inquires, "Please tell me the product's stock status," the server determines the intent as "checking stock" and the emotion as "neutral."

[0291] A means of sending the generated answer

[0292] Based on the determined intent and emotion, the server retrieves relevant information from a database and generates a response in an appropriate tone.

[0293] The server generates a response based on the acquired information and sends it to the user's terminal.

[0294] The terminal displays the received response to the user.

[0295] A means of storing queries and generated responses in a database

[0296] The server permanently stores the query content and the results of sentiment analysis in a database, along with the final answer.

[0297] A means of analyzing stored data

[0298] The server periodically analyzes the accumulated inquiry, response, and sentiment analysis data to extract inquiry trends and patterns.

[0299] Based on the analysis results, we will build a system that proactively provides relevant information that responds to future inquiries and emotional states.

[0300] Specific examples

[0301] Product inquiries

[0302] User: The user enters "Is product A in stock?" and presses the send button.

[0303] Terminal: The terminal sends this query to the server.

[0304] Server: The server temporarily stores the query content in a database and passes it to the natural language processing model.

[0305] Server: The NLP model determines the intent, "Check inventory," and the emotion engine determines the emotion, "Neutral."

[0306] Server: The server retrieves the stock information of product A from the database, generates a response in a neutral tone such as "Currently, there are 15 units of product A in stock," and sends it to the user's device.

[0307] Terminal: The terminal displays this answer to the user.

[0308] Server: The final query, response, and sentiment analysis results are stored in a database for later data analysis.

[0309] Emotion-conscious response to inquiries

[0310] User: The user types, "The delivery of product B is delayed. What's going on?" and submits.

[0311] Terminal: The terminal sends this query to the server.

[0312] Server: The server temporarily stores the query content in a database and passes it to the natural language processing model.

[0313] Server: The NLP model determines the intent, "check delivery status," and the emotion engine determines the emotion, "anger."

[0314] Server: The server retrieves the delivery status of Product B from the delivery system database, generates a polite response saying, "Product B is currently being delivered. We apologize for any inconvenience. The expected delivery date is next Tuesday," and sends it to the user's device.

[0315] Terminal: The terminal displays this answer to the user.

[0316] Server: The final query, response, and sentiment analysis results are stored in a database for later data analysis.

[0317] This allows the system to respond to inquiries taking emotions into consideration, contributing to improved customer satisfaction.

[0318] The processing flow will be explained below.

[0319] Step 1:

[0320] User: Enter the inquiry details and click the send button.

[0321] For example, enter "Please let me know the product availability."

[0322] Step 2:

[0323] Terminal: Sends the entered inquiry to the server.

[0324] Specifically, the inquiry data is sent via the API using an HTTP POST request.

[0325] Step 3:

[0326] Server: Temporarily stores the received inquiry in a database.

[0327] Specifically, the query content is recorded in the query table using an SQL insert statement.

[0328] Example: INSERT INTO inquiries (user_id, inquiry_text) VALUES ('1234', 'What is the product availability?')

[0329] Step 4:

[0330] Server: Passes the received query content to a natural language processing (NLP) model for analysis.

[0331] Specifically, the query is passed to the NLP model API and the analysis results are received in JSON format.

[0332] Example: "Product availability" is parsed as "Intent: Check inventory."

[0333] Step 5:

[0334] Server: Passes the query content to the emotion engine and analyzes the user's emotions.

[0335] Specifically, the query is passed to the emotion engine's API, and the results of the emotion analysis are received in JSON format.

[0336] Example: "What is the product availability?" is parsed as "Neutral."

[0337] Step 6:

[0338] Server: Determines the intent of the query and the user's emotions based on the analysis results of the NLP model and the emotion engine.

[0339] For example, "Intention: Check inventory" and "Emotion: Neutral" are determined.

[0340] Step 7:

[0341] Server: Based on the intent of the query, retrieves the necessary information from the database.

[0342] Specifically, an SQL query is executed to obtain the stock quantity of the relevant product from the inventory table.

[0343] Example: SELECT stock_quantity FROM products WHERE product_name = 'Product A'

[0344] Step 8:

[0345] Server: Generates answers based on the acquired information and the user's sentiment.

[0346] For example, frame your response in a neutral tone, saying, "We currently have 15 units of product A in stock."

[0347] Step 9:

[0348] Server: Sends the generated answer to the user's device.

[0349] Specifically, the answer is sent as an HTTP response.

[0350] Step 10:

[0351] Terminal: Displays the received answer to the user.

[0352] For example, the chat screen in your browser or app might display "Currently, there are 15 units of product A in stock."

[0353] Step 11:

[0354] Server: The inquiry content and sentiment analysis results, along with the final answer, are permanently stored in a database.

[0355] Use SQL insert or update statements to store queries and their answers.

[0356] Example: UPDATE inquiries SET response = 'Currently, there are 15 units of product A in stock', emotion = 'Neutral' WHERE inquiry_id = '5678'

[0357] Step 12:

[0358] Server: Periodically analyzes the accumulated data to identify trends and patterns in inquiries.

[0359] Specifically, data analysis tools such as Python and R are used to analyze past inquiries in terms of distribution and frequency, as well as emotional trends.

[0360] The above is the specific flow of processing the inquiry content, generating and providing an answer taking the user's feelings into consideration. This system will improve the efficiency of inquiry responses and contribute to improving customer satisfaction.

[0361] Example 2

[0362] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0363] Conventional inquiry systems only provided standardized answers without considering the user's feelings, resulting in low user satisfaction and the inability to respond appropriately to complaints and urgent inquiries in particular. In addition, the stored data was treated as a simple log, and could not be used to proactively respond to future inquiries or for improvements.

[0364] 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.

[0365] In this invention, the server includes means for receiving an inquiry, means for analyzing the content of the received inquiry, means for recognizing the user's emotion, means for generating an answer based on the analyzed content of the inquiry and the recognized emotion, means for transmitting the generated answer, means for saving the content of the inquiry and the generated answer in a database, and means for analyzing the saved data. This makes it possible to generate an appropriate answer that takes the user's emotion into consideration, and further makes it possible to proactively respond to future inquiries by utilizing the accumulated data.

[0366] The "means for receiving an inquiry" is a device or process for receiving the content of an inquiry sent from a user to the server via the terminal.

[0367] "Means for analyzing the received query content" refers to a device or process for understanding the received query text and identifying its intent, often using natural language processing (NLP) techniques.

[0368] The "means for recognizing user emotions" is a device or process for determining the user's emotions from the query text, and often utilizes an emotion analysis engine.

[0369] The "means for generating an answer based on the analyzed query content and the recognized sentiment" is a device or process for generating an appropriate answer based on the intent of the query and the recognized sentiment.

[0370] A "means for transmitting a generated answer" is a device or process for transmitting a generated answer to a user.

[0371] The "means for storing the inquiry contents and generated answers in a database" is a device or process for storing all the inquiry contents and the generated answers thereto in a database.

[0372] A "means for analyzing stored data" is a device or process that uses stored inquiry and response data to extract trends and patterns to improve future inquiry responses.

[0373] "Natural Language Processing (NLP)" is a technology that enables machines to understand, analyze, and generate human language.

[0374] An "emotion analysis engine" is an algorithm or software for identifying a user's emotions from text data.

[0375] A "database" is an information storage system for storing queries and generated responses.

[0376] This invention enables more advanced responses by combining a system that receives and analyzes user inquiries, generates appropriate responses, and responds with an emotion engine that recognizes the user's emotions. The system consists of the following main components:

[0377] System Configuration

[0378] This system is mainly composed of a server, terminals, and users. Specifically, it uses the following devices and software:

[0379] server

[0380] Hardware: A server machine equipped with a high-performance processor, memory, and storage

[0381] software:

[0382] Database: MySQL, PostgreSQL, MongoDB, Elasticsearch, etc.

[0383] Natural Language Processing (NLP) libraries: spaCy, NLTK

[0384] Sentiment analysis engine: Microsoft Azure's Text Analytics API, IBM Watson's Natural Language Understanding API

[0385] Text generation model: OpenAI's GPT-3

[0386] Terminal

[0387] Hardware: Smartphones, tablets, personal computers

[0388] Software: Web browsers (Google Chrome, Safari, etc.), mobile applications

[0389] User

[0390] Interface: inquiry form, etc.

[0391] Program processing

[0392] Receive inquiries

[0393] The user uses their device (smartphone or computer) to enter the details of their inquiry. For example, they might enter "Please tell me the product's stock status" in a web browser and press the send button. The device then sends this inquiry to the server via an API (RESTful API or GraphQL).

[0394] Analyze the incoming inquiries

[0395] The server temporarily stores the query received from the device in a database (MySQL or PostgreSQL). It then uses a natural language processing engine (spaCy or NLTK) to analyze the intent of the query. For example, it classifies the text "Please tell me the stock status of the product" as the intent "Check stock."

[0396] Recognize user emotions

[0397] The server sends the received inquiry content to a sentiment analysis engine, which determines the user's emotions (such as "anger," "joy," or "sadness") from the text. Examples of sentiment analysis engines include Microsoft Azure's Text Analytics API and IBM Watson's Natural Language Understanding API.

[0398] Generate an answer

[0399] The server generates a response using a text generation model (OpenAI's GPT-3) based on the intent and sentiment analysis results. For example, it generates a neutral response such as "Currently, there are 15 units of product A in stock."

[0400] Submit your answer

[0401] The server sends the generated answer to the user's terminal, which displays the answer to the user.

[0402] Save your data

[0403] The server stores the query content and sentiment analysis results along with the final answer in a database (MongoDB or Elasticsearch).

[0404] Analyze the data

[0405] The stored data is periodically analyzed and data analysis tools such as Apache Hadoop and Google BigQuery are used to extract trends and patterns in inquiries, which can then be used to proactively respond to future inquiries.

[0406] Specific examples

[0407] Product inquiries

[0408] A user enters "Do you have product A in stock?" into an inquiry form in Google Chrome and presses the submit button. The device sends this inquiry to the server via a RESTful API. The server temporarily saves the inquiry content in MySQL and uses spaCy to extract the intent "to check stock." It then uses Microsoft Azure's Text Analytics to determine the sentiment of "neutral." It retrieves the stock information for product A from the database and generates a response in a neutral tone, such as "There are currently 15 units of product A in stock." This response is then sent back to the device in JSON format. The device displays this response on the user interface and saves the final inquiry, response, and sentiment analysis results in a MySQL database.

[0409] Inquiries about delivery status

[0410] The user types and submits the query in Safari, "The delivery of Product B is delayed. What's going on?" The device sends this query to the server using the GraphQL API. The server temporarily stores the query content in PostgreSQL and uses NLTK to extract the intent, "Check delivery status." It then uses IBM Watson's Natural Language Understanding to determine the emotion, "anger." It retrieves the delivery status of Product B from the delivery system database and generates a polite response, saying, "Product B is currently being delivered. We apologize for any inconvenience. The expected delivery date is next Tuesday." The response returned to the device in JSON format is displayed on the user interface, and the final query, response, and emotion analysis results are saved in the PostgreSQL database.

[0411] This allows the system to respond to inquiries taking into account the user's emotions, contributing to improved customer satisfaction.

[0412] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0413] Step 1: Receiving an inquiry

[0414] The user uses their own device (smartphone or computer) to enter the details of their inquiry and press the send button. Specifically, the user opens a web browser (e.g., Google Chrome), enters "Please let me know the product's stock status" into the inquiry form, and clicks send.

[0415] Input: User-supplied query text

[0416] Output: Data sent from the device to the server (inquiry details, user ID, timestamp, etc.)

[0417] The device sends the entered inquiry to the server via an API (e.g., RESTful API). At this time, the sent data includes metadata such as the user ID and timestamp.

[0418] Step 2: Analyze the query

[0419] The server temporarily stores the query received from the terminal in a database (e.g., MySQL). This temporary storage is performed as a transaction to maintain data consistency.

[0420] Input: Inquiry data sent from the terminal

[0421] Output: Temporarily saved inquiry data

[0422] The server uses a natural language processing (NLP) engine (e.g., spaCy) to analyze the query. Specifically, it extracts the intent from the query text. For example, the text "Please tell me the product's stock status" is classified as the intent "Check stock."

[0423] Step 3: Recognize the user's emotions

[0424] The server sends the received inquiry content to a sentiment analysis engine (e.g., Microsoft Azure's Text Analytics API) to analyze the user's sentiment.

[0425] Input: Inquiry details

[0426] Output: Emotion analysis result (e.g. "neutral")

[0427] The sentiment analysis engine determines the user's sentiment from the query text, for example, "Please tell me the product availability status," and identifies the sentiment "neutral."

[0428] Step 4: Generate an answer

[0429] The server generates an answer using a text generation model (e.g., OpenAI's GPT-3) based on the results of intent analysis and sentiment analysis.

[0430] Input: Intent analysis results (e.g., "Check inventory"), sentiment analysis results (e.g., "Neutral")

[0431] Output: Generated answer (e.g. "Currently, there are 15 units of product A in stock.")

[0432] The server retrieves the necessary information from a database (e.g., MongoDB) based on the intent. For example, for the intent "check stock," it retrieves stock information for product A. It then uses a text generation model to generate an answer in an appropriate tone. For example, it generates an answer such as "Currently, there are 15 units of product A in stock."

[0433] Step 5: Submit your response

[0434] The server then sends the generated response to the user's device, often in JSON format.

[0435] Input: Generated response data

[0436] Output: Data sent to the terminal

[0437] The terminal displays the received answer on the user interface, allowing the user to check the answer on their screen.

[0438] Step 6: Save your data

[0439] The server stores the final answer, inquiry details, and sentiment analysis results in a database (e.g., Elasticsearch).

[0440] Input: Final response, inquiry details, sentiment analysis results

[0441] Output: Data stored in the database

[0442] The saved data can be used for later data analysis.

[0443] Step 7: Analyze the data

[0444] The server periodically analyzes the accumulated inquiry, response, and sentiment analysis data.

[0445] Input: Data of inquiries, responses, and sentiment analysis results stored in the database

[0446] Output: Analysis results (inquiry trends and patterns)

[0447] Specifically, Apache Hadoop and Google BigQuery are used to process data and extract trends and patterns. The analysis results are then used to respond to future inquiries. For example, if there are many inquiries about a particular season or product, that information can be prepared in advance.

[0448] (Application example 2)

[0449] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0450] For modern online shopping sites, responding to user inquiries quickly and appropriately is an important issue that directly contributes to improving customer satisfaction. However, conventional inquiry response systems do not adequately take user emotions into account, resulting in inconsistent response quality. In particular, when a user is dissatisfied or angry, it is difficult to respond appropriately based on that emotion. Another challenge is efficiently analyzing the content of a large number of inquiries and generating appropriate answers.

[0451] 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.

[0452] In this invention, the server includes means for receiving an inquiry, means for analyzing the received inquiry, means for generating a response based on the analyzed inquiry, means for transmitting the generated response, means for saving the inquiry and the generated response in a database, means for analyzing the saved data, means including an emotion engine for recognizing the user's emotion, and means for adjusting the tone of the response in accordance with the user's emotion, thereby enabling a prompt and appropriate response to an inquiry that takes the user's emotion into consideration.

[0453] The "means for receiving an inquiry" is a function by which the system receives and processes an inquiry from a user.

[0454] The "means for analyzing the content of the received inquiry" is a function for understanding the content of the received inquiry and analyzing it as structured data.

[0455] The "means for generating a response based on the analyzed inquiry content" is a function that automatically generates an appropriate response based on the analysis results.

[0456] The "means for sending the generated answer" is a function for returning the generated answer to the user.

[0457] The "means for saving the inquiry and the generated response in a database" is a function for saving the inquiry and the response as a record in a database.

[0458] The "means for analyzing stored data" is a function for analyzing stored inquiry and response data and extracting significant insights and patterns.

[0459] The "emotion engine that recognizes user emotions" is an engine that analyzes and identifies emotions from the content of a user's inquiry.

[0460] The "means for adjusting the tone of the response according to the user's emotion" is a function for automatically adjusting the expression and tone of the response according to the emotion specified by the user.

[0461] The present invention is a system for responding promptly and appropriately to inquiries from users on an online shopping site. This system analyzes the content of the user's inquiry and recognizes their emotions, thereby generating a response in a tone that corresponds to their emotions and improving the quality of the response.

[0462] System Configuration

[0463] The system mainly consists of the following components:

[0464] 1. Means of receiving inquiries

[0465] 2. Means of analyzing the content of received inquiries

[0466] 3. A means of generating answers based on the parsed query content

[0467] 4. A means of sending the generated answer

[0468] 5. A means of storing queries and generated responses in a database

[0469] 6. Means of analyzing stored data

[0470] 7. Emotion engine that recognizes user emotions

[0471] 8. How to adjust the tone of your response based on the user's emotions

[0472] Hardware and Software

[0473] Hardware used:

[0474] server

[0475] User device (smartphone or PC)

[0476] Software used:

[0477] Python

[0478] Natural Language Processing Library (Hugging Face Transformers)

[0479] Sentiment analysis model (BERT)

[0480] Database System

[0481] Data processing and calculation

[0482] 1. Means of receiving inquiries:

[0483] The user uses a smartphone or computer to enter and submit an inquiry.

[0484] The terminal sends the entered inquiry to the server via API.

[0485] 2. How to analyze the received inquiry:

[0486] The server temporarily stores the inquiry received from the terminal in a database.

[0487] The data stored in the database is analyzed using natural language processing models.

[0488] 3. Emotion engine that recognizes user emotions:

[0489] The server passes the data along with the query to the emotion engine.

[0490] The emotion engine analyzes the emotion from the query and returns the result to the server. The emotion may be "anger," "joy," or "sadness," for example.

[0491] 4. Means of generating an answer based on the parsed query:

[0492] The server determines the intent of the query and the user's emotions based on the analysis results of the NLP model and the emotion engine.

[0493] 5. Means for sending generated answers:

[0494] The server generates a response in an appropriate tone based on the determined intent and emotion.

[0495] The generated answer is sent to the user's terminal.

[0496] 6. How to store queries and generated responses in a database:

[0497] The inquiry details and the results of sentiment analysis are stored in a database along with the final answer.

[0498] Examples of specific examples and prompts

[0499] For example, if a user asks, "Product B hasn't arrived yet. What's going on?", the following process will be performed:

[0500] User inquiry: "I haven't received my delivery of product B yet. What's going on?"

[0501] Emotion engine judgement: "Anger"

[0502] System response: "I see you're angry. We'll deal with it right away."

[0503] Example prompt sentence:

[0504] User Question: "I haven't received my delivery of item B yet, what's going on?"

[0505] System emotion detection: "Anger"

[0506] Generate an appropriate response.

[0507] This allows for prompt and appropriate response to inquiries that take into account the user's feelings.

[0508] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0509] Step 1:

[0510] The user enters an inquiry on a smartphone or PC. When the user enters the inquiry content as text and presses the send button, the device sends the inquiry data to the server via the API.

[0511] Input: The query text entered by the user.

[0512] Output: The query data sent to the server.

[0513] Step 2:

[0514] The server temporarily stores the received inquiry in a database.

[0515] Input: Enquiry data received through API.

[0516] Output: Query data temporarily stored in a database.

[0517] Step 3:

[0518] The server passes the received query content to a natural language processing (NLP) model for analysis, which analyzes the text and extracts the intent of the query.

[0519] Input: Query data stored in the database.

[0520] Output: Analysis result data including query intent.

[0521] Step 4:

[0522] The server passes the query content and sentiment analysis data to the sentiment engine, which analyzes the sentiment from the user's text and returns the results to the server.

[0523] Input: Query text.

[0524] Output: Emotion analysis results from the emotion engine (e.g., "anger," "joy," "sadness," etc.).

[0525] Step 5:

[0526] The server determines the intent of the query and the user's emotions based on the analysis results of the NLP model and the emotion engine.

[0527] Input: Analysis results of the NLP model and the emotion engine.

[0528] Output: Analysis result data including intent and sentiment.

[0529] Step 6:

[0530] Based on the determined intent and emotion, the server retrieves relevant information from a database and generates a response in an appropriate tone.

[0531] Input: Analysis result data including intent and sentiment.

[0532] Output: Answer text generated in the appropriate tone.

[0533] Step 7:

[0534] The server sends the generated answer to the user's terminal, which then displays the received answer to the user.

[0535] Input: The generated answer text.

[0536] Output: The answer text that is displayed on the user's terminal.

[0537] Step 8:

[0538] The server stores the query and the results of the sentiment analysis in a database, along with the final answer.

[0539] Input: Query, generated answer, sentiment analysis results.

[0540] Output: The final data stored in the database.

[0541] 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.

[0542] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0543] 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.

[0544] [Second embodiment]

[0545] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0546] 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.

[0547] 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).

[0548] 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.

[0549] 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.

[0550] 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).

[0551] 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. 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.

[0552] 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.

[0553] 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.

[0554] 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.

[0555] 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.

[0556] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0557] The system of the present invention responds quickly and efficiently by automatically receiving and analyzing inquiries from users and generating and returning appropriate answers. This system is primarily composed of a server, terminals, and users. The following describes the processing of the system program and a specific example.

[0558] System Configuration

[0559] This system consists of the following main components:

[0560] 1. Means of receiving inquiries

[0561] 2. Means of analyzing the content of received inquiries

[0562] 3. A means of generating answers based on the parsed query content

[0563] 4. A means of sending the generated answer

[0564] 5. A means of storing queries and generated responses in a database

[0565] 6. Means of analyzing stored data

[0566] Program processing

[0567] How to receive inquiries

[0568] The user inputs a query using their own terminal and presses the send button, which sends the query to the server.

[0569] The terminal sends the entered inquiry to the server via API.

[0570] A means of analyzing the content of received inquiries

[0571] The server temporarily stores the inquiry received from the terminal in a database.

[0572] The server passes the received query content to a natural language processing (NLP) model to analyze the content.

[0573] A means of generating answers based on the parsed query content

[0574] The server determines the intent of the query based on the analysis results of the NLP model.

[0575] For example, if a user inquires, "Please tell me the stock status of the product," the server determines that the user's intent is "to check stock."

[0576] A means of sending the generated answer

[0577] The server retrieves the relevant information from the database based on the determined intent.

[0578] The server generates a response based on the acquired information and sends it to the user's terminal.

[0579] The terminal displays the received response to the user.

[0580] A means of storing queries and generated responses in a database

[0581] The server permanently stores the final answer together with the inquiry details in a database.

[0582] A means of analyzing stored data

[0583] The server periodically analyzes the accumulated inquiry and response data to extract inquiry trends and patterns.

[0584] Based on the analysis results, we will build a system that predicts future inquiries and provides relevant information in advance.

[0585] Specific examples

[0586] Product inquiries

[0587] User: The user enters "Is product A in stock?" and presses the send button.

[0588] Terminal: The terminal sends this query to the server.

[0589] Server: The server temporarily stores the query content in a database and passes it to the natural language processing model.

[0590] Server: The NLP model determines the intent to "check stock," and the server retrieves stock information for product A from the database.

[0591] Server: Generates a response saying "Currently, there are 15 units of product A in stock" and sends it to the user's device.

[0592] Terminal: The terminal displays this answer to the user.

[0593] Server: The final query and answer are stored in a database for later data analysis.

[0594] Delivery inquiries

[0595] User: The user enters "Please tell me the delivery status of product B" and submits.

[0596] Terminal: The terminal sends this query to the server.

[0597] Server: The server temporarily stores the query content in a database and passes it to the natural language processing model.

[0598] Server: The NLP model determines the intent to "check delivery status," and the server retrieves the delivery status of Product B from the delivery system database.

[0599] Server: Generates a response stating, "Item B is currently being delivered. The estimated delivery date is next Tuesday," and sends it to the user's device.

[0600] Terminal: The terminal displays this answer to the user.

[0601] Server: The final query and answer are stored in a database for later data analysis.

[0602] This allows the system to automate inquiries efficiently and accurately, and the accumulated data can be used to improve customer service in the future.

[0603] The processing flow will be explained below.

[0604] Step 1:

[0605] User: Enter the inquiry details and click the send button.

[0606] For example, enter "Please let me know the product availability."

[0607] Step 2:

[0608] Terminal: Sends the entered inquiry to the server.

[0609] Specifically, the inquiry data is sent via the API using an HTTP POST request.

[0610] Step 3:

[0611] Server: Temporarily stores the received inquiry in a database.

[0612] Specifically, the query content is recorded in the query table using an SQL insert statement.

[0613] INSERT INTO inquiries (user_id, inquiry_text) VALUES ('1234', 'What is the product availability?')

[0614] Step 4:

[0615] Server: Passes the received query content to a natural language processing (NLP) model for analysis.

[0616] Specifically, the query is passed to the NLP model API and the analysis results are received in JSON format.

[0617] Example: "Product availability" is parsed as "Intent: Check inventory."

[0618] Step 5:

[0619] Server: Determines the intent of the query based on the analysis results of the NLP model.

[0620] Based on the analysis results, the corresponding processing is selected.

[0621] Example: Determine the intent "check inventory."

[0622] Step 6:

[0623] Server: Based on the intent of the query, retrieves the necessary information from the database.

[0624] Specifically, an SQL query is executed to obtain the stock quantity of the relevant product from the inventory table.

[0625] SELECT stock_quantity FROM products WHERE product_name = 'A product'

[0626] Step 7:

[0627] Server: Generates an answer based on the information obtained.

[0628] For example, construct an answer such as "Currently, there are 15 units of product A in stock."

[0629] Step 8:

[0630] Server: Sends the generated answer to the user's device.

[0631] Specifically, the answer is sent as an HTTP response.

[0632] Step 9:

[0633] Terminal: Displays the received answer to the user.

[0634] For example, the chat screen in your browser or app might display "Currently, there are 15 units of product A in stock."

[0635] Step 10:

[0636] Server: The query is permanently stored in a database along with the final answer.

[0637] Use SQL insert or update statements to store queries and their answers.

[0638] UPDATE inquiries SET response = 'Currently, there are 15 units of product A in stock' WHERE inquiry_id = '5678'

[0639] Step 11:

[0640] Server: Periodically analyzes the accumulated data to identify trends and patterns in inquiries.

[0641] Specifically, data analysis tools such as Python and R are used to analyze past inquiries in terms of distribution and frequency.

[0642] The above is the specific flow of processing from the inquiry content to returning a response to the user. This system improves the efficiency of inquiry responses and contributes to improving customer satisfaction.

[0643] Example 1

[0644] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0645] Conventional inquiry response systems take time to analyze the content of inquiries and generate responses, making it difficult to respond to user inquiries quickly. Furthermore, there was no established method for effectively utilizing accumulated data, making it difficult to predict future inquiries or provide appropriate information. This resulted in a lack of improvement in the user experience and a decline in the quality of a company's services.

[0646] 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.

[0647] In this invention, the server includes means for receiving an inquiry, means for analyzing the received inquiry, means for generating a response based on the analyzed inquiry, means for transmitting the generated response, means for storing the inquiry and the generated response in an information storage medium, and means for analyzing the stored data. This makes it possible to quickly and accurately analyze the inquiry and generate and return an appropriate response. Furthermore, by effectively utilizing the accumulated data, it is possible to predict future inquiries and provide useful information, thereby significantly improving the user experience.

[0648] The "means for receiving an inquiry" is a device or software that has the function of transmitting an inquiry from a user to a server as digital data.

[0649] "Means for analyzing the content of the received inquiry" refers to a device or software that has the function of analyzing the received digital data using methods such as natural language processing and understanding its intent and content.

[0650] The "means for generating a response based on the analyzed inquiry content" is a device or software that has the function of automatically generating an appropriate response based on the analysis results.

[0651] The "means for transmitting the generated answer" is a device or software that has the function of transmitting the generated answer to the user's terminal.

[0652] The "means for storing the inquiry content and the generated response in an information storage medium" refers to a device or software that has the function of recording the inquiry content and the response in digital form and storing it permanently.

[0653] "Means for analyzing stored data" refers to devices or software that have the ability to analyze accumulated data using statistical or machine learning techniques and extract trends and patterns.

[0654] "Natural language processing" is a technology that enables computers to understand and analyze human language, and involves processing text data using language models and algorithms.

[0655] A "terminal" is a device that allows a user to input and send an inquiry, and includes a computer, smartphone, tablet, etc.

[0656] An "information storage medium" is a device or medium for storing digital data for a long period of time, and includes hard disk drives (HDDs) and solid-state drives (SSDs).

[0657] The present invention relates to a system for improving the efficiency of an inquiry response process. The system is composed of a user, a terminal, and a server, and has the function of automating the reception, analysis, response generation, and response transmission of an inquiry. The program processing of the system and a specific embodiment are described below.

[0658] System Configuration

[0659] The system consists of the following main components:

[0660] 1. Means of receiving inquiries

[0661] 2. Means of analyzing the content of received inquiries

[0662] 3. A means of generating answers based on the parsed query content

[0663] 4. A means of sending the generated answer

[0664] 5. Means for storing the inquiry content and the generated response in an information storage medium

[0665] 6. Means of analyzing stored data

[0666] Receiving inquiries

[0667] The user enters an inquiry into their device and presses the send button. For example, they may enter "Please tell me the product's stock status."

[0668] The device sends the input query to the server via API, structuring the query in JSON format and sending it to the server using an HTTP POST request.

[0669] Saving inquiry details

[0670] The server temporarily stores the query received from the terminal in an information storage medium. Specifically, the server inserts the received query into the "ReceivedQueries" table.

[0671] Analysis of inquiry content

[0672] The server passes the received query content to a natural language processing (NLP) model to analyze the content, for example using TextBlob, spaCy, or a custom NLP model (e.g., the BERT model).

[0673] The server determines the intent of the query based on the output of the NLP model. For example, if a user asks, "Please tell me the stock status of a product," the NLP model determines the intent as "check stock."

[0674] Generate answers

[0675] The server retrieves the relevant information from the information storage medium based on the determined intention. For example, it selects the inventory information of the relevant product from the "Inventory" table.

[0676] The server generates an answer based on the information it has obtained, for example, constructing a sentence such as "There are currently 15 units of this product in stock." It may also generate sentences using templates.

[0677] Submit your answer

[0678] The server sends the generated answer to the user's device in JSON format as an HTTP response.

[0679] The device displays the received response to the user, for example, in a browser or a mobile app UI component.

[0680] Data storage

[0681] The server permanently stores the query and the generated response in an information storage medium. Specifically, the columns to be inserted into the "QueriesAndResponses" table include "user_id", "query", "response", and "timestamp".

[0682] Analyzing the data

[0683] The server periodically analyzes the accumulated query and response data, for example, using Python's Pandas library or SQL queries.

[0684] The server will use the analysis results to predict future inquiries and proactively provide relevant information. For example, machine learning models can be used to predict inquiry trends, improve the automated response system, and update FAQs.

[0685] Example prompt

[0686] An example of a prompt sentence could be "What is the system's procedure when a user asks, 'Is the product in stock?'"

[0687] This enables the system to respond to user inquiries quickly and accurately, enabling efficient inquiry management and future service improvements through the use of data.

[0688] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0689] Step 1:

[0690] The user inputs an inquiry into the terminal and presses the send button. At this time, the user inputs an inquiry such as "Please tell me the inventory status of the product." The input data is the inquiry content (e.g., "Do you have product A in stock?").

[0691] Step 2:

[0692] The device structures the query and sends it to the server via API. Specifically, it converts the query into JSON format and sends it using an HTTP POST request. The input is the user's query, and the output is JSON format data.

[0693] Step 3:

[0694] The server temporarily saves the query received from the terminal in a database. Specifically, it inserts it into the "ReceivedQueries" table. The input is the query in JSON format, and the output is the data saved in the database.

[0695] Step 4:

[0696] The server passes the received query content to a natural language processing (NLP) model to analyze the content. For example, it uses TextBlob, spaCy, or a custom NLP model (e.g., the BERT model). The input is the stored query data, and the output is the analysis result (e.g., the intent "check inventory").

[0697] Step 5:

[0698] The server determines the intent of the query based on the analysis results of the NLP model. The input is the analysis results from the NLP model, and the output is the determined intent (e.g., "check inventory").

[0699] Step 6:

[0700] The server retrieves the relevant information from the database based on the determined intent. For example, SELECT the inventory information of a specific product from the "Inventory" table. The input is the determined intent, and the output is the retrieved inventory information (e.g., "15 units in stock").

[0701] Step 7:

[0702] The server generates a response based on the acquired information. For example, it generates a sentence such as "Currently, there are 15 units of product A in stock." The input is the acquired inventory information, and the output is the generated response sentence.

[0703] Step 8:

[0704] The server sends the generated answer to the terminal in JSON format as an HTTP response. The input is the generated answer text, and the output is the JSON format response.

[0705] Step 9:

[0706] The device displays the received answer to the user, for example, in a UI component of a browser or mobile app. The input is a JSON-formatted response, and the output is the answer text that is displayed to the user.

[0707] Step 10:

[0708] The server stores the query and generated response in a database permanently. Specifically, it inserts the query and generated response into the "QueriesAndResponses" table. The input is the query and generated response, and the output is the record stored in the database.

[0709] Step 11:

[0710] The server periodically analyzes the accumulated data using Python's Pandas library and SQL queries. The input is the accumulated inquiry and response data, and the output is the analysis results and trend analysis.

[0711] Step 12:

[0712] The server uses the analysis results to predict future inquiries and builds a system that proactively provides relevant information. For example, a machine learning model can be used to predict inquiry trends, improve the automated response system, and update FAQs. The input is the analysis results, and the output is the predictive model and an improved system.

[0713] (Application example 1)

[0714] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0715] Conventional inquiry response systems often required manual response, resulting in the problem of long response times. It was also difficult to provide an appropriate response immediately based on the inquiry, which could lead to a decline in customer satisfaction. Furthermore, because data was not centrally managed, it was difficult to later analyze inquiries and their responses.

[0716] 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.

[0717] In this invention, the server includes means for using a natural language processing model to identify the intent of the inquiry, means for retrieving information from a database based on the intent of the inquiry, and means for returning the generated answer in JSON format. This enables the provision of an immediate and appropriate answer to the inquiry, and furthermore, centralized management and analysis of data enables the efficiency of subsequent inquiry responses and improvement of customer satisfaction.

[0718] The "means for receiving an inquiry" is a system that has the function of receiving an inquiry from a user via a digital device and transmitting the content of the inquiry to a server.

[0719] The "means for analyzing the received inquiry content" is a system that understands the inquiry content and analyzes it to determine its meaning and intent.

[0720] The "means for generating a response based on the analyzed inquiry content" is a system that has the function of automatically generating an appropriate response based on the analysis results.

[0721] The "means for transmitting the generated answer" is a system having a function for returning and displaying the generated answer to the user.

[0722] The "means for saving the inquiry content and the generated response in a database" is a system that has the function of recording and saving the inquiry content and the response to the inquiry in a database.

[0723] The "means for analyzing stored data" is a system that analyzes the history of inquiries and responses stored in a database and predicts and optimizes future inquiries.

[0724] The "means of using a natural language processing model to identify the intent of a query" is a system that has the function of using natural language processing technology to identify the intent of a query from its content.

[0725] The "means for retrieving information from a database based on the intent of a query" is a system that has the function of searching and retrieving related information from a database in accordance with the specified intent.

[0726] "Means for returning the generated answer in JSON format" refers to a system that has the function of returning the generated answer to the user as JSON format data.

[0727] This invention relates to a system that automatically receives and analyzes inquiries from users, and generates and transmits answers. In this system, the roles of the server, terminal, and user are clarified, and specific implementation means are described below.

[0728] System Configuration

[0729] This system consists of the following main components:

[0730] 1. Means of receiving inquiries

[0731] 2. A means of analyzing the content of the received inquiry (using a natural language processing model)

[0732] 3. A means of generating answers based on the parsed query content

[0733] 4. How to send the generated answer (returned in JSON format)

[0734] 5. A means of storing queries and generated responses in a database

[0735] 6. Means of analyzing stored data

[0736] Hardware and software used

[0737] Server: High-performance cloud server (e.g. AWS, Google Cloud)

[0738] Device: The user's smartphone, tablet, or PC

[0739] Database: SQLite or MySQL

[0740] Natural language processing models: SpaCy, Transformers library (e.g., Hugging Face transformers)

[0741] Framework: Python, Flask (Web application framework)

[0742] Program processing overview

[0743] The server receives queries sent from the user's device via an API and temporarily stores them in a database. It then passes the queries to a natural language processing model (e.g., SpaCy or Transformers) to analyze them and identify their intent. It then retrieves relevant information from the database based on the analysis results and generates an appropriate answer. This answer is sent to the user's device in JSON format. The query and the generated answer are also stored in a database for use in subsequent data analysis.

[0744] Specific examples

[0745] For product inquiries:

[0746] A user makes a query through a smartphone application, asking, "Please tell me the stock of product A." This query is sent from the device to the server, which temporarily stores the query in a database. Using a natural language processing model, the intent "to check stock" is identified and the stock information of product A is retrieved from the database. Finally, the server generates a response saying, "There are currently 15 units of product A in stock," and sends it back to the user's device in JSON format.

[0747] Example prompt sentence:

[0748] "Please let me know the stock of product A."

[0749] For shipping inquiries:

[0750] A user makes a query from their PC asking, "Please tell me the delivery status of product B." This query is sent from the device to the server, which temporarily stores the query in a database. Using a natural language processing model, the intent "check delivery status" is identified and the delivery information for product B is retrieved from the database. Finally, the server generates a response stating, "Product B is currently being delivered. The expected delivery date is next Tuesday," and sends it back to the user's device in JSON format.

[0751] Example prompt sentence:

[0752] "Please let me know the delivery status of product B."

[0753] This will enable immediate and appropriate responses to inquiries, and through centralized data management and analysis, will improve customer satisfaction and make future inquiries more efficient.

[0754] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0755] Step 1:

[0756] The user enters an inquiry from their device (smartphone, tablet, or PC) and presses the send button. At this time, the entered inquiry content is sent to the server as an API request.

[0757] Input: The inquiry entered by the user (e.g., "Please tell me the inventory of product A.")

[0758] Output: Query data in API request format

[0759] Step 2:

[0760] The terminal sends the received inquiry to the server via API and temporarily stores it in a database.

[0761] Input: Inquiry data in API request format

[0762] Output: A database containing the query results

[0763] Specific operation: The server receives a request from a terminal and records the contents in a specific table in the database.

[0764] Step 3:

[0765] The server passes the query content stored in the database to a natural language processing (NLP) model to analyze the content.

[0766] Input: Inquiry details saved in the database

[0767] Output: Analysis results from the natural language processing model (intent and extracted entities)

[0768] Specific operation: The server passes the query content to a natural language processing model (e.g., SpaCy or Hugging Face's transformers model), determines the intent as "check inventory," etc., and extracts entities such as product names.

[0769] Step 4:

[0770] The server determines the intent of the query based on the analysis results of the NLP model and retrieves the relevant information from the database.

[0771] Input: Analysis results from natural language processing model

[0772] Output: Result of retrieving related information (e.g., inventory quantity and delivery status)

[0773] Specific behavior: The server generates and executes a query to the database based on the identified intent, and retrieves the relevant data (e.g., product inventory and delivery status).

[0774] Step 5:

[0775] The server generates a response based on the acquired information and sends it to the user's terminal.

[0776] Input: Related information (e.g., inventory quantity and delivery status)

[0777] Output: Generated answer (e.g. "There are 15 units of product A in stock.")

[0778] Specific operation: The server uses the acquired data to create a corresponding answer using a template or plain text generation, and returns it to the user's device in JSON format.

[0779] Step 6:

[0780] The terminal displays the received response to the user.

[0781] Input: Response data sent from the server

[0782] Output: User-visible answers

[0783] Specific operation: The terminal analyzes the received JSON formatted response data and displays it in a format that is easy for the user to understand via the user interface.

[0784] Step 7:

[0785] The server stores the final query and response in a database for later data analysis.

[0786] Input: Final query and generated answer

[0787] Output: A record of the query and answer stored in a database

[0788] Specific operation: The server records pairs of queries and their answers in a database and uses them as data for future trend analysis and system improvement.

[0789] Example prompt sentence:

[0790] "Please let me know the stock of product A."

[0791] 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.

[0792] This invention provides a system that automatically receives and analyzes user inquiries, generates appropriate responses, and returns them with an emotion engine that recognizes the user's emotions, enabling even more sophisticated responses. This system is primarily composed of a server, terminals, and users. Below are the system's program processing and specific examples.

[0793] System Configuration

[0794] This system consists of the following main components:

[0795] 1. Means of receiving inquiries

[0796] 2. Means of analyzing the content of received inquiries

[0797] 3. A means of generating answers based on the parsed query content

[0798] 4. A means of sending the generated answer

[0799] 5. A means of storing queries and generated responses in a database

[0800] 6. Means of analyzing stored data

[0801] 7. Emotion engine that recognizes user emotions

[0802] Program processing

[0803] How to receive inquiries

[0804] The user inputs a query using their own terminal and presses the send button, which sends the query to the server.

[0805] The terminal sends the entered inquiry to the server via API.

[0806] A means of analyzing the content of received inquiries

[0807] The server temporarily stores the inquiry received from the terminal in a database.

[0808] The server passes the received query content to a natural language processing (NLP) model to analyze the content.

[0809] Emotion engine that recognizes user emotions

[0810] The server passes the data, along with the query content, to an emotion engine for analyzing the user's emotions.

[0811] The emotion engine analyzes emotions from the user's text and returns the results to the server, identifying emotions such as "anger," "joy," and "sadness."

[0812] A means of generating answers based on the parsed query content

[0813] The server determines the intent of the query and the user's emotions based on the analysis results of the NLP model and the emotion engine.

[0814] For example, if a user inquires, "Please tell me the product's stock status," the server determines the intent as "checking stock" and the emotion as "neutral."

[0815] A means of sending the generated answer

[0816] Based on the determined intent and emotion, the server retrieves relevant information from a database and generates a response in an appropriate tone.

[0817] The server generates a response based on the acquired information and sends it to the user's terminal.

[0818] The terminal displays the received response to the user.

[0819] A means of storing queries and generated responses in a database

[0820] The server permanently stores the query content and the results of sentiment analysis in a database, along with the final answer.

[0821] A means of analyzing stored data

[0822] The server periodically analyzes the accumulated inquiry, response, and sentiment analysis data to extract inquiry trends and patterns.

[0823] Based on the analysis results, we will build a system that proactively provides relevant information that responds to future inquiries and emotional states.

[0824] Specific examples

[0825] Product inquiries

[0826] User: The user enters "Is product A in stock?" and presses the send button.

[0827] Terminal: The terminal sends this query to the server.

[0828] Server: The server temporarily stores the query content in a database and passes it to the natural language processing model.

[0829] Server: The NLP model determines the intent, "Check inventory," and the emotion engine determines the emotion, "Neutral."

[0830] Server: The server retrieves the stock information of product A from the database, generates a response in a neutral tone such as "Currently, there are 15 units of product A in stock," and sends it to the user's device.

[0831] Terminal: The terminal displays this answer to the user.

[0832] Server: The final query, response, and sentiment analysis results are stored in a database for later data analysis.

[0833] Emotion-conscious response to inquiries

[0834] User: The user types, "The delivery of product B is delayed. What's going on?" and submits.

[0835] Terminal: The terminal sends this query to the server.

[0836] Server: The server temporarily stores the query content in a database and passes it to the natural language processing model.

[0837] Server: The NLP model determines the intent, "check delivery status," and the emotion engine determines the emotion, "anger."

[0838] Server: The server retrieves the delivery status of Product B from the delivery system database, generates a polite response saying, "Product B is currently being delivered. We apologize for any inconvenience. The expected delivery date is next Tuesday," and sends it to the user's device.

[0839] Terminal: The terminal displays this answer to the user.

[0840] Server: The final query, response, and sentiment analysis results are stored in a database for later data analysis.

[0841] This allows the system to respond to inquiries taking emotions into consideration, contributing to improved customer satisfaction.

[0842] The processing flow will be explained below.

[0843] Step 1:

[0844] User: Enter the inquiry details and click the send button.

[0845] For example, enter "Please let me know the product availability."

[0846] Step 2:

[0847] Terminal: Sends the entered inquiry to the server.

[0848] Specifically, the inquiry data is sent via the API using an HTTP POST request.

[0849] Step 3:

[0850] Server: Temporarily stores the received inquiry in a database.

[0851] Specifically, the query content is recorded in the query table using an SQL insert statement.

[0852] Example: INSERT INTO inquiries (user_id, inquiry_text) VALUES ('1234', 'What is the product availability?')

[0853] Step 4:

[0854] Server: Passes the received query content to a natural language processing (NLP) model for analysis.

[0855] Specifically, the query is passed to the NLP model API and the analysis results are received in JSON format.

[0856] Example: "Product availability" is parsed as "Intent: Check inventory."

[0857] Step 5:

[0858] Server: Passes the query content to the emotion engine and analyzes the user's emotions.

[0859] Specifically, the query is passed to the emotion engine's API, and the results of the emotion analysis are received in JSON format.

[0860] Example: "What is the product availability?" is parsed as "Neutral."

[0861] Step 6:

[0862] Server: Determines the intent of the query and the user's emotions based on the analysis results of the NLP model and the emotion engine.

[0863] For example, "Intention: Check inventory" and "Emotion: Neutral" are determined.

[0864] Step 7:

[0865] Server: Based on the intent of the query, retrieves the necessary information from the database.

[0866] Specifically, an SQL query is executed to obtain the stock quantity of the relevant product from the inventory table.

[0867] Example: SELECT stock_quantity FROM products WHERE product_name = 'Product A'

[0868] Step 8:

[0869] Server: Generates answers based on the acquired information and the user's sentiment.

[0870] For example, frame your response in a neutral tone, saying, "We currently have 15 units of product A in stock."

[0871] Step 9:

[0872] Server: Sends the generated answer to the user's device.

[0873] Specifically, the answer is sent as an HTTP response.

[0874] Step 10:

[0875] Terminal: Displays the received answer to the user.

[0876] For example, the chat screen in your browser or app might display "Currently, there are 15 units of product A in stock."

[0877] Step 11:

[0878] Server: The inquiry content and sentiment analysis results, along with the final answer, are permanently stored in a database.

[0879] Use SQL insert or update statements to store queries and their answers.

[0880] Example: UPDATE inquiries SET response = 'Currently, there are 15 units of product A in stock', emotion = 'Neutral' WHERE inquiry_id = '5678'

[0881] Step 12:

[0882] Server: Periodically analyzes the accumulated data to identify trends and patterns in inquiries.

[0883] Specifically, data analysis tools such as Python and R are used to analyze past inquiries in terms of distribution and frequency, as well as emotional trends.

[0884] The above is the specific flow of processing the inquiry content, generating and providing an answer taking the user's feelings into consideration. This system will improve the efficiency of inquiry responses and contribute to improving customer satisfaction.

[0885] Example 2

[0886] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0887] Conventional inquiry systems only provided standardized answers without considering the user's feelings, resulting in low user satisfaction and the inability to respond appropriately to complaints and urgent inquiries in particular. In addition, the stored data was treated as a simple log, and could not be used to proactively respond to future inquiries or for improvements.

[0888] 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.

[0889] In this invention, the server includes means for receiving an inquiry, means for analyzing the content of the received inquiry, means for recognizing the user's emotion, means for generating an answer based on the analyzed content of the inquiry and the recognized emotion, means for transmitting the generated answer, means for saving the content of the inquiry and the generated answer in a database, and means for analyzing the saved data. This makes it possible to generate an appropriate answer that takes the user's emotion into consideration, and further makes it possible to proactively respond to future inquiries by utilizing the accumulated data.

[0890] The "means for receiving an inquiry" is a device or process for receiving the content of an inquiry sent from a user to the server via the terminal.

[0891] "Means for analyzing the received query content" refers to a device or process for understanding the received query text and identifying its intent, often using natural language processing (NLP) techniques.

[0892] The "means for recognizing user emotions" is a device or process for determining the user's emotions from the query text, and often utilizes an emotion analysis engine.

[0893] The "means for generating an answer based on the analyzed query content and the recognized sentiment" is a device or process for generating an appropriate answer based on the intent of the query and the recognized sentiment.

[0894] A "means for transmitting a generated answer" is a device or process for transmitting a generated answer to a user.

[0895] The "means for storing the inquiry contents and generated answers in a database" is a device or process for storing all the inquiry contents and the generated answers thereto in a database.

[0896] A "means for analyzing stored data" is a device or process that uses stored inquiry and response data to extract trends and patterns to improve future inquiry responses.

[0897] "Natural Language Processing (NLP)" is a technology that enables machines to understand, analyze, and generate human language.

[0898] An "emotion analysis engine" is an algorithm or software for identifying a user's emotions from text data.

[0899] A "database" is an information storage system for storing queries and generated responses.

[0900] This invention enables more advanced responses by combining a system that receives and analyzes user inquiries, generates appropriate responses, and responds with an emotion engine that recognizes the user's emotions. The system consists of the following main components:

[0901] System Configuration

[0902] This system is mainly composed of a server, terminals, and users. Specifically, it uses the following devices and software:

[0903] server

[0904] Hardware: A server machine equipped with a high-performance processor, memory, and storage

[0905] software:

[0906] Database: MySQL, PostgreSQL, MongoDB, Elasticsearch, etc.

[0907] Natural Language Processing (NLP) libraries: spaCy, NLTK

[0908] Sentiment analysis engine: Microsoft Azure's Text Analytics API, IBM Watson's Natural Language Understanding API

[0909] Text generation model: OpenAI's GPT-3

[0910] Terminal

[0911] Hardware: Smartphones, tablets, personal computers

[0912] Software: Web browsers (Google Chrome, Safari, etc.), mobile applications

[0913] User

[0914] Interface: inquiry form, etc.

[0915] Program processing

[0916] Receive inquiries

[0917] The user uses their device (smartphone or computer) to enter the details of their inquiry. For example, they might enter "Please tell me the product's stock status" in a web browser and press the send button. The device then sends this inquiry to the server via an API (RESTful API or GraphQL).

[0918] Analyze the incoming inquiries

[0919] The server temporarily stores the query received from the device in a database (MySQL or PostgreSQL). It then uses a natural language processing engine (spaCy or NLTK) to analyze the intent of the query. For example, it classifies the text "Please tell me the stock status of the product" as the intent "Check stock."

[0920] Recognize user emotions

[0921] The server sends the received inquiry content to a sentiment analysis engine, which determines the user's emotions (such as "anger," "joy," or "sadness") from the text. Examples of sentiment analysis engines include Microsoft Azure's Text Analytics API and IBM Watson's Natural Language Understanding API.

[0922] Generate an answer

[0923] The server generates a response using a text generation model (OpenAI's GPT-3) based on the intent and sentiment analysis results. For example, it generates a neutral response such as "Currently, there are 15 units of product A in stock."

[0924] Submit your answer

[0925] The server sends the generated answer to the user's terminal, which displays the answer to the user.

[0926] Save your data

[0927] The server stores the query content and sentiment analysis results along with the final answer in a database (MongoDB or Elasticsearch).

[0928] Analyze the data

[0929] The stored data is periodically analyzed and data analysis tools such as Apache Hadoop and Google BigQuery are used to extract trends and patterns in inquiries, which can then be used to proactively respond to future inquiries.

[0930] Specific examples

[0931] Product inquiries

[0932] A user enters "Do you have product A in stock?" into an inquiry form in Google Chrome and presses the submit button. The device sends this inquiry to the server via a RESTful API. The server temporarily saves the inquiry content in MySQL and uses spaCy to extract the intent "to check stock." It then uses Microsoft Azure's Text Analytics to determine the sentiment of "neutral." It retrieves the stock information for product A from the database and generates a response in a neutral tone, such as "There are currently 15 units of product A in stock." This response is then sent back to the device in JSON format. The device displays this response on the user interface and saves the final inquiry, response, and sentiment analysis results in a MySQL database.

[0933] Inquiries about delivery status

[0934] The user types and submits the query in Safari, "The delivery of Product B is delayed. What's going on?" The device sends this query to the server using the GraphQL API. The server temporarily stores the query content in PostgreSQL and uses NLTK to extract the intent, "Check delivery status." It then uses IBM Watson's Natural Language Understanding to determine the emotion, "anger." It retrieves the delivery status of Product B from the delivery system database and generates a polite response, saying, "Product B is currently being delivered. We apologize for any inconvenience. The expected delivery date is next Tuesday." The response returned to the device in JSON format is displayed on the user interface, and the final query, response, and emotion analysis results are saved in the PostgreSQL database.

[0935] This allows the system to respond to inquiries taking into account the user's emotions, contributing to improved customer satisfaction.

[0936] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0937] Step 1: Receiving an inquiry

[0938] The user uses their own device (smartphone or computer) to enter the details of their inquiry and press the send button. Specifically, the user opens a web browser (e.g., Google Chrome), enters "Please let me know the product's stock status" into the inquiry form, and clicks send.

[0939] Input: User-supplied query text

[0940] Output: Data sent from the device to the server (inquiry details, user ID, timestamp, etc.)

[0941] The device sends the entered inquiry to the server via an API (e.g., RESTful API). At this time, the sent data includes metadata such as the user ID and timestamp.

[0942] Step 2: Analyze the query

[0943] The server temporarily stores the query received from the terminal in a database (e.g., MySQL). This temporary storage is performed as a transaction to maintain data consistency.

[0944] Input: Inquiry data sent from the terminal

[0945] Output: Temporarily saved inquiry data

[0946] The server uses a natural language processing (NLP) engine (e.g., spaCy) to analyze the query. Specifically, it extracts the intent from the query text. For example, the text "Please tell me the product's stock status" is classified as the intent "Check stock."

[0947] Step 3: Recognize the user's emotions

[0948] The server sends the received inquiry content to a sentiment analysis engine (e.g., Microsoft Azure's Text Analytics API) to analyze the user's sentiment.

[0949] Input: Inquiry details

[0950] Output: Emotion analysis result (e.g. "neutral")

[0951] The sentiment analysis engine determines the user's sentiment from the query text, for example, "Please tell me the product availability status," and identifies the sentiment "neutral."

[0952] Step 4: Generate an answer

[0953] The server generates an answer using a text generation model (e.g., OpenAI's GPT-3) based on the results of intent analysis and sentiment analysis.

[0954] Input: Intent analysis results (e.g., "Check inventory"), sentiment analysis results (e.g., "Neutral")

[0955] Output: Generated answer (e.g. "Currently, there are 15 units of product A in stock.")

[0956] The server retrieves the necessary information from a database (e.g., MongoDB) based on the intent. For example, for the intent "check stock," it retrieves stock information for product A. It then uses a text generation model to generate an answer in an appropriate tone. For example, it generates an answer such as "Currently, there are 15 units of product A in stock."

[0957] Step 5: Submit your response

[0958] The server then sends the generated response to the user's device, often in JSON format.

[0959] Input: Generated response data

[0960] Output: Data sent to the terminal

[0961] The terminal displays the received answer on the user interface, allowing the user to check the answer on their screen.

[0962] Step 6: Save your data

[0963] The server stores the final answer, inquiry details, and sentiment analysis results in a database (e.g., Elasticsearch).

[0964] Input: Final response, inquiry details, sentiment analysis results

[0965] Output: Data stored in the database

[0966] The saved data can be used for later data analysis.

[0967] Step 7: Analyze the data

[0968] The server periodically analyzes the accumulated inquiry, response, and sentiment analysis data.

[0969] Input: Data of inquiries, responses, and sentiment analysis results stored in the database

[0970] Output: Analysis results (inquiry trends and patterns)

[0971] Specifically, Apache Hadoop and Google BigQuery are used to process data and extract trends and patterns. The analysis results are then used to respond to future inquiries. For example, if there are many inquiries about a particular season or product, that information can be prepared in advance.

[0972] (Application example 2)

[0973] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0974] For modern online shopping sites, responding to user inquiries quickly and appropriately is an important issue that directly contributes to improving customer satisfaction. However, conventional inquiry response systems do not adequately take user emotions into account, resulting in inconsistent response quality. In particular, when a user is dissatisfied or angry, it is difficult to respond appropriately based on that emotion. Another challenge is efficiently analyzing the content of a large number of inquiries and generating appropriate answers.

[0975] 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.

[0976] In this invention, the server includes means for receiving an inquiry, means for analyzing the received inquiry, means for generating a response based on the analyzed inquiry, means for transmitting the generated response, means for saving the inquiry and the generated response in a database, means for analyzing the saved data, means including an emotion engine for recognizing the user's emotion, and means for adjusting the tone of the response in accordance with the user's emotion, thereby enabling a prompt and appropriate response to an inquiry that takes the user's emotion into consideration.

[0977] The "means for receiving an inquiry" is a function by which the system receives and processes an inquiry from a user.

[0978] The "means for analyzing the content of the received inquiry" is a function for understanding the content of the received inquiry and analyzing it as structured data.

[0979] The "means for generating a response based on the analyzed inquiry content" is a function that automatically generates an appropriate response based on the analysis results.

[0980] The "means for sending the generated answer" is a function for returning the generated answer to the user.

[0981] The "means for saving the inquiry and the generated response in a database" is a function for saving the inquiry and the response as a record in a database.

[0982] The "means for analyzing stored data" is a function for analyzing stored inquiry and response data and extracting significant insights and patterns.

[0983] The "emotion engine that recognizes user emotions" is an engine that analyzes and identifies emotions from the content of a user's inquiry.

[0984] The "means for adjusting the tone of the response according to the user's emotion" is a function for automatically adjusting the expression and tone of the response according to the emotion specified by the user.

[0985] The present invention is a system for responding promptly and appropriately to inquiries from users on an online shopping site. This system analyzes the content of the user's inquiry and recognizes their emotions, thereby generating a response in a tone that corresponds to their emotions and improving the quality of the response.

[0986] System Configuration

[0987] The system mainly consists of the following components:

[0988] 1. Means of receiving inquiries

[0989] 2. Means of analyzing the content of received inquiries

[0990] 3. A means of generating answers based on the parsed query content

[0991] 4. A means of sending the generated answer

[0992] 5. A means of storing queries and generated responses in a database

[0993] 6. Means of analyzing stored data

[0994] 7. Emotion engine that recognizes user emotions

[0995] 8. How to adjust the tone of your response based on the user's emotions

[0996] Hardware and Software

[0997] Hardware used:

[0998] server

[0999] User device (smartphone or PC)

[1000] Software used:

[1001] Python

[1002] Natural Language Processing Library (Hugging Face Transformers)

[1003] Sentiment analysis model (BERT)

[1004] Database System

[1005] Data processing and calculation

[1006] 1. Means of receiving inquiries:

[1007] The user uses a smartphone or computer to enter and submit an inquiry.

[1008] The terminal sends the entered inquiry to the server via API.

[1009] 2. How to analyze the received inquiry:

[1010] The server temporarily stores the inquiry received from the terminal in a database.

[1011] The data stored in the database is analyzed using natural language processing models.

[1012] 3. Emotion engine that recognizes user emotions:

[1013] The server passes the data along with the query to the emotion engine.

[1014] The emotion engine analyzes the emotion from the query and returns the result to the server. The emotion may be "anger," "joy," or "sadness," for example.

[1015] 4. Means of generating an answer based on the parsed query:

[1016] The server determines the intent of the query and the user's emotions based on the analysis results of the NLP model and the emotion engine.

[1017] 5. Means for sending generated answers:

[1018] The server generates a response in an appropriate tone based on the determined intent and emotion.

[1019] The generated answer is sent to the user's terminal.

[1020] 6. How to store queries and generated responses in a database:

[1021] The inquiry details and the results of sentiment analysis are stored in a database along with the final answer.

[1022] Examples of specific examples and prompts

[1023] For example, if a user asks, "Product B hasn't arrived yet. What's going on?", the following process will be performed:

[1024] User inquiry: "I haven't received my delivery of product B yet. What's going on?"

[1025] Emotion engine judgement: "Anger"

[1026] System response: "I see you're angry. We'll deal with it right away."

[1027] Example prompt sentence:

[1028] User Question: "I haven't received my delivery of item B yet, what's going on?"

[1029] System emotion detection: "Anger"

[1030] Generate an appropriate response.

[1031] This allows for prompt and appropriate response to inquiries that take into account the user's feelings.

[1032] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1033] Step 1:

[1034] The user enters an inquiry on a smartphone or PC. When the user enters the inquiry content as text and presses the send button, the device sends the inquiry data to the server via the API.

[1035] Input: The query text entered by the user.

[1036] Output: The query data sent to the server.

[1037] Step 2:

[1038] The server temporarily stores the received inquiry in a database.

[1039] Input: Enquiry data received through API.

[1040] Output: Query data temporarily stored in a database.

[1041] Step 3:

[1042] The server passes the received query content to a natural language processing (NLP) model for analysis, which analyzes the text and extracts the intent of the query.

[1043] Input: Query data stored in the database.

[1044] Output: Analysis result data including query intent.

[1045] Step 4:

[1046] The server passes the query content and sentiment analysis data to the sentiment engine, which analyzes the sentiment from the user's text and returns the results to the server.

[1047] Input: Query text.

[1048] Output: Emotion analysis results from the emotion engine (e.g., "anger," "joy," "sadness," etc.).

[1049] Step 5:

[1050] The server determines the intent of the query and the user's emotions based on the analysis results of the NLP model and the emotion engine.

[1051] Input: Analysis results of the NLP model and the emotion engine.

[1052] Output: Analysis result data including intent and sentiment.

[1053] Step 6:

[1054] Based on the determined intent and emotion, the server retrieves relevant information from a database and generates a response in an appropriate tone.

[1055] Input: Analysis result data including intent and sentiment.

[1056] Output: Answer text generated in the appropriate tone.

[1057] Step 7:

[1058] The server sends the generated answer to the user's terminal, which then displays the received answer to the user.

[1059] Input: The generated answer text.

[1060] Output: The answer text that is displayed on the user's terminal.

[1061] Step 8:

[1062] The server stores the query and the results of the sentiment analysis in a database, along with the final answer.

[1063] Input: Query, generated answer, sentiment analysis results.

[1064] Output: The final data stored in the database.

[1065] 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.

[1066] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1067] 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.

[1068] [Third embodiment]

[1069] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1070] 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.

[1071] 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).

[1072] 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.

[1073] 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.

[1074] 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).

[1075] 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. 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.

[1076] 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.

[1077] 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.

[1078] 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.

[1079] 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.

[1080] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[1081] The system of the present invention responds quickly and efficiently by automatically receiving and analyzing inquiries from users and generating and returning appropriate answers. This system is primarily composed of a server, terminals, and users. The following describes the processing of the system program and a specific example.

[1082] System Configuration

[1083] This system consists of the following main components:

[1084] 1. Means of receiving inquiries

[1085] 2. Means of analyzing the content of received inquiries

[1086] 3. A means of generating answers based on the parsed query content

[1087] 4. A means of sending the generated answer

[1088] 5. A means of storing queries and generated responses in a database

[1089] 6. Means of analyzing stored data

[1090] Program processing

[1091] How to receive inquiries

[1092] The user inputs a query using their own terminal and presses the send button, which sends the query to the server.

[1093] The terminal sends the entered inquiry to the server via API.

[1094] A means of analyzing the content of received inquiries

[1095] The server temporarily stores the inquiry received from the terminal in a database.

[1096] The server passes the received query content to a natural language processing (NLP) model to analyze the content.

[1097] A means of generating answers based on the parsed query content

[1098] The server determines the intent of the query based on the analysis results of the NLP model.

[1099] For example, if a user inquires, "Please tell me the stock status of the product," the server determines that the user's intent is "to check stock."

[1100] A means of sending the generated answer

[1101] The server retrieves the relevant information from the database based on the determined intent.

[1102] The server generates a response based on the acquired information and sends it to the user's terminal.

[1103] The terminal displays the received response to the user.

[1104] A means of storing queries and generated responses in a database

[1105] The server permanently stores the final answer together with the inquiry details in a database.

[1106] A means of analyzing stored data

[1107] The server periodically analyzes the accumulated inquiry and response data to extract inquiry trends and patterns.

[1108] Based on the analysis results, we will build a system that predicts future inquiries and provides relevant information in advance.

[1109] Specific examples

[1110] Product inquiries

[1111] User: The user enters "Is product A in stock?" and presses the send button.

[1112] Terminal: The terminal sends this query to the server.

[1113] Server: The server temporarily stores the query content in a database and passes it to the natural language processing model.

[1114] Server: The NLP model determines the intent to "check stock," and the server retrieves stock information for product A from the database.

[1115] Server: Generates a response saying "Currently, there are 15 units of product A in stock" and sends it to the user's device.

[1116] Terminal: The terminal displays this answer to the user.

[1117] Server: The final query and answer are stored in a database for later data analysis.

[1118] Delivery inquiries

[1119] User: The user enters "Please tell me the delivery status of product B" and submits.

[1120] Terminal: The terminal sends this query to the server.

[1121] Server: The server temporarily stores the query content in a database and passes it to the natural language processing model.

[1122] Server: The NLP model determines the intent to "check delivery status," and the server retrieves the delivery status of Product B from the delivery system database.

[1123] Server: Generates a response stating, "Item B is currently being delivered. The estimated delivery date is next Tuesday," and sends it to the user's device.

[1124] Terminal: The terminal displays this answer to the user.

[1125] Server: The final query and answer are stored in a database for later data analysis.

[1126] This allows the system to automate inquiries efficiently and accurately, and the accumulated data can be used to improve customer service in the future.

[1127] The processing flow will be explained below.

[1128] Step 1:

[1129] User: Enter the inquiry details and click the send button.

[1130] For example, enter "Please let me know the product availability."

[1131] Step 2:

[1132] Terminal: Sends the entered inquiry to the server.

[1133] Specifically, the inquiry data is sent via the API using an HTTP POST request.

[1134] Step 3:

[1135] Server: Temporarily stores the received inquiry in a database.

[1136] Specifically, the query content is recorded in the query table using an SQL insert statement.

[1137] INSERT INTO inquiries (user_id, inquiry_text) VALUES ('1234', 'What is the product availability?')

[1138] Step 4:

[1139] Server: Passes the received query content to a natural language processing (NLP) model for analysis.

[1140] Specifically, the query is passed to the NLP model API and the analysis results are received in JSON format.

[1141] Example: "Product availability" is parsed as "Intent: Check inventory."

[1142] Step 5:

[1143] Server: Determines the intent of the query based on the analysis results of the NLP model.

[1144] Based on the analysis results, the corresponding processing is selected.

[1145] Example: Determine the intent "check inventory."

[1146] Step 6:

[1147] Server: Based on the intent of the query, retrieves the necessary information from the database.

[1148] Specifically, an SQL query is executed to obtain the stock quantity of the relevant product from the inventory table.

[1149] SELECT stock_quantity FROM products WHERE product_name = 'A product'

[1150] Step 7:

[1151] Server: Generates an answer based on the information obtained.

[1152] For example, construct an answer such as "Currently, there are 15 units of product A in stock."

[1153] Step 8:

[1154] Server: Sends the generated answer to the user's device.

[1155] Specifically, the answer is sent as an HTTP response.

[1156] Step 9:

[1157] Terminal: Displays the received answer to the user.

[1158] For example, the chat screen in your browser or app might display "Currently, there are 15 units of product A in stock."

[1159] Step 10:

[1160] Server: The query is permanently stored in a database along with the final answer.

[1161] Use SQL insert or update statements to store queries and their answers.

[1162] UPDATE inquiries SET response = 'Currently, there are 15 units of product A in stock' WHERE inquiry_id = '5678'

[1163] Step 11:

[1164] Server: Periodically analyzes the accumulated data to identify trends and patterns in inquiries.

[1165] Specifically, data analysis tools such as Python and R are used to analyze past inquiries in terms of distribution and frequency.

[1166] The above is the specific flow of processing from the inquiry content to returning a response to the user. This system improves the efficiency of inquiry responses and contributes to improving customer satisfaction.

[1167] Example 1

[1168] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1169] Conventional inquiry response systems take time to analyze the content of inquiries and generate responses, making it difficult to respond to user inquiries quickly. Furthermore, there was no established method for effectively utilizing accumulated data, making it difficult to predict future inquiries or provide appropriate information. This resulted in a lack of improvement in the user experience and a decline in the quality of a company's services.

[1170] 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.

[1171] In this invention, the server includes means for receiving an inquiry, means for analyzing the received inquiry, means for generating a response based on the analyzed inquiry, means for transmitting the generated response, means for storing the inquiry and the generated response in an information storage medium, and means for analyzing the stored data. This makes it possible to quickly and accurately analyze the inquiry and generate and return an appropriate response. Furthermore, by effectively utilizing the accumulated data, it is possible to predict future inquiries and provide useful information, thereby significantly improving the user experience.

[1172] The "means for receiving an inquiry" is a device or software that has the function of transmitting an inquiry from a user to a server as digital data.

[1173] "Means for analyzing the content of the received inquiry" refers to a device or software that has the function of analyzing the received digital data using methods such as natural language processing and understanding its intent and content.

[1174] The "means for generating a response based on the analyzed inquiry content" is a device or software that has the function of automatically generating an appropriate response based on the analysis results.

[1175] The "means for transmitting the generated answer" is a device or software that has the function of transmitting the generated answer to the user's terminal.

[1176] The "means for storing the inquiry content and the generated response in an information storage medium" refers to a device or software that has the function of recording the inquiry content and the response in digital form and storing it permanently.

[1177] "Means for analyzing stored data" refers to devices or software that have the ability to analyze accumulated data using statistical or machine learning techniques and extract trends and patterns.

[1178] "Natural language processing" is a technology that enables computers to understand and analyze human language, and involves processing text data using language models and algorithms.

[1179] A "terminal" is a device that allows a user to input and send an inquiry, and includes a computer, smartphone, tablet, etc.

[1180] An "information storage medium" is a device or medium for storing digital data for a long period of time, and includes hard disk drives (HDDs) and solid-state drives (SSDs).

[1181] The present invention relates to a system for improving the efficiency of an inquiry response process. The system is composed of a user, a terminal, and a server, and has the function of automating the reception, analysis, response generation, and response transmission of an inquiry. The program processing of the system and a specific embodiment are described below.

[1182] System Configuration

[1183] The system consists of the following main components:

[1184] 1. Means of receiving inquiries

[1185] 2. Means of analyzing the content of received inquiries

[1186] 3. A means of generating answers based on the parsed query content

[1187] 4. A means of sending the generated answer

[1188] 5. Means for storing the inquiry content and the generated response in an information storage medium

[1189] 6. Means of analyzing stored data

[1190] Receiving inquiries

[1191] The user enters an inquiry into their device and presses the send button. For example, they may enter "Please tell me the product's stock status."

[1192] The device sends the input query to the server via API, structuring the query in JSON format and sending it to the server using an HTTP POST request.

[1193] Saving inquiry details

[1194] The server temporarily stores the query received from the terminal in an information storage medium. Specifically, the server inserts the received query into the "ReceivedQueries" table.

[1195] Analysis of inquiry content

[1196] The server passes the received query content to a natural language processing (NLP) model to analyze the content, for example using TextBlob, spaCy, or a custom NLP model (e.g., the BERT model).

[1197] The server determines the intent of the query based on the output of the NLP model. For example, if a user asks, "Please tell me the stock status of a product," the NLP model determines the intent as "check stock."

[1198] Generate answers

[1199] The server retrieves the relevant information from the information storage medium based on the determined intention. For example, it selects the inventory information of the relevant product from the "Inventory" table.

[1200] The server generates an answer based on the information it has obtained, for example, constructing a sentence such as "There are currently 15 units of this product in stock." It may also generate sentences using templates.

[1201] Submit your answer

[1202] The server sends the generated answer to the user's device in JSON format as an HTTP response.

[1203] The device displays the received response to the user, for example, in a browser or a mobile app UI component.

[1204] Data storage

[1205] The server permanently stores the query and the generated response in an information storage medium. Specifically, the columns to be inserted into the "QueriesAndResponses" table include "user_id", "query", "response", and "timestamp".

[1206] Analyzing the data

[1207] The server periodically analyzes the accumulated query and response data, for example, using Python's Pandas library or SQL queries.

[1208] The server will use the analysis results to predict future inquiries and proactively provide relevant information. For example, machine learning models can be used to predict inquiry trends, improve the automated response system, and update FAQs.

[1209] Example prompt

[1210] An example of a prompt sentence could be "What is the system's procedure when a user asks, 'Is the product in stock?'"

[1211] This enables the system to respond to user inquiries quickly and accurately, enabling efficient inquiry management and future service improvements through the use of data.

[1212] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1213] Step 1:

[1214] The user inputs an inquiry into the terminal and presses the send button. At this time, the user inputs an inquiry such as "Please tell me the inventory status of the product." The input data is the inquiry content (e.g., "Do you have product A in stock?").

[1215] Step 2:

[1216] The device structures the query and sends it to the server via API. Specifically, it converts the query into JSON format and sends it using an HTTP POST request. The input is the user's query, and the output is JSON format data.

[1217] Step 3:

[1218] The server temporarily saves the query received from the terminal in a database. Specifically, it inserts it into the "ReceivedQueries" table. The input is the query in JSON format, and the output is the data saved in the database.

[1219] Step 4:

[1220] The server passes the received query content to a natural language processing (NLP) model to analyze the content. For example, it uses TextBlob, spaCy, or a custom NLP model (e.g., the BERT model). The input is the stored query data, and the output is the analysis result (e.g., the intent "check inventory").

[1221] Step 5:

[1222] The server determines the intent of the query based on the analysis results of the NLP model. The input is the analysis results from the NLP model, and the output is the determined intent (e.g., "check inventory").

[1223] Step 6:

[1224] The server retrieves the relevant information from the database based on the determined intent. For example, SELECT the inventory information of a specific product from the "Inventory" table. The input is the determined intent, and the output is the retrieved inventory information (e.g., "15 units in stock").

[1225] Step 7:

[1226] The server generates a response based on the acquired information. For example, it generates a sentence such as "Currently, there are 15 units of product A in stock." The input is the acquired inventory information, and the output is the generated response sentence.

[1227] Step 8:

[1228] The server sends the generated answer to the terminal in JSON format as an HTTP response. The input is the generated answer text, and the output is the JSON format response.

[1229] Step 9:

[1230] The device displays the received answer to the user, for example, in a UI component of a browser or mobile app. The input is a JSON-formatted response, and the output is the answer text that is displayed to the user.

[1231] Step 10:

[1232] The server stores the query and generated response in a database permanently. Specifically, it inserts the query and generated response into the "QueriesAndResponses" table. The input is the query and generated response, and the output is the record stored in the database.

[1233] Step 11:

[1234] The server periodically analyzes the accumulated data using Python's Pandas library and SQL queries. The input is the accumulated inquiry and response data, and the output is the analysis results and trend analysis.

[1235] Step 12:

[1236] The server uses the analysis results to predict future inquiries and builds a system that proactively provides relevant information. For example, a machine learning model can be used to predict inquiry trends, improve the automated response system, and update FAQs. The input is the analysis results, and the output is the predictive model and an improved system.

[1237] (Application example 1)

[1238] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1239] Conventional inquiry response systems often required manual response, resulting in the problem of long response times. It was also difficult to provide an appropriate response immediately based on the inquiry, which could lead to a decline in customer satisfaction. Furthermore, because data was not centrally managed, it was difficult to later analyze inquiries and their responses.

[1240] 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.

[1241] In this invention, the server includes means for using a natural language processing model to identify the intent of the inquiry, means for retrieving information from a database based on the intent of the inquiry, and means for returning the generated answer in JSON format. This enables the provision of an immediate and appropriate answer to the inquiry, and furthermore, centralized management and analysis of data enables the efficiency of subsequent inquiry responses and improvement of customer satisfaction.

[1242] The "means for receiving an inquiry" is a system that has the function of receiving an inquiry from a user via a digital device and transmitting the content of the inquiry to a server.

[1243] The "means for analyzing the received inquiry content" is a system that understands the inquiry content and analyzes it to determine its meaning and intent.

[1244] The "means for generating a response based on the analyzed inquiry content" is a system that has the function of automatically generating an appropriate response based on the analysis results.

[1245] The "means for transmitting the generated answer" is a system having a function for returning and displaying the generated answer to the user.

[1246] The "means for saving the inquiry content and the generated response in a database" is a system that has the function of recording and saving the inquiry content and the response to the inquiry in a database.

[1247] The "means for analyzing stored data" is a system that analyzes the history of inquiries and responses stored in a database and predicts and optimizes future inquiries.

[1248] The "means of using a natural language processing model to identify the intent of a query" is a system that has the function of using natural language processing technology to identify the intent of a query from its content.

[1249] The "means for retrieving information from a database based on the intent of a query" is a system that has the function of searching and retrieving related information from a database in accordance with the specified intent.

[1250] "Means for returning the generated answer in JSON format" refers to a system that has the function of returning the generated answer to the user as JSON format data.

[1251] This invention relates to a system that automatically receives and analyzes inquiries from users, and generates and transmits answers. In this system, the roles of the server, terminal, and user are clarified, and specific implementation means are described below.

[1252] System Configuration

[1253] This system consists of the following main components:

[1254] 1. Means of receiving inquiries

[1255] 2. A means of analyzing the content of the received inquiry (using a natural language processing model)

[1256] 3. A means of generating answers based on the parsed query content

[1257] 4. How to send the generated answer (returned in JSON format)

[1258] 5. A means of storing queries and generated responses in a database

[1259] 6. Means of analyzing stored data

[1260] Hardware and software used

[1261] Server: High-performance cloud server (e.g. AWS, Google Cloud)

[1262] Device: The user's smartphone, tablet, or PC

[1263] Database: SQLite or MySQL

[1264] Natural language processing models: SpaCy, Transformers library (e.g., Hugging Face transformers)

[1265] Framework: Python, Flask (Web application framework)

[1266] Program processing overview

[1267] The server receives queries sent from the user's device via an API and temporarily stores them in a database. It then passes the queries to a natural language processing model (e.g., SpaCy or Transformers) to analyze them and identify their intent. It then retrieves relevant information from the database based on the analysis results and generates an appropriate answer. This answer is sent to the user's device in JSON format. The query and the generated answer are also stored in a database for use in subsequent data analysis.

[1268] Specific examples

[1269] For product inquiries:

[1270] A user makes a query through a smartphone application, asking, "Please tell me the stock of product A." This query is sent from the device to the server, which temporarily stores the query in a database. Using a natural language processing model, the intent "to check stock" is identified and the stock information of product A is retrieved from the database. Finally, the server generates a response saying, "There are currently 15 units of product A in stock," and sends it back to the user's device in JSON format.

[1271] Example prompt sentence:

[1272] "Please let me know the stock of product A."

[1273] For shipping inquiries:

[1274] A user makes a query from their PC asking, "Please tell me the delivery status of product B." This query is sent from the device to the server, which temporarily stores the query in a database. Using a natural language processing model, the intent "check delivery status" is identified and the delivery information for product B is retrieved from the database. Finally, the server generates a response stating, "Product B is currently being delivered. The expected delivery date is next Tuesday," and sends it back to the user's device in JSON format.

[1275] Example prompt sentence:

[1276] "Please let me know the delivery status of product B."

[1277] This will enable immediate and appropriate responses to inquiries, and through centralized data management and analysis, will improve customer satisfaction and make future inquiries more efficient.

[1278] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1279] Step 1:

[1280] The user enters an inquiry from their device (smartphone, tablet, or PC) and presses the send button. At this time, the entered inquiry content is sent to the server as an API request.

[1281] Input: The inquiry entered by the user (e.g., "Please tell me the inventory of product A.")

[1282] Output: Query data in API request format

[1283] Step 2:

[1284] The terminal sends the received inquiry to the server via API and temporarily stores it in a database.

[1285] Input: Inquiry data in API request format

[1286] Output: A database containing the query results

[1287] Specific operation: The server receives a request from a terminal and records the contents in a specific table in the database.

[1288] Step 3:

[1289] The server passes the query content stored in the database to a natural language processing (NLP) model to analyze the content.

[1290] Input: Inquiry details saved in the database

[1291] Output: Analysis results from the natural language processing model (intent and extracted entities)

[1292] Specific operation: The server passes the query content to a natural language processing model (e.g., SpaCy or Hugging Face's transformers model), determines the intent as "check inventory," etc., and extracts entities such as product names.

[1293] Step 4:

[1294] The server determines the intent of the query based on the analysis results of the NLP model and retrieves the relevant information from the database.

[1295] Input: Analysis results from natural language processing model

[1296] Output: Result of retrieving related information (e.g., inventory quantity and delivery status)

[1297] Specific behavior: The server generates and executes a query to the database based on the identified intent, and retrieves the relevant data (e.g., product inventory and delivery status).

[1298] Step 5:

[1299] The server generates a response based on the acquired information and sends it to the user's terminal.

[1300] Input: Related information (e.g., inventory quantity and delivery status)

[1301] Output: Generated answer (e.g. "There are 15 units of product A in stock.")

[1302] Specific operation: The server uses the acquired data to create a corresponding answer using a template or plain text generation, and returns it to the user's device in JSON format.

[1303] Step 6:

[1304] The terminal displays the received response to the user.

[1305] Input: Response data sent from the server

[1306] Output: User-visible answers

[1307] Specific operation: The terminal analyzes the received JSON formatted response data and displays it in a format that is easy for the user to understand via the user interface.

[1308] Step 7:

[1309] The server stores the final query and response in a database for later data analysis.

[1310] Input: Final query and generated answer

[1311] Output: A record of the query and answer stored in a database

[1312] Specific operation: The server records pairs of queries and their answers in a database and uses them as data for future trend analysis and system improvement.

[1313] Example prompt sentence:

[1314] "Please let me know the stock of product A."

[1315] 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.

[1316] This invention provides a system that automatically receives and analyzes user inquiries, generates appropriate responses, and returns them with an emotion engine that recognizes the user's emotions, enabling even more sophisticated responses. This system is primarily composed of a server, terminals, and users. Below are the system's program processing and specific examples.

[1317] System Configuration

[1318] This system consists of the following main components:

[1319] 1. Means of receiving inquiries

[1320] 2. Means of analyzing the content of received inquiries

[1321] 3. A means of generating answers based on the parsed query content

[1322] 4. A means of sending the generated answer

[1323] 5. A means of storing queries and generated responses in a database

[1324] 6. Means of analyzing stored data

[1325] 7. Emotion engine that recognizes user emotions

[1326] Program processing

[1327] How to receive inquiries

[1328] The user inputs a query using their own terminal and presses the send button, which sends the query to the server.

[1329] The terminal sends the entered inquiry to the server via API.

[1330] A means of analyzing the content of received inquiries

[1331] The server temporarily stores the inquiry received from the terminal in a database.

[1332] The server passes the received query content to a natural language processing (NLP) model to analyze the content.

[1333] Emotion engine that recognizes user emotions

[1334] The server passes the data, along with the query content, to an emotion engine for analyzing the user's emotions.

[1335] The emotion engine analyzes emotions from the user's text and returns the results to the server, identifying emotions such as "anger," "joy," and "sadness."

[1336] A means of generating answers based on the parsed query content

[1337] The server determines the intent of the query and the user's emotions based on the analysis results of the NLP model and the emotion engine.

[1338] For example, if a user inquires, "Please tell me the product's stock status," the server determines the intent as "checking stock" and the emotion as "neutral."

[1339] A means of sending the generated answer

[1340] Based on the determined intent and emotion, the server retrieves relevant information from a database and generates a response in an appropriate tone.

[1341] The server generates a response based on the acquired information and sends it to the user's terminal.

[1342] The terminal displays the received response to the user.

[1343] A means of storing queries and generated responses in a database

[1344] The server permanently stores the query content and the results of sentiment analysis in a database, along with the final answer.

[1345] A means of analyzing stored data

[1346] The server periodically analyzes the accumulated inquiry, response, and sentiment analysis data to extract inquiry trends and patterns.

[1347] Based on the analysis results, we will build a system that proactively provides relevant information that responds to future inquiries and emotional states.

[1348] Specific examples

[1349] Product inquiries

[1350] User: The user enters "Is product A in stock?" and presses the send button.

[1351] Terminal: The terminal sends this query to the server.

[1352] Server: The server temporarily stores the query content in a database and passes it to the natural language processing model.

[1353] Server: The NLP model determines the intent, "Check inventory," and the emotion engine determines the emotion, "Neutral."

[1354] Server: The server retrieves the stock information of product A from the database, generates a response in a neutral tone such as "Currently, there are 15 units of product A in stock," and sends it to the user's device.

[1355] Terminal: The terminal displays this answer to the user.

[1356] Server: The final query, response, and sentiment analysis results are stored in a database for later data analysis.

[1357] Emotion-conscious response to inquiries

[1358] User: The user types, "The delivery of product B is delayed. What's going on?" and submits.

[1359] Terminal: The terminal sends this query to the server.

[1360] Server: The server temporarily stores the query content in a database and passes it to the natural language processing model.

[1361] Server: The NLP model determines the intent, "check delivery status," and the emotion engine determines the emotion, "anger."

[1362] Server: The server retrieves the delivery status of Product B from the delivery system database, generates a polite response saying, "Product B is currently being delivered. We apologize for any inconvenience. The expected delivery date is next Tuesday," and sends it to the user's device.

[1363] Terminal: The terminal displays this answer to the user.

[1364] Server: The final query, response, and sentiment analysis results are stored in a database for later data analysis.

[1365] This allows the system to respond to inquiries taking emotions into consideration, contributing to improved customer satisfaction.

[1366] The processing flow will be explained below.

[1367] Step 1:

[1368] User: Enter the inquiry details and click the send button.

[1369] For example, enter "Please let me know the product availability."

[1370] Step 2:

[1371] Terminal: Sends the entered inquiry to the server.

[1372] Specifically, the inquiry data is sent via the API using an HTTP POST request.

[1373] Step 3:

[1374] Server: Temporarily stores the received inquiry in a database.

[1375] Specifically, the query content is recorded in the query table using an SQL insert statement.

[1376] Example: INSERT INTO inquiries (user_id, inquiry_text) VALUES ('1234', 'What is the product availability?')

[1377] Step 4:

[1378] Server: Passes the received query content to a natural language processing (NLP) model for analysis.

[1379] Specifically, the query is passed to the NLP model API and the analysis results are received in JSON format.

[1380] Example: "Product availability" is parsed as "Intent: Check inventory."

[1381] Step 5:

[1382] Server: Passes the query content to the emotion engine and analyzes the user's emotions.

[1383] Specifically, the query is passed to the emotion engine's API, and the results of the emotion analysis are received in JSON format.

[1384] Example: "What is the product availability?" is parsed as "Neutral."

[1385] Step 6:

[1386] Server: Determines the intent of the query and the user's emotions based on the analysis results of the NLP model and the emotion engine.

[1387] For example, "Intention: Check inventory" and "Emotion: Neutral" are determined.

[1388] Step 7:

[1389] Server: Based on the intent of the query, retrieves the necessary information from the database.

[1390] Specifically, an SQL query is executed to obtain the stock quantity of the relevant product from the inventory table.

[1391] Example: SELECT stock_quantity FROM products WHERE product_name = 'Product A'

[1392] Step 8:

[1393] Server: Generates answers based on the acquired information and the user's sentiment.

[1394] For example, frame your response in a neutral tone, saying, "We currently have 15 units of product A in stock."

[1395] Step 9:

[1396] Server: Sends the generated answer to the user's device.

[1397] Specifically, the answer is sent as an HTTP response.

[1398] Step 10:

[1399] Terminal: Displays the received answer to the user.

[1400] For example, the chat screen in your browser or app might display "Currently, there are 15 units of product A in stock."

[1401] Step 11:

[1402] Server: The inquiry content and sentiment analysis results, along with the final answer, are permanently stored in a database.

[1403] Use SQL insert or update statements to store queries and their answers.

[1404] Example: UPDATE inquiries SET response = 'Currently, there are 15 units of product A in stock', emotion = 'Neutral' WHERE inquiry_id = '5678'

[1405] Step 12:

[1406] Server: Periodically analyzes the accumulated data to identify trends and patterns in inquiries.

[1407] Specifically, data analysis tools such as Python and R are used to analyze past inquiries in terms of distribution and frequency, as well as emotional trends.

[1408] The above is the specific flow of processing the inquiry content, generating and providing an answer taking the user's feelings into consideration. This system will improve the efficiency of inquiry responses and contribute to improving customer satisfaction.

[1409] Example 2

[1410] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1411] Conventional inquiry systems only provided standardized answers without considering the user's feelings, resulting in low user satisfaction and the inability to respond appropriately to complaints and urgent inquiries in particular. In addition, the stored data was treated as a simple log, and could not be used to proactively respond to future inquiries or for improvements.

[1412] 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.

[1413] In this invention, the server includes means for receiving an inquiry, means for analyzing the content of the received inquiry, means for recognizing the user's emotion, means for generating an answer based on the analyzed content of the inquiry and the recognized emotion, means for transmitting the generated answer, means for saving the content of the inquiry and the generated answer in a database, and means for analyzing the saved data. This makes it possible to generate an appropriate answer that takes the user's emotion into consideration, and further makes it possible to proactively respond to future inquiries by utilizing the accumulated data.

[1414] The "means for receiving an inquiry" is a device or process for receiving the content of an inquiry sent from a user to the server via the terminal.

[1415] "Means for analyzing the received query content" refers to a device or process for understanding the received query text and identifying its intent, often using natural language processing (NLP) techniques.

[1416] The "means for recognizing user emotions" is a device or process for determining the user's emotions from the query text, and often utilizes an emotion analysis engine.

[1417] The "means for generating an answer based on the analyzed query content and the recognized sentiment" is a device or process for generating an appropriate answer based on the intent of the query and the recognized sentiment.

[1418] A "means for transmitting a generated answer" is a device or process for transmitting a generated answer to a user.

[1419] The "means for storing the inquiry contents and generated answers in a database" is a device or process for storing all the inquiry contents and the generated answers thereto in a database.

[1420] A "means for analyzing stored data" is a device or process that uses stored inquiry and response data to extract trends and patterns to improve future inquiry responses.

[1421] "Natural Language Processing (NLP)" is a technology that enables machines to understand, analyze, and generate human language.

[1422] An "emotion analysis engine" is an algorithm or software for identifying a user's emotions from text data.

[1423] A "database" is an information storage system for storing queries and generated responses.

[1424] This invention enables more advanced responses by combining a system that receives and analyzes user inquiries, generates appropriate responses, and responds with an emotion engine that recognizes the user's emotions. The system consists of the following main components:

[1425] System Configuration

[1426] This system is mainly composed of a server, terminals, and users. Specifically, it uses the following devices and software:

[1427] server

[1428] Hardware: A server machine equipped with a high-performance processor, memory, and storage

[1429] software:

[1430] Database: MySQL, PostgreSQL, MongoDB, Elasticsearch, etc.

[1431] Natural Language Processing (NLP) libraries: spaCy, NLTK

[1432] Sentiment analysis engine: Microsoft Azure's Text Analytics API, IBM Watson's Natural Language Understanding API

[1433] Text generation model: OpenAI's GPT-3

[1434] Terminal

[1435] Hardware: Smartphones, tablets, personal computers

[1436] Software: Web browsers (Google Chrome, Safari, etc.), mobile applications

[1437] User

[1438] Interface: inquiry form, etc.

[1439] Program processing

[1440] Receive inquiries

[1441] The user uses their device (smartphone or computer) to enter the details of their inquiry. For example, they might enter "Please tell me the product's stock status" in a web browser and press the send button. The device then sends this inquiry to the server via an API (RESTful API or GraphQL).

[1442] Analyze the incoming inquiries

[1443] The server temporarily stores the query received from the device in a database (MySQL or PostgreSQL). It then uses a natural language processing engine (spaCy or NLTK) to analyze the intent of the query. For example, it classifies the text "Please tell me the stock status of the product" as the intent "Check stock."

[1444] Recognize user emotions

[1445] The server sends the received inquiry content to a sentiment analysis engine, which determines the user's emotions (such as "anger," "joy," or "sadness") from the text. Examples of sentiment analysis engines include Microsoft Azure's Text Analytics API and IBM Watson's Natural Language Understanding API.

[1446] Generate an answer

[1447] The server generates a response using a text generation model (OpenAI's GPT-3) based on the intent and sentiment analysis results. For example, it generates a neutral response such as "Currently, there are 15 units of product A in stock."

[1448] Submit your answer

[1449] The server sends the generated answer to the user's terminal, which displays the answer to the user.

[1450] Save your data

[1451] The server stores the query content and sentiment analysis results along with the final answer in a database (MongoDB or Elasticsearch).

[1452] Analyze the data

[1453] The stored data is periodically analyzed and data analysis tools such as Apache Hadoop and Google BigQuery are used to extract trends and patterns in inquiries, which can then be used to proactively respond to future inquiries.

[1454] Specific examples

[1455] Product inquiries

[1456] A user enters "Do you have product A in stock?" into an inquiry form in Google Chrome and presses the submit button. The device sends this inquiry to the server via a RESTful API. The server temporarily saves the inquiry content in MySQL and uses spaCy to extract the intent "to check stock." It then uses Microsoft Azure's Text Analytics to determine the sentiment of "neutral." It retrieves the stock information for product A from the database and generates a response in a neutral tone, such as "There are currently 15 units of product A in stock." This response is then sent back to the device in JSON format. The device displays this response on the user interface and saves the final inquiry, response, and sentiment analysis results in a MySQL database.

[1457] Inquiries about delivery status

[1458] The user types and submits the query in Safari, "The delivery of Product B is delayed. What's going on?" The device sends this query to the server using the GraphQL API. The server temporarily stores the query content in PostgreSQL and uses NLTK to extract the intent, "Check delivery status." It then uses IBM Watson's Natural Language Understanding to determine the emotion, "anger." It retrieves the delivery status of Product B from the delivery system database and generates a polite response, saying, "Product B is currently being delivered. We apologize for any inconvenience. The expected delivery date is next Tuesday." The response returned to the device in JSON format is displayed on the user interface, and the final query, response, and emotion analysis results are saved in the PostgreSQL database.

[1459] This allows the system to respond to inquiries taking into account the user's emotions, contributing to improved customer satisfaction.

[1460] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1461] Step 1: Receiving an inquiry

[1462] The user uses their own device (smartphone or computer) to enter the details of their inquiry and press the send button. Specifically, the user opens a web browser (e.g., Google Chrome), enters "Please let me know the product's stock status" into the inquiry form, and clicks send.

[1463] Input: User-supplied query text

[1464] Output: Data sent from the device to the server (inquiry details, user ID, timestamp, etc.)

[1465] The device sends the entered inquiry to the server via an API (e.g., RESTful API). At this time, the sent data includes metadata such as the user ID and timestamp.

[1466] Step 2: Analyze the query

[1467] The server temporarily stores the query received from the terminal in a database (e.g., MySQL). This temporary storage is performed as a transaction to maintain data consistency.

[1468] Input: Inquiry data sent from the terminal

[1469] Output: Temporarily saved inquiry data

[1470] The server uses a natural language processing (NLP) engine (e.g., spaCy) to analyze the query. Specifically, it extracts the intent from the query text. For example, the text "Please tell me the product's stock status" is classified as the intent "Check stock."

[1471] Step 3: Recognize the user's emotions

[1472] The server sends the received inquiry content to a sentiment analysis engine (e.g., Microsoft Azure's Text Analytics API) to analyze the user's sentiment.

[1473] Input: Inquiry details

[1474] Output: Emotion analysis result (e.g. "neutral")

[1475] The sentiment analysis engine determines the user's sentiment from the query text, for example, "Please tell me the product availability status," and identifies the sentiment "neutral."

[1476] Step 4: Generate an answer

[1477] The server generates an answer using a text generation model (e.g., OpenAI's GPT-3) based on the results of intent analysis and sentiment analysis.

[1478] Input: Intent analysis results (e.g., "Check inventory"), sentiment analysis results (e.g., "Neutral")

[1479] Output: Generated answer (e.g. "Currently, there are 15 units of product A in stock.")

[1480] The server retrieves the necessary information from a database (e.g., MongoDB) based on the intent. For example, for the intent "check stock," it retrieves stock information for product A. It then uses a text generation model to generate an answer in an appropriate tone. For example, it generates an answer such as "Currently, there are 15 units of product A in stock."

[1481] Step 5: Submit your response

[1482] The server then sends the generated response to the user's device, often in JSON format.

[1483] Input: Generated response data

[1484] Output: Data sent to the terminal

[1485] The terminal displays the received answer on the user interface, allowing the user to check the answer on their screen.

[1486] Step 6: Save your data

[1487] The server stores the final answer, inquiry details, and sentiment analysis results in a database (e.g., Elasticsearch).

[1488] Input: Final response, inquiry details, sentiment analysis results

[1489] Output: Data stored in the database

[1490] The saved data can be used for later data analysis.

[1491] Step 7: Analyze the data

[1492] The server periodically analyzes the accumulated inquiry, response, and sentiment analysis data.

[1493] Input: Data of inquiries, responses, and sentiment analysis results stored in the database

[1494] Output: Analysis results (inquiry trends and patterns)

[1495] Specifically, Apache Hadoop and Google BigQuery are used to process data and extract trends and patterns. The analysis results are then used to respond to future inquiries. For example, if there are many inquiries about a particular season or product, that information can be prepared in advance.

[1496] (Application example 2)

[1497] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1498] For modern online shopping sites, responding to user inquiries quickly and appropriately is an important issue that directly contributes to improving customer satisfaction. However, conventional inquiry response systems do not adequately take user emotions into account, resulting in inconsistent response quality. In particular, when a user is dissatisfied or angry, it is difficult to respond appropriately based on that emotion. Another challenge is efficiently analyzing the content of a large number of inquiries and generating appropriate answers.

[1499] 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.

[1500] In this invention, the server includes means for receiving an inquiry, means for analyzing the received inquiry, means for generating a response based on the analyzed inquiry, means for transmitting the generated response, means for saving the inquiry and the generated response in a database, means for analyzing the saved data, means including an emotion engine for recognizing the user's emotion, and means for adjusting the tone of the response in accordance with the user's emotion, thereby enabling a prompt and appropriate response to an inquiry that takes the user's emotion into consideration.

[1501] The "means for receiving an inquiry" is a function by which the system receives and processes an inquiry from a user.

[1502] The "means for analyzing the content of the received inquiry" is a function for understanding the content of the received inquiry and analyzing it as structured data.

[1503] The "means for generating a response based on the analyzed inquiry content" is a function that automatically generates an appropriate response based on the analysis results.

[1504] The "means for sending the generated answer" is a function for returning the generated answer to the user.

[1505] The "means for saving the inquiry and the generated response in a database" is a function for saving the inquiry and the response as a record in a database.

[1506] The "means for analyzing stored data" is a function for analyzing stored inquiry and response data and extracting significant insights and patterns.

[1507] The "emotion engine that recognizes user emotions" is an engine that analyzes and identifies emotions from the content of a user's inquiry.

[1508] The "means for adjusting the tone of the response according to the user's emotion" is a function for automatically adjusting the expression and tone of the response according to the emotion specified by the user.

[1509] The present invention is a system for responding promptly and appropriately to inquiries from users on an online shopping site. This system analyzes the content of the user's inquiry and recognizes their emotions, thereby generating a response in a tone that corresponds to their emotions and improving the quality of the response.

[1510] System Configuration

[1511] The system mainly consists of the following components:

[1512] 1. Means of receiving inquiries

[1513] 2. Means of analyzing the content of received inquiries

[1514] 3. A means of generating answers based on the parsed query content

[1515] 4. A means of sending the generated answer

[1516] 5. A means of storing queries and generated responses in a database

[1517] 6. Means of analyzing stored data

[1518] 7. Emotion engine that recognizes user emotions

[1519] 8. How to adjust the tone of your response based on the user's emotions

[1520] Hardware and Software

[1521] Hardware used:

[1522] server

[1523] User device (smartphone or PC)

[1524] Software used:

[1525] Python

[1526] Natural Language Processing Library (Hugging Face Transformers)

[1527] Sentiment analysis model (BERT)

[1528] Database System

[1529] Data processing and calculation

[1530] 1. Means of receiving inquiries:

[1531] The user uses a smartphone or computer to enter and submit an inquiry.

[1532] The terminal sends the entered inquiry to the server via API.

[1533] 2. How to analyze the received inquiry:

[1534] The server temporarily stores the inquiry received from the terminal in a database.

[1535] The data stored in the database is analyzed using natural language processing models.

[1536] 3. Emotion engine that recognizes user emotions:

[1537] The server passes the data along with the query to the emotion engine.

[1538] The emotion engine analyzes the emotion from the query and returns the result to the server. The emotion may be "anger," "joy," or "sadness," for example.

[1539] 4. Means of generating an answer based on the parsed query:

[1540] The server determines the intent of the query and the user's emotions based on the analysis results of the NLP model and the emotion engine.

[1541] 5. Means for sending generated answers:

[1542] The server generates a response in an appropriate tone based on the determined intent and emotion.

[1543] The generated answer is sent to the user's terminal.

[1544] 6. How to store queries and generated responses in a database:

[1545] The inquiry details and the results of sentiment analysis are stored in a database along with the final answer.

[1546] Examples of specific examples and prompts

[1547] For example, if a user asks, "Product B hasn't arrived yet. What's going on?", the following process will be performed:

[1548] User inquiry: "I haven't received my delivery of product B yet. What's going on?"

[1549] Emotion engine judgement: "Anger"

[1550] System response: "I see you're angry. We'll deal with it right away."

[1551] Example prompt sentence:

[1552] User Question: "I haven't received my delivery of item B yet, what's going on?"

[1553] System emotion detection: "Anger"

[1554] Generate an appropriate response.

[1555] This allows for prompt and appropriate response to inquiries that take into account the user's feelings.

[1556] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1557] Step 1:

[1558] The user enters an inquiry on a smartphone or PC. When the user enters the inquiry content as text and presses the send button, the device sends the inquiry data to the server via the API.

[1559] Input: The query text entered by the user.

[1560] Output: The query data sent to the server.

[1561] Step 2:

[1562] The server temporarily stores the received inquiry in a database.

[1563] Input: Enquiry data received through API.

[1564] Output: Query data temporarily stored in a database.

[1565] Step 3:

[1566] The server passes the received query content to a natural language processing (NLP) model for analysis, which analyzes the text and extracts the intent of the query.

[1567] Input: Query data stored in the database.

[1568] Output: Analysis result data including query intent.

[1569] Step 4:

[1570] The server passes the query content and sentiment analysis data to the sentiment engine, which analyzes the sentiment from the user's text and returns the results to the server.

[1571] Input: Query text.

[1572] Output: Emotion analysis results from the emotion engine (e.g., "anger," "joy," "sadness," etc.).

[1573] Step 5:

[1574] The server determines the intent of the query and the user's emotions based on the analysis results of the NLP model and the emotion engine.

[1575] Input: Analysis results of the NLP model and the emotion engine.

[1576] Output: Analysis result data including intent and sentiment.

[1577] Step 6:

[1578] Based on the determined intent and emotion, the server retrieves relevant information from a database and generates a response in an appropriate tone.

[1579] Input: Analysis result data including intent and sentiment.

[1580] Output: Answer text generated in the appropriate tone.

[1581] Step 7:

[1582] The server sends the generated answer to the user's terminal, which then displays the received answer to the user.

[1583] Input: The generated answer text.

[1584] Output: The answer text that is displayed on the user's terminal.

[1585] Step 8:

[1586] The server stores the query and the results of the sentiment analysis in a database, along with the final answer.

[1587] Input: Query, generated answer, sentiment analysis results.

[1588] Output: The final data stored in the database.

[1589] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1590] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1591] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1592] [Fourth embodiment]

[1593] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1594] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1595] 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).

[1596] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1597] 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.

[1598] 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).

[1599] 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. 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.

[1600] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1601] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1602] 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.

[1603] 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.

[1604] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1605] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1606] The system of the present invention responds quickly and efficiently by automatically receiving and analyzing inquiries from users and generating and returning appropriate answers. This system is primarily composed of a server, terminals, and users. The following describes the processing of the system program and a specific example.

[1607] System Configuration

[1608] This system consists of the following main components:

[1609] 1. Means of receiving inquiries

[1610] 2. Means of analyzing the content of received inquiries

[1611] 3. A means of generating answers based on the parsed query content

[1612] 4. A means of sending the generated answer

[1613] 5. A means of storing queries and generated responses in a database

[1614] 6. Means of analyzing stored data

[1615] Program processing

[1616] How to receive inquiries

[1617] The user inputs a query using their own terminal and presses the send button, which sends the query to the server.

[1618] The terminal sends the entered inquiry to the server via API.

[1619] A means of analyzing the content of received inquiries

[1620] The server temporarily stores the inquiry received from the terminal in a database.

[1621] The server passes the received query content to a natural language processing (NLP) model to analyze the content.

[1622] A means of generating answers based on the parsed query content

[1623] The server determines the intent of the query based on the analysis results of the NLP model.

[1624] For example, if a user inquires, "Please tell me the stock status of the product," the server determines that the user's intent is "to check stock."

[1625] A means of sending the generated answer

[1626] The server retrieves the relevant information from the database based on the determined intent.

[1627] The server generates a response based on the acquired information and sends it to the user's terminal.

[1628] The terminal displays the received response to the user.

[1629] A means of storing queries and generated responses in a database

[1630] The server permanently stores the final answer together with the inquiry details in a database.

[1631] A means of analyzing stored data

[1632] The server periodically analyzes the accumulated inquiry and response data to extract inquiry trends and patterns.

[1633] Based on the analysis results, we will build a system that predicts future inquiries and provides relevant information in advance.

[1634] Specific examples

[1635] Product inquiries

[1636] User: The user enters "Is product A in stock?" and presses the send button.

[1637] Terminal: The terminal sends this query to the server.

[1638] Server: The server temporarily stores the query content in a database and passes it to the natural language processing model.

[1639] Server: The NLP model determines the intent to "check stock," and the server retrieves stock information for product A from the database.

[1640] Server: Generates a response saying "Currently, there are 15 units of product A in stock" and sends it to the user's device.

[1641] Terminal: The terminal displays this answer to the user.

[1642] Server: The final query and answer are stored in a database for later data analysis.

[1643] Delivery inquiries

[1644] User: The user enters "Please tell me the delivery status of product B" and submits.

[1645] Terminal: The terminal sends this query to the server.

[1646] Server: The server temporarily stores the query content in a database and passes it to the natural language processing model.

[1647] Server: The NLP model determines the intent to "check delivery status," and the server retrieves the delivery status of Product B from the delivery system database.

[1648] Server: Generates a response stating, "Item B is currently being delivered. The estimated delivery date is next Tuesday," and sends it to the user's device.

[1649] Terminal: The terminal displays this answer to the user.

[1650] Server: The final query and answer are stored in a database for later data analysis.

[1651] This allows the system to automate inquiries efficiently and accurately, and the accumulated data can be used to improve customer service in the future.

[1652] The processing flow will be explained below.

[1653] Step 1:

[1654] User: Enter the inquiry details and click the send button.

[1655] For example, enter "Please let me know the product availability."

[1656] Step 2:

[1657] Terminal: Sends the entered inquiry to the server.

[1658] Specifically, the inquiry data is sent via the API using an HTTP POST request.

[1659] Step 3:

[1660] Server: Temporarily stores the received inquiry in a database.

[1661] Specifically, the query content is recorded in the query table using an SQL insert statement.

[1662] INSERT INTO inquiries (user_id, inquiry_text) VALUES ('1234', 'What is the product availability?')

[1663] Step 4:

[1664] Server: Passes the received query content to a natural language processing (NLP) model for analysis.

[1665] Specifically, the query is passed to the NLP model API and the analysis results are received in JSON format.

[1666] Example: "Product availability" is parsed as "Intent: Check inventory."

[1667] Step 5:

[1668] Server: Determines the intent of the query based on the analysis results of the NLP model.

[1669] Based on the analysis results, the corresponding processing is selected.

[1670] Example: Determine the intent "check inventory."

[1671] Step 6:

[1672] Server: Based on the intent of the query, retrieves the necessary information from the database.

[1673] Specifically, an SQL query is executed to obtain the stock quantity of the relevant product from the inventory table.

[1674] SELECT stock_quantity FROM products WHERE product_name = 'A product'

[1675] Step 7:

[1676] Server: Generates an answer based on the information obtained.

[1677] For example, construct an answer such as "Currently, there are 15 units of product A in stock."

[1678] Step 8:

[1679] Server: Sends the generated answer to the user's device.

[1680] Specifically, the answer is sent as an HTTP response.

[1681] Step 9:

[1682] Terminal: Displays the received answer to the user.

[1683] For example, the chat screen in your browser or app might display "Currently, there are 15 units of product A in stock."

[1684] Step 10:

[1685] Server: The query is permanently stored in a database along with the final answer.

[1686] Use SQL insert or update statements to store queries and their answers.

[1687] UPDATE inquiries SET response = 'Currently, there are 15 units of product A in stock' WHERE inquiry_id = '5678'

[1688] Step 11:

[1689] Server: Periodically analyzes the accumulated data to identify trends and patterns in inquiries.

[1690] Specifically, data analysis tools such as Python and R are used to analyze past inquiries in terms of distribution and frequency.

[1691] The above is the specific flow of processing from the inquiry content to returning a response to the user. This system improves the efficiency of inquiry responses and contributes to improving customer satisfaction.

[1692] Example 1

[1693] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1694] Conventional inquiry response systems take time to analyze the content of inquiries and generate responses, making it difficult to respond to user inquiries quickly. Furthermore, there was no established method for effectively utilizing accumulated data, making it difficult to predict future inquiries or provide appropriate information. This resulted in a lack of improvement in the user experience and a decline in the quality of a company's services.

[1695] 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.

[1696] In this invention, the server includes means for receiving an inquiry, means for analyzing the received inquiry, means for generating a response based on the analyzed inquiry, means for transmitting the generated response, means for storing the inquiry and the generated response in an information storage medium, and means for analyzing the stored data. This makes it possible to quickly and accurately analyze the inquiry and generate and return an appropriate response. Furthermore, by effectively utilizing the accumulated data, it is possible to predict future inquiries and provide useful information, thereby significantly improving the user experience.

[1697] The "means for receiving an inquiry" is a device or software that has the function of transmitting an inquiry from a user to a server as digital data.

[1698] "Means for analyzing the content of the received inquiry" refers to a device or software that has the function of analyzing the received digital data using methods such as natural language processing and understanding its intent and content.

[1699] The "means for generating a response based on the analyzed inquiry content" is a device or software that has the function of automatically generating an appropriate response based on the analysis results.

[1700] The "means for transmitting the generated answer" is a device or software that has the function of transmitting the generated answer to the user's terminal.

[1701] The "means for storing the inquiry content and the generated response in an information storage medium" refers to a device or software that has the function of recording the inquiry content and the response in digital form and storing it permanently.

[1702] "Means for analyzing stored data" refers to devices or software that have the ability to analyze accumulated data using statistical or machine learning techniques and extract trends and patterns.

[1703] "Natural language processing" is a technology that enables computers to understand and analyze human language, and involves processing text data using language models and algorithms.

[1704] A "terminal" is a device that allows a user to input and send an inquiry, and includes a computer, smartphone, tablet, etc.

[1705] An "information storage medium" is a device or medium for storing digital data for a long period of time, and includes hard disk drives (HDDs) and solid-state drives (SSDs).

[1706] The present invention relates to a system for improving the efficiency of an inquiry response process. The system is composed of a user, a terminal, and a server, and has the function of automating the reception, analysis, response generation, and response transmission of an inquiry. The program processing of the system and a specific embodiment are described below.

[1707] System Configuration

[1708] The system consists of the following main components:

[1709] 1. Means of receiving inquiries

[1710] 2. Means of analyzing the content of received inquiries

[1711] 3. A means of generating answers based on the parsed query content

[1712] 4. A means of sending the generated answer

[1713] 5. Means for storing the inquiry content and the generated response in an information storage medium

[1714] 6. Means of analyzing stored data

[1715] Receiving inquiries

[1716] The user enters an inquiry into their device and presses the send button. For example, they may enter "Please tell me the product's stock status."

[1717] The device sends the input query to the server via API, structuring the query in JSON format and sending it to the server using an HTTP POST request.

[1718] Saving inquiry details

[1719] The server temporarily stores the query received from the terminal in an information storage medium. Specifically, the server inserts the received query into the "ReceivedQueries" table.

[1720] Analysis of inquiry content

[1721] The server passes the received query content to a natural language processing (NLP) model to analyze the content, for example using TextBlob, spaCy, or a custom NLP model (e.g., the BERT model).

[1722] The server determines the intent of the query based on the output of the NLP model. For example, if a user asks, "Please tell me the stock status of a product," the NLP model determines the intent as "check stock."

[1723] Generate answers

[1724] The server retrieves the relevant information from the information storage medium based on the determined intention. For example, it selects the inventory information of the relevant product from the "Inventory" table.

[1725] The server generates an answer based on the information it has obtained, for example, constructing a sentence such as "There are currently 15 units of this product in stock." It may also generate sentences using templates.

[1726] Submit your answer

[1727] The server sends the generated answer to the user's device in JSON format as an HTTP response.

[1728] The device displays the received response to the user, for example, in a browser or a mobile app UI component.

[1729] Data storage

[1730] The server permanently stores the query and the generated response in an information storage medium. Specifically, the columns to be inserted into the "QueriesAndResponses" table include "user_id", "query", "response", and "timestamp".

[1731] Analyzing the data

[1732] The server periodically analyzes the accumulated query and response data, for example, using Python's Pandas library or SQL queries.

[1733] The server will use the analysis results to predict future inquiries and proactively provide relevant information. For example, machine learning models can be used to predict inquiry trends, improve the automated response system, and update FAQs.

[1734] Example prompt

[1735] An example of a prompt sentence could be "What is the system's procedure when a user asks, 'Is the product in stock?'"

[1736] This enables the system to respond to user inquiries quickly and accurately, enabling efficient inquiry management and future service improvements through the use of data.

[1737] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1738] Step 1:

[1739] The user inputs an inquiry into the terminal and presses the send button. At this time, the user inputs an inquiry such as "Please tell me the inventory status of the product." The input data is the inquiry content (e.g., "Do you have product A in stock?").

[1740] Step 2:

[1741] The device structures the query and sends it to the server via API. Specifically, it converts the query into JSON format and sends it using an HTTP POST request. The input is the user's query, and the output is JSON format data.

[1742] Step 3:

[1743] The server temporarily saves the query received from the terminal in a database. Specifically, it inserts it into the "ReceivedQueries" table. The input is the query in JSON format, and the output is the data saved in the database.

[1744] Step 4:

[1745] The server passes the received query content to a natural language processing (NLP) model to analyze the content. For example, it uses TextBlob, spaCy, or a custom NLP model (e.g., the BERT model). The input is the stored query data, and the output is the analysis result (e.g., the intent "check inventory").

[1746] Step 5:

[1747] The server determines the intent of the query based on the analysis results of the NLP model. The input is the analysis results from the NLP model, and the output is the determined intent (e.g., "check inventory").

[1748] Step 6:

[1749] The server retrieves the relevant information from the database based on the determined intent. For example, SELECT the inventory information of a specific product from the "Inventory" table. The input is the determined intent, and the output is the retrieved inventory information (e.g., "15 units in stock").

[1750] Step 7:

[1751] The server generates a response based on the acquired information. For example, it generates a sentence such as "Currently, there are 15 units of product A in stock." The input is the acquired inventory information, and the output is the generated response sentence.

[1752] Step 8:

[1753] The server sends the generated answer to the terminal in JSON format as an HTTP response. The input is the generated answer text, and the output is the JSON format response.

[1754] Step 9:

[1755] The device displays the received answer to the user, for example, in a UI component of a browser or mobile app. The input is a JSON-formatted response, and the output is the answer text that is displayed to the user.

[1756] Step 10:

[1757] The server stores the query and generated response in a database permanently. Specifically, it inserts the query and generated response into the "QueriesAndResponses" table. The input is the query and generated response, and the output is the record stored in the database.

[1758] Step 11:

[1759] The server periodically analyzes the accumulated data using Python's Pandas library and SQL queries. The input is the accumulated inquiry and response data, and the output is the analysis results and trend analysis.

[1760] Step 12:

[1761] The server uses the analysis results to predict future inquiries and builds a system that proactively provides relevant information. For example, a machine learning model can be used to predict inquiry trends, improve the automated response system, and update FAQs. The input is the analysis results, and the output is the predictive model and an improved system.

[1762] (Application example 1)

[1763] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1764] Conventional inquiry response systems often required manual response, resulting in the problem of long response times. It was also difficult to provide an appropriate response immediately based on the inquiry, which could lead to a decline in customer satisfaction. Furthermore, because data was not centrally managed, it was difficult to later analyze inquiries and their responses.

[1765] 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.

[1766] In this invention, the server includes means for using a natural language processing model to identify the intent of the inquiry, means for retrieving information from a database based on the intent of the inquiry, and means for returning the generated answer in JSON format. This enables the provision of an immediate and appropriate answer to the inquiry, and furthermore, centralized management and analysis of data enables the efficiency of subsequent inquiry responses and improvement of customer satisfaction.

[1767] The "means for receiving an inquiry" is a system that has the function of receiving an inquiry from a user via a digital device and transmitting the content of the inquiry to a server.

[1768] The "means for analyzing the received inquiry content" is a system that understands the inquiry content and analyzes it to determine its meaning and intent.

[1769] The "means for generating a response based on the analyzed inquiry content" is a system that has the function of automatically generating an appropriate response based on the analysis results.

[1770] The "means for transmitting the generated answer" is a system having a function for returning and displaying the generated answer to the user.

[1771] The "means for saving the inquiry content and the generated response in a database" is a system that has the function of recording and saving the inquiry content and the response to the inquiry in a database.

[1772] The "means for analyzing stored data" is a system that analyzes the history of inquiries and responses stored in a database and predicts and optimizes future inquiries.

[1773] The "means of using a natural language processing model to identify the intent of a query" is a system that has the function of using natural language processing technology to identify the intent of a query from its content.

[1774] The "means for retrieving information from a database based on the intent of a query" is a system that has the function of searching and retrieving related information from a database in accordance with the specified intent.

[1775] "Means for returning the generated answer in JSON format" refers to a system that has the function of returning the generated answer to the user as JSON format data.

[1776] This invention relates to a system that automatically receives and analyzes inquiries from users, and generates and transmits answers. In this system, the roles of the server, terminal, and user are clarified, and specific implementation means are described below.

[1777] System Configuration

[1778] This system consists of the following main components:

[1779] 1. Means of receiving inquiries

[1780] 2. A means of analyzing the content of the received inquiry (using a natural language processing model)

[1781] 3. A means of generating answers based on the parsed query content

[1782] 4. How to send the generated answer (returned in JSON format)

[1783] 5. A means of storing queries and generated responses in a database

[1784] 6. Means of analyzing stored data

[1785] Hardware and software used

[1786] Server: High-performance cloud server (e.g. AWS, Google Cloud)

[1787] Device: The user's smartphone, tablet, or PC

[1788] Database: SQLite or MySQL

[1789] Natural language processing models: SpaCy, Transformers library (e.g., Hugging Face transformers)

[1790] Framework: Python, Flask (Web application framework)

[1791] Program processing overview

[1792] The server receives queries sent from the user's device via an API and temporarily stores them in a database. It then passes the queries to a natural language processing model (e.g., SpaCy or Transformers) to analyze them and identify their intent. It then retrieves relevant information from the database based on the analysis results and generates an appropriate answer. This answer is sent to the user's device in JSON format. The query and the generated answer are also stored in a database for use in subsequent data analysis.

[1793] Specific examples

[1794] For product inquiries:

[1795] A user makes a query through a smartphone application, asking, "Please tell me the stock of product A." This query is sent from the device to the server, which temporarily stores the query in a database. Using a natural language processing model, the intent "to check stock" is identified and the stock information of product A is retrieved from the database. Finally, the server generates a response saying, "There are currently 15 units of product A in stock," and sends it back to the user's device in JSON format.

[1796] Example prompt sentence:

[1797] "Please let me know the stock of product A."

[1798] For shipping inquiries:

[1799] A user makes a query from their PC asking, "Please tell me the delivery status of product B." This query is sent from the device to the server, which temporarily stores the query in a database. Using a natural language processing model, the intent "check delivery status" is identified and the delivery information for product B is retrieved from the database. Finally, the server generates a response stating, "Product B is currently being delivered. The expected delivery date is next Tuesday," and sends it back to the user's device in JSON format.

[1800] Example prompt sentence:

[1801] "Please let me know the delivery status of product B."

[1802] This will enable immediate and appropriate responses to inquiries, and through centralized data management and analysis, will improve customer satisfaction and make future inquiries more efficient.

[1803] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1804] Step 1:

[1805] The user enters an inquiry from their device (smartphone, tablet, or PC) and presses the send button. At this time, the entered inquiry content is sent to the server as an API request.

[1806] Input: The inquiry entered by the user (e.g., "Please tell me the inventory of product A.")

[1807] Output: Query data in API request format

[1808] Step 2:

[1809] The terminal sends the received inquiry to the server via API and temporarily stores it in a database.

[1810] Input: Inquiry data in API request format

[1811] Output: A database containing the query results

[1812] Specific operation: The server receives a request from a terminal and records the contents in a specific table in the database.

[1813] Step 3:

[1814] The server passes the query content stored in the database to a natural language processing (NLP) model to analyze the content.

[1815] Input: Inquiry details saved in the database

[1816] Output: Analysis results from the natural language processing model (intent and extracted entities)

[1817] Specific operation: The server passes the query content to a natural language processing model (e.g., SpaCy or Hugging Face's transformers model), determines the intent as "check inventory," etc., and extracts entities such as product names.

[1818] Step 4:

[1819] The server determines the intent of the query based on the analysis results of the NLP model and retrieves the relevant information from the database.

[1820] Input: Analysis results from natural language processing model

[1821] Output: Result of retrieving related information (e.g., inventory quantity and delivery status)

[1822] Specific behavior: The server generates and executes a query to the database based on the identified intent, and retrieves the relevant data (e.g., product inventory and delivery status).

[1823] Step 5:

[1824] The server generates a response based on the acquired information and sends it to the user's terminal.

[1825] Input: Related information (e.g., inventory quantity and delivery status)

[1826] Output: Generated answer (e.g. "There are 15 units of product A in stock.")

[1827] Specific operation: The server uses the acquired data to create a corresponding answer using a template or plain text generation, and returns it to the user's device in JSON format.

[1828] Step 6:

[1829] The terminal displays the received response to the user.

[1830] Input: Response data sent from the server

[1831] Output: User-visible answers

[1832] Specific operation: The terminal analyzes the received JSON formatted response data and displays it in a format that is easy for the user to understand via the user interface.

[1833] Step 7:

[1834] The server stores the final query and response in a database for later data analysis.

[1835] Input: Final query and generated answer

[1836] Output: A record of the query and answer stored in a database

[1837] Specific operation: The server records pairs of queries and their answers in a database and uses them as data for future trend analysis and system improvement.

[1838] Example prompt sentence:

[1839] "Please let me know the stock of product A."

[1840] 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.

[1841] This invention provides a system that automatically receives and analyzes user inquiries, generates appropriate responses, and returns them with an emotion engine that recognizes the user's emotions, enabling even more sophisticated responses. This system is primarily composed of a server, terminals, and users. Below are the system's program processing and specific examples.

[1842] System Configuration

[1843] This system consists of the following main components:

[1844] 1. Means of receiving inquiries

[1845] 2. Means of analyzing the content of received inquiries

[1846] 3. A means of generating answers based on the parsed query content

[1847] 4. A means of sending the generated answer

[1848] 5. A means of storing queries and generated responses in a database

[1849] 6. Means of analyzing stored data

[1850] 7. Emotion engine that recognizes user emotions

[1851] Program processing

[1852] How to receive inquiries

[1853] The user inputs a query using their own terminal and presses the send button, which sends the query to the server.

[1854] The terminal sends the entered inquiry to the server via API.

[1855] A means of analyzing the content of received inquiries

[1856] The server temporarily stores the inquiry received from the terminal in a database.

[1857] The server passes the received query content to a natural language processing (NLP) model to analyze the content.

[1858] Emotion engine that recognizes user emotions

[1859] The server passes the data, along with the query content, to an emotion engine for analyzing the user's emotions.

[1860] The emotion engine analyzes emotions from the user's text and returns the results to the server, identifying emotions such as "anger," "joy," and "sadness."

[1861] A means of generating answers based on the parsed query content

[1862] The server determines the intent of the query and the user's emotions based on the analysis results of the NLP model and the emotion engine.

[1863] For example, if a user inquires, "Please tell me the product's stock status," the server determines the intent as "checking stock" and the emotion as "neutral."

[1864] A means of sending the generated answer

[1865] Based on the determined intent and emotion, the server retrieves relevant information from a database and generates a response in an appropriate tone.

[1866] The server generates a response based on the acquired information and sends it to the user's terminal.

[1867] The terminal displays the received response to the user.

[1868] A means of storing queries and generated responses in a database

[1869] The server permanently stores the query content and the results of sentiment analysis in a database, along with the final answer.

[1870] A means of analyzing stored data

[1871] The server periodically analyzes the accumulated inquiry, response, and sentiment analysis data to extract inquiry trends and patterns.

[1872] Based on the analysis results, we will build a system that proactively provides relevant information that responds to future inquiries and emotional states.

[1873] Specific examples

[1874] Product inquiries

[1875] User: The user enters "Is product A in stock?" and presses the send button.

[1876] Terminal: The terminal sends this query to the server.

[1877] Server: The server temporarily stores the query content in a database and passes it to the natural language processing model.

[1878] Server: The NLP model determines the intent, "Check inventory," and the emotion engine determines the emotion, "Neutral."

[1879] Server: The server retrieves the stock information of product A from the database, generates a response in a neutral tone such as "Currently, there are 15 units of product A in stock," and sends it to the user's device.

[1880] Terminal: The terminal displays this answer to the user.

[1881] Server: The final query, response, and sentiment analysis results are stored in a database for later data analysis.

[1882] Emotion-conscious response to inquiries

[1883] User: The user types, "The delivery of product B is delayed. What's going on?" and submits.

[1884] Terminal: The terminal sends this query to the server.

[1885] Server: The server temporarily stores the query content in a database and passes it to the natural language processing model.

[1886] Server: The NLP model determines the intent, "check delivery status," and the emotion engine determines the emotion, "anger."

[1887] Server: The server retrieves the delivery status of Product B from the delivery system database, generates a polite response saying, "Product B is currently being delivered. We apologize for any inconvenience. The expected delivery date is next Tuesday," and sends it to the user's device.

[1888] Terminal: The terminal displays this answer to the user.

[1889] Server: The final query, response, and sentiment analysis results are stored in a database for later data analysis.

[1890] This allows the system to respond to inquiries taking emotions into consideration, contributing to improved customer satisfaction.

[1891] The processing flow will be explained below.

[1892] Step 1:

[1893] User: Enter the inquiry details and click the send button.

[1894] For example, enter "Please let me know the product availability."

[1895] Step 2:

[1896] Terminal: Sends the entered inquiry to the server.

[1897] Specifically, the inquiry data is sent via the API using an HTTP POST request.

[1898] Step 3:

[1899] Server: Temporarily stores the received inquiry in a database.

[1900] Specifically, the query content is recorded in the query table using an SQL insert statement.

[1901] Example: INSERT INTO inquiries (user_id, inquiry_text) VALUES ('1234', 'What is the product availability?')

[1902] Step 4:

[1903] Server: Passes the received query content to a natural language processing (NLP) model for analysis.

[1904] Specifically, the query is passed to the NLP model API and the analysis results are received in JSON format.

[1905] Example: "Product availability" is parsed as "Intent: Check inventory."

[1906] Step 5:

[1907] Server: Passes the query content to the emotion engine and analyzes the user's emotions.

[1908] Specifically, the query is passed to the emotion engine's API, and the results of the emotion analysis are received in JSON format.

[1909] Example: "What is the product availability?" is parsed as "Neutral."

[1910] Step 6:

[1911] Server: Determines the intent of the query and the user's emotions based on the analysis results of the NLP model and the emotion engine.

[1912] For example, "Intention: Check inventory" and "Emotion: Neutral" are determined.

[1913] Step 7:

[1914] Server: Based on the intent of the query, retrieves the necessary information from the database.

[1915] Specifically, an SQL query is executed to obtain the stock quantity of the relevant product from the inventory table.

[1916] Example: SELECT stock_quantity FROM products WHERE product_name = 'Product A'

[1917] Step 8:

[1918] Server: Generates answers based on the acquired information and the user's sentiment.

[1919] For example, frame your response in a neutral tone, saying, "We currently have 15 units of product A in stock."

[1920] Step 9:

[1921] Server: Sends the generated answer to the user's device.

[1922] Specifically, the answer is sent as an HTTP response.

[1923] Step 10:

[1924] Terminal: Displays the received answer to the user.

[1925] For example, the chat screen in your browser or app might display "Currently, there are 15 units of product A in stock."

[1926] Step 11:

[1927] Server: The inquiry content and sentiment analysis results, along with the final answer, are permanently stored in a database.

[1928] Use SQL insert or update statements to store queries and their answers.

[1929] Example: UPDATE inquiries SET response = 'Currently, there are 15 units of product A in stock', emotion = 'Neutral' WHERE inquiry_id = '5678'

[1930] Step 12:

[1931] Server: Periodically analyzes the accumulated data to identify trends and patterns in inquiries.

[1932] Specifically, data analysis tools such as Python and R are used to analyze past inquiries in terms of distribution and frequency, as well as emotional trends.

[1933] The above is the specific flow of processing the inquiry content, generating and providing an answer taking the user's feelings into consideration. This system will improve the efficiency of inquiry responses and contribute to improving customer satisfaction.

[1934] Example 2

[1935] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1936] Conventional inquiry systems only provided standardized answers without considering the user's feelings, resulting in low user satisfaction and the inability to respond appropriately to complaints and urgent inquiries in particular. In addition, the stored data was treated as a simple log, and could not be used to proactively respond to future inquiries or for improvements.

[1937] 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.

[1938] In this invention, the server includes means for receiving an inquiry, means for analyzing the content of the received inquiry, means for recognizing the user's emotion, means for generating an answer based on the analyzed content of the inquiry and the recognized emotion, means for transmitting the generated answer, means for saving the content of the inquiry and the generated answer in a database, and means for analyzing the saved data. This makes it possible to generate an appropriate answer that takes the user's emotion into consideration, and further makes it possible to proactively respond to future inquiries by utilizing the accumulated data.

[1939] The "means for receiving an inquiry" is a device or process for receiving the content of an inquiry sent from a user to the server via the terminal.

[1940] "Means for analyzing the received query content" refers to a device or process for understanding the received query text and identifying its intent, often using natural language processing (NLP) techniques.

[1941] The "means for recognizing user emotions" is a device or process for determining the user's emotions from the query text, and often utilizes an emotion analysis engine.

[1942] The "means for generating an answer based on the analyzed query content and the recognized sentiment" is a device or process for generating an appropriate answer based on the intent of the query and the recognized sentiment.

[1943] A "means for transmitting a generated answer" is a device or process for transmitting a generated answer to a user.

[1944] The "means for storing the inquiry contents and generated answers in a database" is a device or process for storing all the inquiry contents and the generated answers thereto in a database.

[1945] A "means for analyzing stored data" is a device or process that uses stored inquiry and response data to extract trends and patterns to improve future inquiry responses.

[1946] "Natural Language Processing (NLP)" is a technology that enables machines to understand, analyze, and generate human language.

[1947] An "emotion analysis engine" is an algorithm or software for identifying a user's emotions from text data.

[1948] A "database" is an information storage system for storing queries and generated responses.

[1949] This invention enables more advanced responses by combining a system that receives and analyzes user inquiries, generates appropriate responses, and responds with an emotion engine that recognizes the user's emotions. The system consists of the following main components:

[1950] System Configuration

[1951] This system is mainly composed of a server, terminals, and users. Specifically, it uses the following devices and software:

[1952] server

[1953] Hardware: A server machine equipped with a high-performance processor, memory, and storage

[1954] software:

[1955] Database: MySQL, PostgreSQL, MongoDB, Elasticsearch, etc.

[1956] Natural Language Processing (NLP) libraries: spaCy, NLTK

[1957] Sentiment analysis engine: Microsoft Azure's Text Analytics API, IBM Watson's Natural Language Understanding API

[1958] Text generation model: OpenAI's GPT-3

[1959] Terminal

[1960] Hardware: Smartphones, tablets, personal computers

[1961] Software: Web browsers (Google Chrome, Safari, etc.), mobile applications

[1962] User

[1963] Interface: inquiry form, etc.

[1964] Program processing

[1965] Receive inquiries

[1966] The user uses their device (smartphone or computer) to enter the details of their inquiry. For example, they might enter "Please tell me the product's stock status" in a web browser and press the send button. The device then sends this inquiry to the server via an API (RESTful API or GraphQL).

[1967] Analyze the incoming inquiries

[1968] The server temporarily stores the query received from the device in a database (MySQL or PostgreSQL). It then uses a natural language processing engine (spaCy or NLTK) to analyze the intent of the query. For example, it classifies the text "Please tell me the stock status of the product" as the intent "Check stock."

[1969] Recognize user emotions

[1970] The server sends the received inquiry content to a sentiment analysis engine, which determines the user's emotions (such as "anger," "joy," or "sadness") from the text. Examples of sentiment analysis engines include Microsoft Azure's Text Analytics API and IBM Watson's Natural Language Understanding API.

[1971] Generate an answer

[1972] The server generates a response using a text generation model (OpenAI's GPT-3) based on the intent and sentiment analysis results. For example, it generates a neutral response such as "Currently, there are 15 units of product A in stock."

[1973] Submit your answer

[1974] The server sends the generated answer to the user's terminal, which displays the answer to the user.

[1975] Save your data

[1976] The server stores the query content and sentiment analysis results along with the final answer in a database (MongoDB or Elasticsearch).

[1977] Analyze the data

[1978] The stored data is periodically analyzed and data analysis tools such as Apache Hadoop and Google BigQuery are used to extract trends and patterns in inquiries, which can then be used to proactively respond to future inquiries.

[1979] Specific examples

[1980] Product inquiries

[1981] A user enters "Do you have product A in stock?" into an inquiry form in Google Chrome and presses the submit button. The device sends this inquiry to the server via a RESTful API. The server temporarily saves the inquiry content in MySQL and uses spaCy to extract the intent "to check stock." It then uses Microsoft Azure's Text Analytics to determine the sentiment of "neutral." It retrieves the stock information for product A from the database and generates a response in a neutral tone, such as "There are currently 15 units of product A in stock." This response is then sent back to the device in JSON format. The device displays this response on the user interface and saves the final inquiry, response, and sentiment analysis results in a MySQL database.

[1982] Inquiries about delivery status

[1983] The user types and submits the query in Safari, "The delivery of Product B is delayed. What's going on?" The device sends this query to the server using the GraphQL API. The server temporarily stores the query content in PostgreSQL and uses NLTK to extract the intent, "Check delivery status." It then uses IBM Watson's Natural Language Understanding to determine the emotion, "anger." It retrieves the delivery status of Product B from the delivery system database and generates a polite response, saying, "Product B is currently being delivered. We apologize for any inconvenience. The expected delivery date is next Tuesday." The response returned to the device in JSON format is displayed on the user interface, and the final query, response, and emotion analysis results are saved in the PostgreSQL database.

[1984] This allows the system to respond to inquiries taking into account the user's emotions, contributing to improved customer satisfaction.

[1985] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1986] Step 1: Receiving an inquiry

[1987] The user uses their own device (smartphone or computer) to enter the details of their inquiry and press the send button. Specifically, the user opens a web browser (e.g., Google Chrome), enters "Please let me know the product's stock status" into the inquiry form, and clicks send.

[1988] Input: User-supplied query text

[1989] Output: Data sent from the device to the server (inquiry details, user ID, timestamp, etc.)

[1990] The device sends the entered inquiry to the server via an API (e.g., RESTful API). At this time, the sent data includes metadata such as the user ID and timestamp.

[1991] Step 2: Analyze the query

[1992] The server temporarily stores the query received from the terminal in a database (e.g., MySQL). This temporary storage is performed as a transaction to maintain data consistency.

[1993] Input: Inquiry data sent from the terminal

[1994] Output: Temporarily saved inquiry data

[1995] The server uses a natural language processing (NLP) engine (e.g., spaCy) to analyze the query. Specifically, it extracts the intent from the query text. For example, the text "Please tell me the product's stock status" is classified as the intent "Check stock."

[1996] Step 3: Recognize the user's emotions

[1997] The server sends the received inquiry content to a sentiment analysis engine (e.g., Microsoft Azure's Text Analytics API) to analyze the user's sentiment.

[1998] Input: Inquiry details

[1999] Output: Emotion analysis result (e.g. "neutral")

[2000] The sentiment analysis engine determines the user's sentiment from the query text, for example, "Please tell me the product availability status," and identifies the sentiment "neutral."

[2001] Step 4: Generate an answer

[2002] The server generates an answer using a text generation model (e.g., OpenAI's GPT-3) based on the results of intent analysis and sentiment analysis.

[2003] Input: Intent analysis results (e.g., "Check inventory"), sentiment analysis results (e.g., "Neutral")

[2004] Output: Generated answer (e.g. "Currently, there are 15 units of product A in stock.")

[2005] The server retrieves the necessary information from a database (e.g., MongoDB) based on the intent. For example, for the intent "check stock," it retrieves stock information for product A. It then uses a text generation model to generate an answer in an appropriate tone. For example, it generates an answer such as "Currently, there are 15 units of product A in stock."

[2006] Step 5: Submit your response

[2007] The server then sends the generated response to the user's device, often in JSON format.

[2008] Input: Generated response data

[2009] Output: Data sent to the terminal

[2010] The terminal displays the received answer on the user interface, allowing the user to check the answer on their screen.

[2011] Step 6: Save your data

[2012] The server stores the final answer, inquiry details, and sentiment analysis results in a database (e.g., Elasticsearch).

[2013] Input: Final response, inquiry details, sentiment analysis results

[2014] Output: Data stored in the database

[2015] The saved data can be used for later data analysis.

[2016] Step 7: Analyze the data

[2017] The server periodically analyzes the accumulated inquiry, response, and sentiment analysis data.

[2018] Input: Data of inquiries, responses, and sentiment analysis results stored in the database

[2019] Output: Analysis results (inquiry trends and patterns)

[2020] Specifically, Apache Hadoop and Google BigQuery are used to process data and extract trends and patterns. The analysis results are then used to respond to future inquiries. For example, if there are many inquiries about a particular season or product, that information can be prepared in advance.

[2021] (Application example 2)

[2022] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[2023] For modern online shopping sites, responding to user inquiries quickly and appropriately is an important issue that directly contributes to improving customer satisfaction. However, conventional inquiry response systems do not adequately take user emotions into account, resulting in inconsistent response quality. In particular, when a user is dissatisfied or angry, it is difficult to respond appropriately based on that emotion. Another challenge is efficiently analyzing the content of a large number of inquiries and generating appropriate answers.

[2024] 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.

[2025] In this invention, the server includes means for receiving an inquiry, means for analyzing the received inquiry, means for generating a response based on the analyzed inquiry, means for transmitting the generated response, means for saving the inquiry and the generated response in a database, means for analyzing the saved data, means including an emotion engine for recognizing the user's emotion, and means for adjusting the tone of the response in accordance with the user's emotion, thereby enabling a prompt and appropriate response to an inquiry that takes the user's emotion into consideration.

[2026] The "means for receiving an inquiry" is a function by which the system receives and processes an inquiry from a user.

[2027] The "means for analyzing the content of the received inquiry" is a function for understanding the content of the received inquiry and analyzing it as structured data.

[2028] The "means for generating a response based on the analyzed inquiry content" is a function that automatically generates an appropriate response based on the analysis results.

[2029] The "means for sending the generated answer" is a function for returning the generated answer to the user.

[2030] The "means for saving the inquiry and the generated response in a database" is a function for saving the inquiry and the response as a record in a database.

[2031] The "means for analyzing stored data" is a function for analyzing stored inquiry and response data and extracting significant insights and patterns.

[2032] The "emotion engine that recognizes user emotions" is an engine that analyzes and identifies emotions from the content of a user's inquiry.

[2033] The "means for adjusting the tone of the response according to the user's emotion" is a function for automatically adjusting the expression and tone of the response according to the emotion specified by the user.

[2034] The present invention is a system for responding promptly and appropriately to inquiries from users on an online shopping site. This system analyzes the content of the user's inquiry and recognizes their emotions, thereby generating a response in a tone that corresponds to their emotions and improving the quality of the response.

[2035] System Configuration

[2036] The system mainly consists of the following components:

[2037] 1. Means of receiving inquiries

[2038] 2. Means of analyzing the content of received inquiries

[2039] 3. A means of generating answers based on the parsed query content

[2040] 4. A means of sending the generated answer

[2041] 5. A means of storing queries and generated responses in a database

[2042] 6. Means of analyzing stored data

[2043] 7. Emotion engine that recognizes user emotions

[2044] 8. How to adjust the tone of your response based on the user's emotions

[2045] Hardware and Software

[2046] Hardware used:

[2047] server

[2048] User device (smartphone or PC)

[2049] Software used:

[2050] Python

[2051] Natural Language Processing Library (Hugging Face Transformers)

[2052] Sentiment analysis model (BERT)

[2053] Database System

[2054] Data processing and calculation

[2055] 1. Means of receiving inquiries:

[2056] The user uses a smartphone or computer to enter and submit an inquiry.

[2057] The terminal sends the entered inquiry to the server via API.

[2058] 2. How to analyze the received inquiry:

[2059] The server temporarily stores the inquiry received from the terminal in a database.

[2060] The data stored in the database is analyzed using natural language processing models.

[2061] 3. Emotion engine that recognizes user emotions:

[2062] The server passes the data along with the query to the emotion engine.

[2063] The emotion engine analyzes the emotion from the query and returns the result to the server. The emotion may be "anger," "joy," or "sadness," for example.

[2064] 4. Means of generating an answer based on the parsed query:

[2065] The server determines the intent of the query and the user's emotions based on the analysis results of the NLP model and the emotion engine.

[2066] 5. Means for sending generated answers:

[2067] The server generates a response in an appropriate tone based on the determined intent and emotion.

[2068] The generated answer is sent to the user's terminal.

[2069] 6. How to store queries and generated responses in a database:

[2070] The inquiry details and the results of sentiment analysis are stored in a database along with the final answer.

[2071] Examples of specific examples and prompts

[2072] For example, if a user asks, "Product B hasn't arrived yet. What's going on?", the following process will be performed:

[2073] User inquiry: "I haven't received my delivery of product B yet. What's going on?"

[2074] Emotion engine judgement: "Anger"

[2075] System response: "I see you're angry. We'll deal with it right away."

[2076] Example prompt sentence:

[2077] User Question: "I haven't received my delivery of item B yet, what's going on?"

[2078] System emotion detection: "Anger"

[2079] Generate an appropriate response.

[2080] This allows for prompt and appropriate response to inquiries that take into account the user's feelings.

[2081] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[2082] Step 1:

[2083] The user enters an inquiry on a smartphone or PC. When the user enters the inquiry content as text and presses the send button, the device sends the inquiry data to the server via the API.

[2084] Input: The query text entered by the user.

[2085] Output: The query data sent to the server.

[2086] Step 2:

[2087] The server temporarily stores the received inquiry in a database.

[2088] Input: Enquiry data received through API.

[2089] Output: Query data temporarily stored in a database.

[2090] Step 3:

[2091] The server passes the received query content to a natural language processing (NLP) model for analysis, which analyzes the text and extracts the intent of the query.

[2092] Input: Query data stored in the database.

[2093] Output: Analysis result data including query intent.

[2094] Step 4:

[2095] The server passes the query content and sentiment analysis data to the sentiment engine, which analyzes the sentiment from the user's text and returns the results to the server.

[2096] Input: Query text.

[2097] Output: Emotion analysis results from the emotion engine (e.g., "anger," "joy," "sadness," etc.).

[2098] Step 5:

[2099] The server determines the intent of the query and the user's emotions based on the analysis results of the NLP model and the emotion engine.

[2100] Input: Analysis results of the NLP model and the emotion engine.

[2101] Output: Analysis result data including intent and sentiment.

[2102] Step 6:

[2103] Based on the determined intent and emotion, the server retrieves relevant information from a database and generates a response in an appropriate tone.

[2104] Input: Analysis result data including intent and sentiment.

[2105] Output: Answer text generated in the appropriate tone.

[2106] Step 7:

[2107] The server sends the generated answer to the user's terminal, which then displays the received answer to the user.

[2108] Input: The generated answer text.

[2109] Output: The answer text that is displayed on the user's terminal.

[2110] Step 8:

[2111] The server stores the query and the results of the sentiment analysis in a database, along with the final answer.

[2112] Input: Query, generated answer, sentiment analysis results.

[2113] Output: The final data stored in the database.

[2114] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[2115] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[2116] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[2117] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[2118] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[2119] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[2120] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[2121] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[2122] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[2123] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[2124] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[2125] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[2126] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[2127] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[2128] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[2129] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[2130] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[2131] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[2132] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[2133] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[2134] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[2135] The following is further disclosed regarding the above embodiment.

[2136] (Claim 1)

[2137] means for receiving inquiries;

[2138] means for analyzing the received inquiry;

[2139] means for generating an answer based on the analyzed query content;

[2140] means for transmitting the generated response;

[2141] means for storing the queries and generated responses in a database;

[2142] means for analyzing the stored data;

[2143] A system including:

[2144] (Claim 2)

[2145] 2. The system of claim 1, wherein natural language processing (NLP) is used to analyze the content of the query.

[2146] (Claim 3)

[2147] 10. The system of claim 1, further comprising means for generating and providing useful information based on analysis of the stored data.

[2148] "Example 1"

[2149] (Claim 1)

[2150] means for receiving inquiries;

[2151] means for analyzing the received inquiry;

[2152] means for generating an answer based on the analyzed query content;

[2153] means for transmitting the generated response;

[2154] means for storing the inquiry content and the generated response in an information storage medium;

[2155] means for analyzing the stored data;

[2156] A system including:

[2157] (Claim 2)

[2158] 2. The system of claim 1, wherein natural language processing (NLP) is used to analyze the content of the query.

[2159] (Claim 3)

[2160] 10. The system of claim 1, further comprising means for generating and providing useful information based on analysis of the stored data.

[2161] (Claim 4)

[2162] 2. The system of claim 1, further comprising a terminal for forwarding inquiries sent by users and receiving and displaying replies.

[2163] (Claim 5)

[2164] 2. The system according to claim 1, further comprising means for permanently storing the inquiry contents and responses and utilizing them for later data analysis.

[2165] "Application Example 1"

[2166] (Claim 1)

[2167] means for receiving inquiries;

[2168] means for analyzing the received inquiry;

[2169] means for generating an answer based on the analyzed query content;

[2170] means for transmitting the generated response;

[2171] means for storing the queries and generated responses in a database;

[2172] means for analyzing the stored data;

[2173] a means for using a natural language processing model to identify the intent of the query;

[2174] A means for retrieving information from a database based on the query intent;

[2175] A means to return the generated answer in JSON format;

[2176] A system including:

[2177] (Claim 2)

[2178] 2. The system of claim 1, wherein natural language processing (NLP) is used to analyze the content of the query.

[2179] (Claim 3)

[2180] 10. The system of claim 1, further comprising means for generating and providing useful information based on analysis of the stored data.

[2181] "Example 2: Combining Emotion Engines"

[2182] (Claim 1)

[2183] means for receiving inquiries;

[2184] means for analyzing the received inquiry;

[2185] means for recognizing a user's emotion;

[2186] means for generating an answer based on the analyzed query content and the recognized sentiment;

[2187] means for transmitting the generated response;

[2188] means for storing the queries and generated responses in a database;

[2189] means for analyzing the stored data;

[2190] A system including:

[2191] (Claim 2)

[2192] 2. The system of claim 1, wherein natural language processing (NLP) is used to analyze the content of the query.

[2193] (Claim 3)

[2194] 2. The system according to claim 1, wherein an emotion analysis engine is used to recognize the user's emotions.

[2195] (Claim 4)

[2196] 10. The system of claim 1, further comprising means for generating and providing useful information based on analysis of the stored data.

[2197] "Application example 2 when combining emotion engines"

[2198] (Claim 1)

[2199] means for receiving inquiries;

[2200] means for analyzing the received inquiry;

[2201] means for generating an answer based on the analyzed query content;

[2202] means for transmitting the generated response;

[2203] means for storing the queries and generated responses in a database;

[2204] means for analyzing the stored data;

[2205] means including an emotion engine for recognizing an emotion of a user;

[2206] A means for adjusting the tone of the response depending on the user's emotions;

[2207] A system including:

[2208] (Claim 2)

[2209] 2. The system according to claim 1, wherein natural language processing is used to analyze the content of the inquiry.

[2210] (Claim 3)

[2211] 10. The system of claim 1, further comprising means for generating and providing useful information based on analysis of the stored data. [Explanation of symbols]

[2212] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means for receiving inquiries; means for analyzing the received inquiry; means for generating an answer based on the analyzed query content; means for transmitting the generated response; means for storing the queries and generated responses in a database; means for analyzing the stored data; A system including:

2. 2. The system of claim 1, wherein natural language processing (NLP) is used to analyze the content of the query.

3. 10. The system of claim 1, further comprising means for generating and providing useful information based on analysis of the stored data.

Images