system

Abstract

We provide the system. [Solution] A means for receiving inquiries from users, A means for analyzing the aforementioned query using a natural language processing model, A means for searching for relevant answers from a pre-configured database based on the analyzed query, Means for providing the searched answers to the user, A means for storing the aforementioned inquiry and the aforementioned answer in a database, A system that includes means for training a machine learning model using data stored in the aforementioned database to improve responses to new inquiries.

Description

Technical Field

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

Background Art

[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 an explanation of a character of the chatbot, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance that responds to the user utterance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In a modern business environment, efficient and accurate inquiry response is important. However, conventional manual response systems are time-consuming and labor-intensive, and human errors are inevitable. Also, since the inquiry content covers a wide range, it is difficult to provide an appropriate answer immediately. Furthermore, there is a lack of means for automating specific operations (for example, shipping of alternative machines, new contracts, model changes). Therefore, there is an increasing need for a new system to improve the overall business efficiency.

Means for Solving the Problems

[0005] The present invention provides a system that efficiently processes user inquiries and provides accurate answers. This system includes the following means: means for receiving user inquiries; means for analyzing the inquiries using a natural language processing model; means for searching for relevant answers from a pre-configured database based on the analyzed inquiries; and means for providing the retrieved answers to the user. It also includes means for storing the inquiries and answers in a database, and means for training a machine learning model using the data stored in the database to improve the response to new inquiries. Furthermore, it includes means for the user to request specific operations and for the server to call an appropriate API to perform the processing. In this way, the present invention achieves improved operational efficiency and cost reduction.

[0006] A "user" is an individual or legal entity that uses the system to make an inquiry.

[0007] An "inquiry" refers to a question or request submitted by a user through the system.

[0008] "Natural language processing" is a technology that enables computers to understand and analyze human language.

[0009] A "model" is a computer program or algorithm built to achieve a specific function.

[0010] "Analysis" is the process of understanding the content of a query and extracting its key elements.

[0011] A "database" is a digital record used to systematically store information about inquiries and answers.

[0012] "Answer" refers to the information or explanation provided in response to an inquiry.

[0013] A "server" is a central computer system that handles query processing and data storage and management.

[0014] The "terminal" is a device (such as a PC or a mobile device) through which a user inputs an inquiry and receives a response.

[0015] "Machine learning" is a technology by which a computer improves its performance using empirical data.

[0016] "Training a model" is a process of supplying data to a machine learning algorithm to improve the performance of that model.

[0017] An "API" is an interface for using functions between different software.

[0018] A "replacement device" is a device that is temporarily provided in place of a failed device.

[0019] A "new contract" is an official agreement for a user to start using a new service or product.

[0020] "Device model change" is a process by which a user changes from the currently used device to a new model.

[0021] "Automating" means executing operations by machines or software without human intervention.

[0022] "Processing" means that a system executes a series of operations based on a user's request.

Brief Description of the Drawings

[0023] [Figure 1] It is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] It is a conceptual diagram showing an example of the main functions of a data processing device and a smart device according to the first embodiment. [Figure 3]It is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] It is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] It is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] It is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] It is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] It is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] It shows an emotion map to which a plurality of emotions are mapped. [Figure 10] It shows an emotion map to which a plurality of emotions are mapped. [Figure 11] It is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Example 2 when an emotion engine is combined. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when an emotion engine is combined.

Embodiments for Carrying Out the Invention

[0024] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0025] First, the language used in the following description will be explained.

[0026] In the following embodiments, the signed processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Furthermore, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include CPU (Central Processing Unit), GPU (Graphics Processing Unit), GPGPU (General-Purpose computing on Graphics Processing Units), and APU (Accelerated Processing Unit).

[0027] In the following embodiments, signed RAM (Random Access Memory) is a memory that temporarily stores information and is used as work memory by the processor.

[0028] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0029] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0030] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0031] [First Embodiment]

[0032] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0033] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0034] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0035] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0036] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0037] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0038] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0039] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0040] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0041] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0042] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0043] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0044] This invention is a system that efficiently receives and analyzes user inquiries and provides appropriate responses. This achieves improved operational efficiency and cost reduction. Specific embodiments for implementing this invention are described below.

[0045] System Configuration

[0046] The system consists of user terminals, a server, and a database. Users make queries using the terminals, the server receives and analyzes them, searches for relevant answers, and provides them to the user.

[0047] The user enters an inquiry.

[0048] Users use their devices (e.g., PCs or smartphones) to input language-based inquiries into the system. This allows them to ask questions about seasonal products and services, how to set up an account, and more.

[0049] The device sends the query to the server.

[0050] The terminal sends the entered query content to the server. The server receives this via the API endpoint and begins parsing the query.

[0051] The server parses the query.

[0052] The server analyzes the query using a natural language processing (NLP) model. For example, the NLP model extracts key keywords and context from the text to understand the intent of the query. Techniques such as morphological analysis and syntax analysis are used in this step.

[0053] The server searches for relevant answers.

[0054] The server searches the database for relevant answers based on the analyzed query. The database contains pre-configured FAQs and manuals, and the search engine uses this information to return the most suitable answer.

[0055] The server sends the response to the terminal.

[0056] The server sends the search results to the device. The device displays this to the user as a chat window or notification. This allows the user to quickly obtain the appropriate information.

[0057] The server stores queries and answers in a database.

[0058] The server stores user inquiries and their corresponding responses in a database. This data is used to improve the accuracy of automated responses to future inquiries.

[0059] The server trains the machine learning model.

[0060] The server uses accumulated data to train a machine learning model. This model improves the response generation algorithm to handle new queries, resulting in more accurate automated responses.

[0061] Specific example

[0062] For example, if a user submits a query such as "Please tell me the conditions for applying maintenance," the server analyzes the query and extracts the relevant keywords "maintenance," "application," and "conditions." The server then searches its database for the appropriate answer, such as "Maintenance services are only applicable within the product warranty period," and sends it to the user's terminal. This allows the user to quickly obtain the necessary information.

[0063] Automated operation

[0064] Furthermore, if a user requests specific actions such as sending a replacement device, signing a new contract, or changing models, the server will call the appropriate API to automate these operations. For example, if a user requests "Please send a replacement device," the server will integrate with the customer management system and automatically handle the process of sending the replacement device.

[0065] In this way, the system of the present invention efficiently processes user inquiries and automates appropriate responses and operations, thereby achieving improved operational efficiency and cost reduction.

[0066] The following describes the processing flow.

[0067] Step 1:

[0068] The user enters the inquiry. The user uses a device (e.g., a PC or smartphone) to input questions or requests to the system in text format.

[0069] Step 2:

[0070] The device sends the query to the server. The device sends the entered query content to the server via the API endpoint.

[0071] Step 3:

[0072] The server receives the query. The server retrieves the text data received from the API endpoint.

[0073] Step 4:

[0074] The server analyzes the query. The server uses a natural language processing (NLP) model to analyze the query content. Specifically, it uses morphological analysis and syntax analysis to extract key keywords and context.

[0075] Step 5:

[0076] The server extracts keywords. It extracts important keywords, mainly nouns and verbs, from the query text.

[0077] Step 6:

[0078] The server searches the FAQ database. Based on the extracted keywords, the server searches the FAQ database for relevant answers.

[0079] Step 7:

[0080] The server selects the most appropriate answer. If multiple answer choices exist, the server selects the most relevant answer.

[0081] Step 8:

[0082] The server generates the answer. Based on the selected answer, it generates the final text to be provided to the user.

[0083] Step 9:

[0084] The server sends the response to the device. An API endpoint is used to send the generated response text to the device.

[0085] Step 10:

[0086] The device displays the response to the user. The device displays the received response to the user in a chat window or notification format.

[0087] Step 11:

[0088] The server stores queries and answers in a database. The query content and the corresponding answers are recorded in the database and used for future query processing.

[0089] Step 12:

[0090] The server trains a machine learning model. Using accumulated data, the machine learning model is periodically trained to improve the accuracy of automated responses to inquiries.

[0091] Step 13:

[0092] The user requests a specific action. Specific requests, such as sending a replacement device, signing a new contract, or changing models, are entered in text format.

[0093] Step 14:

[0094] The server processes the user's request. It calls the appropriate API and performs the specific action required for the request (e.g., arranging for a replacement device to be shipped).

[0095] Step 15:

[0096] The server sends the processing results to the terminal. A notification indicating that processing is complete is sent to the terminal.

[0097] Step 16:

[0098] The terminal displays the processing results to the user. The terminal displays the processing results to the user in a chat window or notification format.

[0099] Through the steps described above, the system can efficiently process user inquiries and automate appropriate responses and actions.

[0100] (Example 1)

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

[0102] Conventional inquiry handling systems suffered from long response times to user inquiries and low accuracy in providing appropriate answers. Furthermore, the need for numerous manual operations to find appropriate answers prevented sufficient efficiency improvements and cost reductions. The objective of this invention is to solve these problems and provide users with efficient and highly accurate responses.

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

[0104] In this invention, the server includes means for receiving inquiries from users, means for analyzing the inquiries using a natural language processing model, means for sending the input inquiries to the server via an API endpoint, means for searching for relevant answers from a pre-configured data storage based on the analyzed inquiries, means for providing the retrieved answers to the user, means for storing the inquiries and answers in the data storage, and means for training a computational model using the data stored in the data storage to improve the response to new inquiries. This enables a rapid and accurate response to user inquiries, efficient business operations, and cost reduction.

[0105] An "inquiry" is the act of a user inputting information they want to know or a problem they want to solve into a system using natural language.

[0106] A "natural language processing model" is a type of artificial intelligence that analyzes text written in natural language, understands its content and meaning, and processes it.

[0107] An "API endpoint" is part of an application programming interface and is a connection point for different software systems to communicate with each other.

[0108] A "server" is a computer system that provides services to other computers via a network.

[0109] "Data storage" refers to devices and systems for storing digital data, and includes databases and cloud storage.

[0110] "Analysis" is the act of breaking down input data into its components, understanding its content, and identifying its meaning and intent.

[0111] An "answer" refers to the information or solution provided in response to a user's inquiry.

[0112] A "machine learning model" is an algorithm that learns from data and performs tasks such as prediction and classification.

[0113] A "computational model" is a mathematical model used to perform a specific task through numerical calculations and data processing.

[0114] A "programming interface" is a set of definitions and protocols that enable different software programs to work together smoothly.

[0115] "Accumulation" is the act of collecting data and saving it.

[0116] This invention is a system that efficiently receives and analyzes user inquiries and provides appropriate responses. This achieves improved operational efficiency and cost reduction. Specific embodiments of this invention are described below.

[0117] System Configuration

[0118] This system consists of user terminals, a server, and data storage. Users use the terminals to make inquiries, the server receives and analyzes them, searches for relevant answers, and provides them to the user.

[0119] The user enters an inquiry.

[0120] Users use their devices (such as PCs or smartphones) to input natural language queries into the system. For example, they might ask, "I'd like to know about the product's warranty period."

[0121] The device sends the query to the server.

[0122] The terminal sends the entered query content as an HTTP POST request to the server's API endpoint. The server receives this and begins parsing the query.

[0123] The server parses the query.

[0124] The server uses an NLP (Natural Language Processing) model to analyze the query. This analysis utilizes NLP libraries such as spaCy and Hugging Face in Python. For example, morphological analysis is used to extract the keyword "product warranty period."

[0125] The server searches for relevant answers.

[0126] The server searches for relevant answers in data storage based on the extracted keywords. Data storage may include SQL Server or Elasticsearch®. The server finds the most suitable answer, retrieving information such as "The product warranty period is one year."

[0127] The server sends the response to the terminal.

[0128] The server sends the retrieved response to the device in JSON format. The device then displays this response to the user in a chat window or notification, allowing the user to quickly obtain the necessary information.

[0129] The server stores queries and answers in data storage.

[0130] The server records the query content and the corresponding response in a separate data storage. This data will be used to improve future query responses.

[0131] The server trains the machine learning model.

[0132] The server uses data stored in data storage to train machine learning models. For example, it uses Python's Scikit-learn or TENSORFLOW® to improve the accuracy of the models. This enables sophisticated automated responses to new queries.

[0133] Automated operation

[0134] Furthermore, this system can handle specific user requests such as shipping replacement equipment, signing up for new contracts, or changing models. For example, if a user requests "Please ship a replacement unit," the server will call the appropriate programming interface and automatically execute the operation. This programming interface integrates with customer management systems, logistics systems, and other systems to automate necessary procedures.

[0135] Specific example

[0136] For example, if a user asks, "What are the conditions for maintenance?", the server analyzes the inquiry and extracts the keywords "maintenance," "applicability," and "conditions." The server then searches its database for the answer, "Maintenance services are applicable only within the product warranty period," and sends it to the user's terminal. This allows the user to quickly obtain the necessary information.

[0137] Thus, the present invention enables prompt and accurate responses to user inquiries, thereby improving operational efficiency and reducing costs.

[0138] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0139] Step 1:

[0140] The user enters an inquiry.

[0141] Users enter their inquiries in natural language into a text form using a device such as a PC or smartphone. Specifically, they might type "Please tell me about the product's warranty period" and click the submit button. The entered text is temporarily stored on the user's device.

[0142] Step 2:

[0143] The device sends the query to the server.

[0144] When the user clicks the submit button, the device sends the entered query content to the server as an HTTP POST request. For example, it sends a request with data in JSON format like the following:

[0145] json

[0146] {

[0147] "query": "Please tell me about the product's warranty period."

[0148] }

[0149] The server receives this request and temporarily holds the query details in memory.

[0150] Step 3:

[0151] The server parses the query.

[0152] The server inputs the received query into a natural language processing (NLP) model and begins analysis. Specifically, it uses a Python NLP library (e.g., spaCy) to perform morphological analysis and extract keywords and phrases from the text. For example, the keyword "product warranty period" might be extracted. The input is the received query, and the output is the extracted keywords.

[0153] Step 4:

[0154] The server searches for relevant answers.

[0155] The server searches for information in data storage based on keywords extracted from the analyzed query. It uses SQL Server, Elasticsearch, etc., to retrieve relevant answers. Specifically, it executes SQL queries like the following:

[0156] sql

[0157] SELECT answer FROM faq WHERE question LIKE '%warranty period%'

[0158] The input is the extracted keywords, and the output is the appropriate answer. This query yields the answer, "The product warranty period is one year."

[0159] Step 5:

[0160] The server sends the response to the terminal.

[0161] The server sends the retrieved response to the terminal as a JSON response. Specifically, it sends it in the following format:

[0162] json

[0163] {

[0164] "Answer": "The product warranty period is one year."

[0165] }

[0166] The terminal receives this data and displays it in an inquiry form or a dedicated window. The output is displayed on the user's terminal.

[0167] Step 6:

[0168] The user receives the response.

[0169] The user checks the answer displayed on their device. Specifically, they can see a message such as "The product warranty period is one year" and get the appropriate answer to their inquiry.

[0170] Step 7:

[0171] The server stores queries and answers in a database.

[0172] The server stores the query details and corresponding answers in a database. This data is used to improve future responses. Specifically, the data is stored as follows:

[0173] sql

[0174] INSERT INTO logs (query, answer) VALUES ('Please tell me about the product's warranty period', 'The product's warranty period is 1 year')

[0175] The input consists of the inquiry content and the response, and the output is the storage of the data in the database.

[0176] Step 8:

[0177] The server trains the machine learning model.

[0178] The server trains a machine learning model using historical query data stored in the database. Specifically, it uses Python's Scikit-learn and TensorFlow to improve the model's accuracy. This model is designed to increase the accuracy of generating answers to new queries. The input is historical query data, and the output is the updated machine learning model.

[0179] (Application Example 1)

[0180] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0181] In modern brick-and-mortar stores, it is necessary to respond quickly and accurately to a wide range of customer inquiries, but relying on manual responses makes efficient and cost-effective operation difficult. Furthermore, if customers cannot obtain information about products and services immediately, their willingness to purchase may decrease. In addition, manual processing is time-consuming when performing specific actions based on user inquiries, leading to decreased customer satisfaction. This invention aims to solve these problems and achieve efficiency and automation in customer service in physical stores.

[0182] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0183] In this invention, the server includes means for receiving inquiries from users, means for analyzing inquiries using a natural language processing model, and means for searching for relevant answers from a pre-configured database based on the analyzed inquiries. This enables the rapid provision of answers to customer inquiries. Furthermore, by including means for customer interaction in the transport device, it becomes possible to automate tasks such as guiding customers to product locations and assisting with purchase procedures. This streamlines customer service in physical stores and improves customer satisfaction.

[0184] definition statement

[0185] "Means of receiving user inquiries" refers to the function of receiving any inquiries that users input into the system via their terminals.

[0186] "Methods for analysis using natural language processing models" refers to a function that analyzes input queries using natural language processing technology, extracts key keywords and context, and understands the user's intent.

[0187] "Means of searching for relevant answers from a pre-configured database" refers to a function that searches for the most suitable answer from a pre-configured database based on the analyzed query content.

[0188] "Means of providing information to the user" refers to functions that notify or display the searched answers on the user's device.

[0189] "Means for storing inquiries and answers in a database" refers to a function that saves user inquiries and the corresponding answers in a database.

[0190] "Means of training machine learning models to improve responses to new inquiries" refers to a function that uses accumulated data to train machine learning models and improve the accuracy and effectiveness of responses to future inquiries.

[0191] "Means by which a transport device responds to customers" refers to a function in which a transport device or robot directly responds to customers based on the analyzed inquiry content.

[0192] "Means for automatically generating answers to new inquiries" refers to a function that uses machine learning models to automatically generate appropriate answers.

[0193] "Means of calling an appropriate application program interface to perform processing" refers to a function that automatically executes processes such as shipping replacement equipment, signing new contracts, and changing models by calling an appropriate application program interface in response to a user's operation request.

[0194] Modes for carrying out the invention

[0195] This invention is a system that efficiently receives and analyzes user inquiries and provides appropriate responses. This system is particularly aimed at improving the efficiency of customer service in physical stores. Specific embodiments for implementing this invention are described below.

[0196] System Configuration

[0197] The system consists of user terminals, a server, and a database. Users make queries using the terminals, the server receives and analyzes them, searches for relevant answers, and provides them to the user. Furthermore, a transport device responds to customers as needed.

[0198] The user enters an inquiry.

[0199] Users use a device (e.g., smartphone, tablet, or transport device) to input language-based inquiries into the system. This allows them to ask questions such as the location of products or detailed information about services.

[0200] The device sends the query to the server.

[0201] The terminal sends the entered query content to the server. The server receives this via the API endpoint and begins parsing the query.

[0202] The server parses the query.

[0203] The server analyzes the query content using a natural language processing (NLP) model. For example, the NLP model extracts key keywords and context from the text to understand the intent of the query. Techniques such as morphological analysis and syntax analysis are used in this step. Libraries such as spaCy are examples of NLP models that can be used.

[0204] The server searches for relevant answers.

[0205] The server searches the database for relevant answers based on the analyzed query. The database contains pre-configured frequently asked questions (FAQs) and store manuals, and the search engine returns the most suitable answer based on this information. SQLite and MySQL (registered trademarks) are used as database management systems (DBMS).

[0206] The server sends the response to the terminal.

[0207] The server sends the search results to the device. The device displays this to the user as a chat window or notification. This allows the user to quickly obtain the appropriate information.

[0208] The server stores queries and answers in a database.

[0209] The server stores user inquiries and their corresponding responses in a database. This data is used to improve the accuracy of automated responses to future inquiries.

[0210] The server trains the machine learning model.

[0211] The server trains a machine learning model using the accumulated data. This model improves the response generation algorithm to handle new queries, resulting in more accurate automated responses. Machine learning libraries such as TensorFlow and PyTorch are used.

[0212] The conveying equipment responds to the customer.

[0213] A transport device (e.g., a robot) can guide customers within a store based on the analyzed inquiry. For example, if a customer asks, "Where is this product?", the device can guide the customer to the location where the product is located based on the analyzed results.

[0214] Specific example

[0215] For example, if a user submits a request such as "Please tell me which items are on sale," the server analyzes the request and extracts relevant keywords like "sale" and "item." The server then searches its database for an answer such as "The current sale items are A, B, and C" and sends it to the user's terminal. This allows the user to quickly obtain the information they need.

[0216] Example of a prompt

[0217] "Please analyze the following user inquiry and provide an appropriate answer: 'What items are on sale?'"

[0218] The above describes the embodiments for carrying out the present invention. This system enables efficient customer service in physical stores and improves customer satisfaction.

[0219] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0220] Program processing flow

[0221] Step 1:

[0222] The user enters a language-based inquiry using a terminal. The entered inquiry is captured as text data by the user's terminal's input interface. For example, the user enters the inquiry, "Please tell me what items are on sale."

[0223] Step 2:

[0224] The terminal sends the entered query content to the server. Here, the query data is sent to the server's API endpoint via an HTTP POST request. The input data is the text of the query.

[0225] Step 3:

[0226] The server analyzes the received query using a natural language processing (NLP) model. Specifically, it uses an NLP model (e.g., spaCy) to analyze the text data and extract the main keywords "sale" and "product." In this step, the text of the input data is converted into the main keywords.

[0227] Step 4:

[0228] The server searches for relevant answers from a pre-configured database based on the parsed query. For example, it might use an SQLite database to search for information on "sale items." In this step, it uses the entered keywords to execute a database query that contains the appropriate answers.

[0229] Step 5:

[0230] The server sends the search results to the terminal. Specifically, it sends the answers back to the terminal as an HTTP response. In this step, the answers retrieved from the database are sent to the terminal as output data.

[0231] Step 6:

[0232] The terminal displays the received response to the user. The response content is displayed on the user's terminal screen. For example, it might say, "Current sale items are A, B, and C."

[0233] Step 7:

[0234] The server stores user inquiries and their corresponding answers in a database. This improves the accuracy of automated responses to future inquiries. In this step, the input data (inquiry content) and output data (answer) are stored in the database as a single record.

[0235] Step 8:

[0236] The server trains machine learning models using the accumulated data. Specifically, it uses libraries such as TensorFlow and PyTorch to train the models and improve their accuracy. In this step, the historical data stored in the database is used as input data for the machine learning models, and the model parameters are updated.

[0237] Step 9:

[0238] The transport system responds to the customer based on the analyzed inquiry. For example, a robot guides the customer to the location where the product is located. In this step, the transport system physically operates based on the input data (analyzed inquiry).

[0239] Through the processing steps described above, this system can streamline customer service in physical stores and provide timely and appropriate information. Furthermore, machine learning using accumulated data continuously improves the system's response accuracy.

[0240] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0241] This invention is a system that efficiently receives user inquiries, provides appropriate answers, and optimizes responses by recognizing user emotions. This achieves improved operational efficiency and enhanced user experience. Specific embodiments for implementing this invention are described below.

[0242] System Configuration

[0243] The system consists of a user terminal, a server, a database, and an emotion engine. Users make queries using the terminal, the server receives and analyzes them, searches for relevant answers, and provides them to the user. The emotion engine also recognizes the user's emotions and provides appropriate responses accordingly.

[0244] The user enters an inquiry.

[0245] Users use their devices (e.g., PCs or smartphones) to input language-based inquiries into the system. This allows them to ask questions about product and service details, account setup instructions, and more.

[0246] The device sends the query to the server.

[0247] The terminal sends the entered query content to the server. The server receives this via the API endpoint and begins parsing the query.

[0248] The server parses the query.

[0249] The server uses a natural language processing (NLP) model to analyze the query. Specifically, it uses morphological and syntax analysis to extract key keywords and context to understand the intent of the query.

[0250] The server uses an emotion engine to recognize emotions.

[0251] The server uses an emotion engine to recognize emotions from the user's inquiry. For example, it uses text analysis to determine whether the user is angry, anxious, or calm.

[0252] The server searches the FAQ database.

[0253] The server searches the database for relevant answers based on the analyzed query. The database contains pre-configured FAQs and manuals, and the search engine uses this information to return the most suitable answer.

[0254] The server generates responses based on emotions.

[0255] The server adjusts its responses and tone to be more user-friendly based on the analysis results of the emotion engine. For example, if the user is angry, it will generate a response that includes a polite apology.

[0256] The server sends the response to the terminal.

[0257] The server sends the generated response to the device. The device displays this to the user as a chat window or notification. This allows the user to quickly obtain the appropriate information.

[0258] The server stores queries and answers in a database.

[0259] The server stores user inquiries and their corresponding responses in a database. This data is used to improve the accuracy of automated responses to future inquiries.

[0260] The server trains the machine learning model.

[0261] The server uses accumulated data to train a machine learning model. This model improves the response generation algorithm to handle new queries, resulting in more accurate automated responses.

[0262] Automated operation

[0263] Furthermore, if a user requests specific actions such as sending a replacement device, signing a new contract, or changing models, the server will call the appropriate API to automate these operations. For example, if a user requests "Please send a replacement device," the server will integrate with the customer management system and automatically handle the process of sending the replacement device.

[0264] Specific example

[0265] For example, if a user submits an inquiry such as "Please tell me the conditions for maintenance coverage," the server analyzes the inquiry and extracts relevant keywords such as "maintenance," "coverage," and "conditions." The server then uses an emotion engine to analyze the user's emotions and recognizes that the user is feeling anxious. Next, the server searches its database for an appropriate answer, such as "Maintenance services are only applicable within the product warranty period," and generates a response that includes a more considerate explanation that takes the user's feelings into account. This response is then sent to the terminal and displayed to the user. This allows the user to obtain the necessary information with peace of mind.

[0266] In this way, the system of the present invention efficiently processes user inquiries, automates appropriate responses and operations, and simultaneously improves operational efficiency and user experience by recognizing user emotions and optimizing responses.

[0267] The following describes the processing flow.

[0268] Step 1:

[0269] The user enters the inquiry. The user uses a device (for example, a PC or smartphone) to enter a question in text format, such as "Please tell me the conditions for applying maintenance."

[0270] Step 2:

[0271] The terminal sends the query to the server. The terminal sends the entered query text to the server via the API.

[0272] Step 3:

[0273] The server receives the query. The server receives the query text from the API endpoint and prepares to parse it.

[0274] Step 4:

[0275] The server analyzes the query using a natural language processing (NLP) model. The server performs morphological analysis and extracts major keywords (e.g., maintenance, application, conditions).

[0276] Step 5:

[0277] The server uses an emotion engine to recognize the user's emotion. The server analyzes the query text using pattern matching or an emotion dictionary and determines that the user may be feeling, for example, "uneasy".

[0278] Step 6:

[0279] The server searches the FAQ database. The server searches the FAQ database for relevant answers based on the extracted keywords.

[0280] Step 7:

[0281] The server selects the optimal answer. The server selects the most appropriate one from the multiple found answers.

[0282] Step 8:

[0283] The server adjusts the answer based on the emotion recognition result. Based on the result of the emotion engine, the server adjusts the answer content by, for example, adding a gentle tone or polite words.

[0284] Step 9:

[0285] The server generates an answer. The server generates an optimal answer sentence and prepares to send it to the terminal in text format.

[0286] Step 10:

[0287] The server sends the answer to the terminal. The server sends the generated answer text to the terminal via an API.

[0288] Step 11:

[0289] The device displays the response to the user. The device displays the received response to the user in a chat window or notification format. For example, it may display a polite explanation such as, "Maintenance service is only applicable within the product warranty period. Please contact us anytime if you have any concerns."

[0290] Step 12:

[0291] The server stores inquiries and answers in a database. The server records the details of the current inquiry and the corresponding answer in the database, and uses this information to handle future inquiries.

[0292] Step 13:

[0293] The server trains a machine learning model. The server periodically trains the machine learning model using accumulated data to improve the accuracy of automated responses.

[0294] Step 14:

[0295] The user requests a specific action. For example, the user might request a specific action through the chat window, such as "Please send a replacement device."

[0296] Step 15:

[0297] The terminal sends an operation request to the server. The terminal sends an operation request to the server via an API.

[0298] Step 16:

[0299] The server processes the request. The server calls the appropriate API (e.g., customer management system) and automatically performs the requested operation (e.g., arranging the shipment of a replacement unit).

[0300] Step 17:

[0301] The server sends the processing result to the terminal. It sends a notice indicating that the processing has been completed to the terminal.

[0302] Step 18:

[0303] The terminal displays the processing result to the user. The terminal displays the processing result to the user in a chat window or in a notification format. For example, it displays "The replacement device has been shipped. Please wait a while until it arrives."

[0304] Through the above steps, the system efficiently processes inquiries from users, automates appropriate responses and operations, and in addition, recognizes the emotions of users and optimizes responses, thereby simultaneously achieving an improvement in business efficiency and an improvement in the user experience.

[0305] (Example 2)

[0306] Next, Example 2 will be described. In the following description, the data processing device 12 is referred to as the "server", and the smart device 14 is referred to as the "terminal".

[0307] In the conventional inquiry response system, it often provides a uniform response without considering the emotions of users, and thus cannot sufficiently improve the user satisfaction. Also, when the quality and quantity of the training data of the machine learning model are insufficient, there is a problem that the accuracy of the automatic response to new inquiries decreases. Furthermore, since there is a lack of means to automate appropriate processing for specific operation requests from users, the efficiency of the business has not been sufficiently improved.

[0308] The specific processing by the specific processing unit 290 of the data processing device 12 in Example 2 is realized by the following respective means.

[0309] In this invention, the server includes means for receiving inquiries from users, means for analyzing inquiries using a natural language processing model, means for searching for relevant answers from a pre-configured database based on the analyzed inquiries, means for providing the retrieved answers to the user, means for storing inquiries and answers in a database, means for training a machine learning model using the data stored in the database to improve responses to new inquiries, means for using an emotion analysis engine that recognizes emotions from the content of the user's inquiry, and means for adjusting the tone and content of the response based on the emotion analysis results. This enables appropriate responses that take into account the user's emotions, thereby increasing satisfaction, improving the accuracy of the machine learning model, and enabling quick and accurate responses to new inquiries. Furthermore, automated processing of specific operation requests can be realized, thus improving operational efficiency.

[0310] A "user" is an individual or organization that uses the system to make an inquiry.

[0311] An "inquiry" refers to a question or request that a user makes to a system.

[0312] A "terminal" refers to an electronic device (such as a PC or smartphone) that a user uses to enter an inquiry.

[0313] A "server" is a computer system that receives user inquiries, analyzes them, and provides appropriate answers.

[0314] A "natural language processing model" is an algorithm or software used to analyze, understand, and generate responses to human language.

[0315] "Analysis" is the process of examining the content of an inquiry and extracting its intent and meaning.

[0316] A "database" is a system for systematically storing inquiries, their answers, and other related information.

[0317] A "sentiment analysis engine" is software or an algorithm that recognizes emotions from user input text and provides the results.

[0318] A "machine learning model" is a collection of algorithms that learn from large amounts of data, extract patterns and knowledge, and then make predictions and classifications based on new data.

[0319] "Searching" is the process of finding specific information from a database.

[0320] "Tone" refers to the way responses are expressed and the writing style, which is adjusted to match the user's emotions.

[0321] An "API" is an interface that allows different software systems to communicate with each other.

[0322] "Automated processing" refers to operations or procedures that a system performs automatically with minimal human intervention.

[0323] This invention provides a system that efficiently receives user inquiries and provides appropriate responses, and further optimizes responses by recognizing user emotions. This achieves improved operational efficiency and enhanced user experience.

[0324] System Configuration

[0325] The system consists of user terminals, servers, a database, and an emotion analysis engine.

[0326] Hardware and software used

[0327] User terminal: A device such as a PC or smartphone used by the user to enter their inquiry.

[0328] Server: A server runs multiple software programs. Details are provided below.

[0329] Natural language processing models: "spaCy" and "NLTK" are used for morphological and syntax analysis.

[0330] Sentiment Analysis Engine: To analyze user emotions, we use "IBM Watson® Tone Analyzer".

[0331] FAQ database search engine: Use "Elasticsearch" for searching.

[0332] Machine learning models: TensorFlow and PyTorch are used for data training and prediction.

[0333] Program processing flow

[0334] The user enters an inquiry.

[0335] Users enter inquiries into the system using their own devices (e.g., PCs or smartphones). For example, a user might enter a question such as, "Please tell me the conditions for applying maintenance."

[0336] The device sends the query to the server.

[0337] The terminal sends the entered query content to the server using an HTTP POST request. The request contains the query content and is sent to the specified API endpoint (e.g., / api / v1 / inquiries).

[0338] The server parses the query.

[0339] The server passes the received request to a natural language processing (NLP) model for analysis. For example, it uses the Python library "spaCy" to perform morphological and syntax analysis, extracting key keywords and context. To understand the intent of the query, the keywords are categorized into specific groups.

[0340] The server recognizes emotions using an emotion engine.

[0341] The server passes the analysis results to a sentiment analysis engine (e.g., IBM Watson Tone Analyzer) to recognize the user's emotions. For example, if an inquiry frequently uses language that includes anxiety or anger, the server will identify those emotions.

[0342] The server searches the database.

[0343] The server searches the FAQ database based on the analysis results. It uses Elasticsearch to quickly find answers related to the inquiry. For example, if a user inquires about maintenance services, the server retrieves information from the database such as "Maintenance services are only applicable within the product warranty period."

[0344] The server generates responses based on emotions.

[0345] The server uses a generative AI model (e.g., GPT-3®) based on the sentiment analysis results to generate responses in an appropriate tone for the user. For example, if the user is feeling anxious, it might generate a response such as, "To alleviate your concerns, please rest assured that maintenance services are only applicable within the product warranty period."

[0346] The server sends the response to the terminal.

[0347] The server sends the generated response to the device and displays it to the user. The response is displayed on the user's device in a chat window or as a notification.

[0348] The server stores queries and answers in a database.

[0349] The server stores user inquiries and generated responses in a database. This information can be reused for future inquiries. MySQL and PostgreSQL are commonly used databases.

[0350] The server trains the machine learning model.

[0351] The server uses the accumulated data to train machine learning models. For example, it uses tools like TensorFlow or PyTorch to improve the model's performance and enable it to respond quickly and accurately to new queries.

[0352] Automated operation

[0353] The server automatically responds to specific user requests (e.g., shipping a replacement device, signing a new contract, changing models, etc.). For example, if a user requests "ship a replacement device," the server will integrate with a customer management system (e.g., Salesforce) and automatically initiate the process of shipping the replacement device.

[0354] Specific example

[0355] For example, if a user submits an inquiry such as "Please tell me the conditions for maintenance coverage," the server analyzes the inquiry and extracts relevant keywords such as "maintenance," "coverage," and "conditions." The server then uses a sentiment analysis engine to analyze the user's emotions and recognizes that the user is feeling anxious. Next, the server searches its database for an appropriate answer, such as "Maintenance services are only applicable within the product warranty period," and generates a response that includes a more considerate and detailed explanation that takes the user's feelings into account. This response is then sent to the terminal and displayed to the user. This allows the user to obtain the necessary information with peace of mind.

[0356] In this way, the system of the present invention efficiently processes user inquiries, automates appropriate responses and operations, and simultaneously improves operational efficiency and user experience by recognizing user emotions and optimizing responses.

[0357] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0358] Step 1:

[0359] The user enters an inquiry.

[0360] Users enter inquiries into the system using their own devices (PCs, smartphones, etc.). Specifically, users open a web browser or application and enter questions or requests in text format into input fields. For example, a user might enter the prompt message, "Please tell me the conditions for maintenance application."

[0361] Step 2:

[0362] The device sends the query to the server.

[0363] The terminal sends the entered query content to the server as an HTTP POST request. The input data (user's query text) is included in the request body. The request is sent to the specified API endpoint (e.g., / api / v1 / inquiries). The server receives this request and prepares to process the query content.

[0364] Step 3:

[0365] The server parses the query.

[0366] The server passes the received query content to a natural language processing (NLP) model for analysis. The query text is provided as input, and morphological and syntax analysis is performed. Specifically, the Python library "spaCy" is used to extract key keywords and context. The output provides the analyzed keywords and contextual information.

[0367] Step 4:

[0368] The server recognizes emotions using an emotion engine.

[0369] The server passes the parsed query to the sentiment analysis engine to recognize the user's emotions. The parsed query text is used as input, and the sentiment analysis engine (e.g., IBM Watson Tone Analyzer) analyzes the text. The output is data indicating the user's emotions (anger, anxiety, calmness, etc.).

[0370] Step 5:

[0371] The server searches the FAQ database.

[0372] The server searches the FAQ database for relevant answers based on the analysis results. The query keywords and contextual information are used as input, and a search engine (e.g., Elasticsearch) extracts the appropriate answers. The output is the relevant answers retrieved from the database.

[0373] Step 6:

[0374] The server generates responses based on emotions.

[0375] The server generates responses to the user in an appropriate tone based on the searched answers and sentiment analysis results. The inputs used are answers and sentiment data from a database. A generative AI model (e.g., GPT-3) generates text in an appropriate tone based on this input data. The output is a sentiment-sensitive response text. For example, a response such as "Please rest assured that maintenance services are only applicable within the product warranty period" might be generated.

[0376] Step 7:

[0377] The server sends the response to the terminal.

[0378] The server sends the generated response to the device. The generated response text is used as input and returned to the device as an HTTP response. The device receives this response and displays the response to the user. Specifically, it is displayed as a chat window on a web page or as a notification in an application.

[0379] Step 8:

[0380] The server stores queries and answers in a database.

[0381] The server stores user inquiries and generated responses in a database. Inquiry text and response text are used as input and stored in a database (e.g., MySQL or PostgreSQL). This data is accumulated in the database as output. This accumulated data helps improve future query responses.

[0382] Step 9:

[0383] The server trains the machine learning model.

[0384] The server uses accumulated data to train a machine learning model. Past query and response data is used as input, and the model is trained using a machine learning framework (e.g., TensorFlow or PyTorch). The output is a trained model capable of generating highly accurate responses.

[0385] Step 10:

[0386] The server automates the operation.

[0387] The server automatically responds to specific user requests (e.g., shipping a replacement device, signing a new contract, changing models, etc.). The user's request text is used as input, and the server calls the appropriate API to process it. For example, if a user enters "Please ship a replacement device," the server will integrate with the customer management system and automatically initiate the shipping process. The user receives a notification that the requested operation has been completed.

[0388] (Application Example 2)

[0389] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0390] Conventional inquiry handling systems struggle to respond to user inquiries quickly and appropriately, particularly lacking in responses that take emotions into consideration. Therefore, there is a need for improved user experience and operational efficiency. Furthermore, in security services, emotional recognition is especially crucial in handling emergency inquiries, requiring quick and appropriate responses. This often leads to challenges such as a lack of user confidence and delays in responses.

[0391] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0392] In this invention, the server includes means for recognizing emotions from the content of a user's inquiry, means for adjusting the tone of the response and generating an emotion-appropriate answer based on the recognized emotion, and means for training a machine learning model using data stored in the database to improve responses to new inquiries. This enables quick and appropriate responses that take into account the user's emotions, improving the user experience and operational efficiency.

[0393] "Means for receiving user inquiries" refers to a function that provides an interface for end users to make inquiries to the system via text or voice.

[0394] "Methods for analysis using natural language processing models" refer to functions that convert linguistic information contained in user inquiries into a format that a computer can understand, and then apply a model to analyze the intent and content of the inquiry.

[0395] "Means for searching for relevant answers" refers to a function that finds appropriate answers from a database based on the analyzed query content.

[0396] "Means of providing information to users" refers to functions that display search results in a user-friendly format and provide information to users.

[0397] "Means of recognizing emotions" refers to a function that analyzes the user's inquiry to determine their emotional state, such as anger, anxiety, or calmness.

[0398] "Means for generating emotionally responsive answers" refers to a function that creates more appropriate and friendly responses based on the recognized emotions of the user.

[0399] "Means for storing inquiries and answers in a database" refers to a function that records received inquiries and their responses in a database, which can then be used to handle future inquiries.

[0400] "Methods for training machine learning models" refers to the function of training models that improve the accuracy of automated responses using machine learning algorithms based on a large amount of query and answer data accumulated in a database.

[0401] This invention is a system that efficiently receives user inquiries, provides appropriate answers, and further recognizes user emotions to optimize responses. The system consists of a user terminal, a server, a database, and an emotion engine. Detailed embodiments of this system are described below.

[0402] System Configuration

[0403] Users input language-based queries to the system using devices such as PCs and smartphones. The device sends the entered query content to the server. The server receives this via an API endpoint and parses the query.

[0404] Hardware and software to be used

[0405] Hardware:

[0406] Smartphone or PC (device)

[0407] server

[0408] software:

[0409] Natural Language Processing (NLP) Model: Using the Python transformers library

[0410] Emotion recognition model: Uses the transformers library from Hugging Face.

[0411] Databases: Relational databases such as MySQL and PostgreSQL

[0412] Machine learning models: TensorFlow or PyTorch

[0413] Data processing and data calculation

[0414] Inquiry Analysis

[0415] The server analyzes user inquiries using a natural language processing model. Specifically, it performs morphological and syntax analysis to extract key keywords and their context.

[0416] emotion recognition

[0417] After analyzing the content of the inquiry, an emotion engine is used to recognize the user's emotions. Through text analysis, it is determined whether the user is angry, anxious, or calm.

[0418] Generating an answer

[0419] The server searches for relevant answers based on the analyzed query and the perceived emotion, and adjusts the tone of the response accordingly. For example, if the user is angry, it might include a polite apology.

[0420] Data storage and training of machine learning models

[0421] All inquiries and answers are stored in a database. This data is used to train machine learning models, improving the accuracy of responses to new inquiries.

[0422] Specific example

[0423] Specific inquiry scenarios and responses

[0424] User inquiry: "I heard gunshots nearby."

[0425] System response: "Please stay calm. Close windows and doors and move to a safe place immediately."

[0426] Examples of prompts for a generative AI model:

[0427] "When a user reports hearing gunshots nearby, analyze their emotions and generate a quick response to ensure their safety."

[0428] In this way, the system of the invention can efficiently and quickly process user inquiries and generate appropriate responses. This system optimizes responses to urgent user inquiries, particularly in the field of security services, and provides users with peace of mind.

[0429] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0430] Step 1:

[0431] The user enters their inquiry.

[0432] The user enters their inquiry in text format using a terminal. After input, the inquiry data is sent from the terminal to the server.

[0433] Input: Text data entered by the user on the device.

[0434] Output: Query data sent to the server

[0435] Step 2:

[0436] The server receives the query.

[0437] The server receives query data sent from the terminal via the API endpoint.

[0438] Input: Inquiry data sent from the terminal

[0439] Output: Text data received by the server

[0440] Step 3:

[0441] The server analyzes the query using a natural language processing model.

[0442] The server uses an NLP model (transformers library) to perform morphological and syntax analysis on the query content. It extracts key keywords and context to understand the intent of the query.

[0443] Input: Text data received by the server

[0444] Output: Intent and key keywords of the query

[0445] Step 4:

[0446] The server uses an emotion recognition model to recognize emotions.

[0447] The server uses an emotion recognition model (transformers library) to determine the user's emotions from the analyzed query content. For example, it can determine whether the user is angry, anxious, or calm.

[0448] Input: Intent and main keywords of the inquiry

[0449] Output: User's emotional state

[0450] Step 5:

[0451] The server searches the FAQ database.

[0452] The server searches the database for relevant answers based on the analyzed query content and sentiment state. If a relevant answer is found, it retrieves it.

[0453] Input: Inquiry intent, key keywords, user sentiment

[0454] Output: Related answers

[0455] Step 6:

[0456] The server generates responses based on emotions.

[0457] Based on the emotion recognition results, the server adjusts the tone to be user-friendly and generates an appropriate response. For example, if the user is angry, it will generate a response that includes a polite apology.

[0458] Input: Related answers, user's emotional state

[0459] Output: Emotionally adjusted response

[0460] Step 7:

[0461] The server sends the response to the terminal.

[0462] The server sends the generated response to the device. The device then displays this to the user as a chat window or notification.

[0463] Input: Emotionally adjusted response

[0464] Output: The answer displayed on the user's device.

[0465] Step 8:

[0466] The server stores inquiries and answers in a database.

[0467] The server records the received inquiry and the generated response in a database. This data will be used to improve the accuracy of future inquiry handling.

[0468] Input: Inquiry details, generated response

[0469] Output: Queries and answers stored in the database

[0470] Step 9:

[0471] The server trains the machine learning model.

[0472] The server uses the accumulated data to train a machine learning model. This model improves the response generation algorithm to handle new queries.

[0473] Input: Inquiries and answers stored in the database

[0474] Output: Trained machine learning model

[0475] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0476] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0477] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0478] [Second Embodiment]

[0479] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0480] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0481] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0482] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0483] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0484] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0485] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0486] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0487] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0488] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0489] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0490] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0491] This invention is a system that efficiently receives and analyzes user inquiries and provides appropriate responses. This achieves improved operational efficiency and cost reduction. Specific embodiments for implementing this invention are described below.

[0492] System Configuration

[0493] The system consists of user terminals, a server, and a database. Users make queries using the terminals, the server receives and analyzes them, searches for relevant answers, and provides them to the user.

[0494] The user enters an inquiry.

[0495] Users use their devices (e.g., PCs or smartphones) to input language-based inquiries into the system. This allows them to ask questions about seasonal products and services, how to set up an account, and more.

[0496] The device sends the query to the server.

[0497] The terminal sends the entered query content to the server. The server receives this via the API endpoint and begins parsing the query.

[0498] The server parses the query.

[0499] The server analyzes the query using a natural language processing (NLP) model. For example, the NLP model extracts key keywords and context from the text to understand the intent of the query. Techniques such as morphological analysis and syntax analysis are used in this step.

[0500] The server searches for relevant answers.

[0501] The server searches the database for relevant answers based on the analyzed query. The database contains pre-configured FAQs and manuals, and the search engine uses this information to return the most suitable answer.

[0502] The server sends the response to the terminal.

[0503] The server sends the search results to the device. The device displays this to the user as a chat window or notification. This allows the user to quickly obtain the appropriate information.

[0504] The server stores queries and answers in a database.

[0505] The server stores user inquiries and their corresponding responses in a database. This data is used to improve the accuracy of automated responses to future inquiries.

[0506] The server trains the machine learning model.

[0507] The server uses accumulated data to train a machine learning model. This model improves the response generation algorithm to handle new queries, resulting in more accurate automated responses.

[0508] Specific example

[0509] For example, if a user submits a query such as "Please tell me the conditions for applying maintenance," the server analyzes the query and extracts the relevant keywords "maintenance," "application," and "conditions." The server then searches its database for the appropriate answer, such as "Maintenance services are only applicable within the product warranty period," and sends it to the user's terminal. This allows the user to quickly obtain the necessary information.

[0510] Automated operation

[0511] Furthermore, if a user requests specific actions such as sending a replacement device, signing a new contract, or changing models, the server will call the appropriate API to automate these operations. For example, if a user requests "Please send a replacement device," the server will integrate with the customer management system and automatically handle the process of sending the replacement device.

[0512] In this way, the system of the present invention efficiently processes user inquiries and automates appropriate responses and operations, thereby achieving improved operational efficiency and cost reduction.

[0513] The following describes the processing flow.

[0514] Step 1:

[0515] The user enters the inquiry. The user uses a device (e.g., a PC or smartphone) to input questions or requests to the system in text format.

[0516] Step 2:

[0517] The device sends the query to the server. The device sends the entered query content to the server via the API endpoint.

[0518] Step 3:

[0519] The server receives the query. The server retrieves the text data received from the API endpoint.

[0520] Step 4:

[0521] The server analyzes the query. The server uses a natural language processing (NLP) model to analyze the query content. Specifically, it uses morphological analysis and syntax analysis to extract key keywords and context.

[0522] Step 5:

[0523] The server extracts keywords. It extracts important keywords, mainly nouns and verbs, from the query text.

[0524] Step 6:

[0525] The server searches the FAQ database. Based on the extracted keywords, the server searches the FAQ database for relevant answers.

[0526] Step 7:

[0527] The server selects the most appropriate answer. If multiple answer choices exist, the server selects the most relevant answer.

[0528] Step 8:

[0529] The server generates the answer. Based on the selected answer, it generates the final text to be provided to the user.

[0530] Step 9:

[0531] The server sends the response to the device. An API endpoint is used to send the generated response text to the device.

[0532] Step 10:

[0533] The device displays the response to the user. The device displays the received response to the user in a chat window or notification format.

[0534] Step 11:

[0535] The server stores queries and answers in a database. The query content and the corresponding answers are recorded in the database and used for future query processing.

[0536] Step 12:

[0537] The server trains a machine learning model. Using accumulated data, the machine learning model is periodically trained to improve the accuracy of automated responses to inquiries.

[0538] Step 13:

[0539] The user requests a specific action. Specific requests, such as sending a replacement device, signing a new contract, or changing models, are entered in text format.

[0540] Step 14:

[0541] The server processes the user's request. It calls the appropriate API and performs the specific action required for the request (e.g., arranging for a replacement device to be shipped).

[0542] Step 15:

[0543] The server sends the processing results to the terminal. A notification indicating that processing is complete is sent to the terminal.

[0544] Step 16:

[0545] The terminal displays the processing results to the user. The terminal displays the processing results to the user in a chat window or notification format.

[0546] Through the steps described above, the system can efficiently process user inquiries and automate appropriate responses and actions.

[0547] (Example 1)

[0548] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0549] Conventional inquiry handling systems suffered from long response times to user inquiries and low accuracy in providing appropriate answers. Furthermore, the need for numerous manual operations to find appropriate answers prevented sufficient efficiency improvements and cost reductions. The objective of this invention is to solve these problems and provide users with efficient and highly accurate responses.

[0550] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0551] In this invention, the server includes means for receiving inquiries from users, means for analyzing the inquiries using a natural language processing model, means for sending the input inquiries to the server via an API endpoint, means for searching for relevant answers from a pre-configured data storage based on the analyzed inquiries, means for providing the retrieved answers to the user, means for storing the inquiries and answers in the data storage, and means for training a computational model using the data stored in the data storage to improve the response to new inquiries. This enables a rapid and accurate response to user inquiries, efficient business operations, and cost reduction.

[0552] An "inquiry" is the act of a user inputting information they want to know or a problem they want to solve into a system using natural language.

[0553] A "natural language processing model" is a type of artificial intelligence that analyzes text written in natural language, understands its content and meaning, and processes it.

[0554] An "API endpoint" is part of an application programming interface and is a connection point for different software systems to communicate with each other.

[0555] A "server" is a computer system that provides services to other computers via a network.

[0556] "Data storage" refers to devices and systems for storing digital data, and includes databases and cloud storage.

[0557] "Analysis" is the act of breaking down input data into its components, understanding its content, and identifying its meaning and intent.

[0558] An "answer" refers to the information or solution provided in response to a user's inquiry.

[0559] A "machine learning model" is an algorithm that learns from data and performs tasks such as prediction and classification.

[0560] A "computational model" is a mathematical model used to perform a specific task through numerical calculations and data processing.

[0561] A "programming interface" is a set of definitions and protocols that enable different software programs to work together smoothly.

[0562] "Accumulation" is the act of collecting data and saving it.

[0563] This invention is a system that efficiently receives and analyzes user inquiries and provides appropriate responses. This achieves improved operational efficiency and cost reduction. Specific embodiments of this invention are described below.

[0564] System Configuration

[0565] This system consists of user terminals, a server, and data storage. Users use the terminals to make inquiries, the server receives and analyzes them, searches for relevant answers, and provides them to the user.

[0566] The user enters an inquiry.

[0567] Users use their devices (such as PCs or smartphones) to input natural language queries into the system. For example, they might ask, "I'd like to know about the product's warranty period."

[0568] The device sends the query to the server.

[0569] The terminal sends the entered query content as an HTTP POST request to the server's API endpoint. The server receives this and begins parsing the query.

[0570] The server parses the query.

[0571] The server uses an NLP (Natural Language Processing) model to analyze the query. This analysis utilizes NLP libraries such as spaCy and Hugging Face in Python. For example, morphological analysis is used to extract the keyword "product warranty period."

[0572] The server searches for relevant answers.

[0573] The server searches for relevant answers in data storage based on the extracted keywords. Data storage may include SQL Server or Elasticsearch. The server finds the best answer, retrieving information such as "The product warranty period is one year."

[0574] The server sends the response to the terminal.

[0575] The server sends the retrieved response to the device in JSON format. The device then displays this response to the user in a chat window or notification, allowing the user to quickly obtain the necessary information.

[0576] The server stores queries and answers in data storage.

[0577] The server records the query content and the corresponding response in a separate data storage. This data will be used to improve future query responses.

[0578] The server trains the machine learning model.

[0579] The server uses data stored in data storage to train machine learning models. For example, it uses Python's Scikit-learn or TensorFlow to improve the accuracy of the models. This enables sophisticated automated responses to new queries.

[0580] Automated operation

[0581] Furthermore, this system can handle specific user requests such as shipping replacement equipment, signing up for new contracts, or changing models. For example, if a user requests "Please ship a replacement unit," the server will call the appropriate programming interface and automatically execute the operation. This programming interface integrates with customer management systems, logistics systems, and other systems to automate necessary procedures.

[0582] Specific example

[0583] For example, if a user asks, "What are the conditions for maintenance?", the server analyzes the inquiry and extracts the keywords "maintenance," "applicability," and "conditions." The server then searches its database for the answer, "Maintenance services are applicable only within the product warranty period," and sends it to the user's terminal. This allows the user to quickly obtain the necessary information.

[0584] Thus, the present invention enables prompt and accurate responses to user inquiries, thereby improving operational efficiency and reducing costs.

[0585] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0586] Step 1:

[0587] The user enters an inquiry.

[0588] Users enter their inquiries in natural language into a text form using a device such as a PC or smartphone. Specifically, they might type "Please tell me about the product's warranty period" and click the submit button. The entered text is temporarily stored on the user's device.

[0589] Step 2:

[0590] The device sends the query to the server.

[0591] When the user clicks the submit button, the device sends the entered query content to the server as an HTTP POST request. For example, it sends a request with data in JSON format like the following:

[0592] json

[0593] {

[0594] "query": "Please tell me about the product's warranty period."

[0595] }

[0596] The server receives this request and temporarily holds the query details in memory.

[0597] Step 3:

[0598] The server parses the query.

[0599] The server inputs the received query into a natural language processing (NLP) model and begins analysis. Specifically, it uses a Python NLP library (e.g., spaCy) to perform morphological analysis and extract keywords and phrases from the text. For example, the keyword "product warranty period" might be extracted. The input is the received query, and the output is the extracted keywords.

[0600] Step 4:

[0601] The server searches for relevant answers.

[0602] The server searches for information in data storage based on keywords extracted from the analyzed query. It uses SQL Server, Elasticsearch, etc., to retrieve relevant answers. Specifically, it executes SQL queries like the following:

[0603] sql

[0604] SELECT answer FROM faq WHERE question LIKE '%warranty period%'

[0605] The input is the extracted keywords, and the output is the appropriate answer. This query yields the answer, "The product warranty period is one year."

[0606] Step 5:

[0607] The server sends the response to the terminal.

[0608] The server sends the retrieved response to the terminal as a JSON response. Specifically, it sends it in the following format:

[0609] json

[0610] {

[0611] "Answer": "The product warranty period is one year."

[0612] }

[0613] The terminal receives this data and displays it in an inquiry form or a dedicated window. The output is displayed on the user's terminal.

[0614] Step 6:

[0615] The user receives the response.

[0616] The user checks the answer displayed on their device. Specifically, they can see a message such as "The product warranty period is one year" and get the appropriate answer to their inquiry.

[0617] Step 7:

[0618] The server stores queries and answers in a database.

[0619] The server stores the query details and corresponding answers in a database. This data is used to improve future responses. Specifically, the data is stored as follows:

[0620] sql

[0621] INSERT INTO logs (query, answer) VALUES ('Please tell me about the product's warranty period', 'The product's warranty period is 1 year')

[0622] The input consists of the inquiry content and the response, and the output is the storage of the data in the database.

[0623] Step 8:

[0624] The server trains the machine learning model.

[0625] The server trains a machine learning model using historical query data stored in the database. Specifically, it uses Python's Scikit-learn and TensorFlow to improve the model's accuracy. This model is designed to increase the accuracy of generating answers to new queries. The input is historical query data, and the output is the updated machine learning model.

[0626] (Application Example 1)

[0627] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0628] In modern brick-and-mortar stores, it is necessary to respond quickly and accurately to a wide range of customer inquiries, but relying on manual responses makes efficient and cost-effective operation difficult. Furthermore, if customers cannot obtain information about products and services immediately, their willingness to purchase may decrease. In addition, manual processing is time-consuming when performing specific actions based on user inquiries, leading to decreased customer satisfaction. This invention aims to solve these problems and achieve efficiency and automation in customer service in physical stores.

[0629] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0630] In this invention, the server includes means for receiving inquiries from users, means for analyzing inquiries using a natural language processing model, and means for searching for relevant answers from a pre-configured database based on the analyzed inquiries. This enables the rapid provision of answers to customer inquiries. Furthermore, by including means for customer interaction in the transport device, it becomes possible to automate tasks such as guiding customers to product locations and assisting with purchase procedures. This streamlines customer service in physical stores and improves customer satisfaction.

[0631] definition statement

[0632] "Means of receiving user inquiries" refers to the function of receiving any inquiries that users input into the system via their terminals.

[0633] "Methods for analysis using natural language processing models" refers to a function that analyzes input queries using natural language processing technology, extracts key keywords and context, and understands the user's intent.

[0634] "Means of searching for relevant answers from a pre-configured database" refers to a function that searches for the most suitable answer from a pre-configured database based on the analyzed query content.

[0635] "Means of providing information to the user" refers to functions that notify or display the searched answers on the user's device.

[0636] "Means for storing inquiries and answers in a database" refers to a function that saves user inquiries and the corresponding answers in a database.

[0637] "Means of training machine learning models to improve responses to new inquiries" refers to a function that uses accumulated data to train machine learning models and improve the accuracy and effectiveness of responses to future inquiries.

[0638] "Means by which a transport device responds to customers" refers to a function in which a transport device or robot directly responds to customers based on the analyzed inquiry content.

[0639] "Means for automatically generating answers to new inquiries" refers to a function that uses machine learning models to automatically generate appropriate answers.

[0640] "Means of calling an appropriate application program interface to perform processing" refers to a function that automatically executes processes such as shipping replacement equipment, signing new contracts, and changing models by calling an appropriate application program interface in response to a user's operation request.

[0641] Modes for carrying out the invention

[0642] This invention is a system that efficiently receives and analyzes user inquiries and provides appropriate responses. This system is particularly aimed at improving the efficiency of customer service in physical stores. Specific embodiments for implementing this invention are described below.

[0643] System Configuration

[0644] The system consists of user terminals, a server, and a database. Users make queries using the terminals, the server receives and analyzes them, searches for relevant answers, and provides them to the user. Furthermore, a transport device responds to customers as needed.

[0645] The user enters an inquiry.

[0646] Users use a device (e.g., smartphone, tablet, or transport device) to input language-based inquiries into the system. This allows them to ask questions such as the location of products or detailed information about services.

[0647] The device sends the query to the server.

[0648] The terminal sends the entered query content to the server. The server receives this via the API endpoint and begins parsing the query.

[0649] The server parses the query.

[0650] The server analyzes the query content using a natural language processing (NLP) model. For example, the NLP model extracts key keywords and context from the text to understand the intent of the query. Techniques such as morphological analysis and syntax analysis are used in this step. Libraries such as spaCy are examples of NLP models that can be used.

[0651] The server searches for relevant answers.

[0652] The server searches the database for relevant answers based on the analyzed query. The database contains pre-configured frequently asked questions (FAQs) and store manuals, and the search engine uses this information to return the most appropriate answer. SQLite and MySQL are used as database management systems (DBMS).

[0653] The server sends the response to the terminal.

[0654] The server sends the search results to the device. The device displays this to the user as a chat window or notification. This allows the user to quickly obtain the appropriate information.

[0655] The server stores queries and answers in a database.

[0656] The server stores user inquiries and their corresponding responses in a database. This data is used to improve the accuracy of automated responses to future inquiries.

[0657] The server trains the machine learning model.

[0658] The server trains a machine learning model using the accumulated data. This model improves the response generation algorithm to handle new queries, resulting in more accurate automated responses. Machine learning libraries such as TensorFlow and PyTorch are used.

[0659] The conveying equipment responds to the customer.

[0660] A transport device (e.g., a robot) can guide customers within a store based on the analyzed inquiry. For example, if a customer asks, "Where is this product?", the device can guide the customer to the location where the product is located based on the analyzed results.

[0661] Specific example

[0662] For example, if a user submits a request such as "Please tell me which items are on sale," the server analyzes the request and extracts relevant keywords like "sale" and "item." The server then searches its database for an answer such as "The current sale items are A, B, and C" and sends it to the user's terminal. This allows the user to quickly obtain the information they need.

[0663] Example of a prompt

[0664] "Please analyze the following user inquiry and provide an appropriate answer: 'What items are on sale?'"

[0665] The above describes the embodiments for carrying out the present invention. This system enables efficient customer service in physical stores and improves customer satisfaction.

[0666] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0667] Program processing flow

[0668] Step 1:

[0669] The user enters a language-based inquiry using a terminal. The entered inquiry is captured as text data by the user's terminal's input interface. For example, the user enters the inquiry, "Please tell me what items are on sale."

[0670] Step 2:

[0671] The terminal sends the entered query content to the server. Here, the query data is sent to the server's API endpoint via an HTTP POST request. The input data is the text of the query.

[0672] Step 3:

[0673] The server analyzes the received query using a natural language processing (NLP) model. Specifically, it uses an NLP model (e.g., spaCy) to analyze the text data and extract the main keywords "sale" and "product." In this step, the text of the input data is converted into the main keywords.

[0674] Step 4:

[0675] The server searches for relevant answers from a pre-configured database based on the parsed query. For example, it might use an SQLite database to search for information on "sale items." In this step, it uses the entered keywords to execute a database query that contains the appropriate answers.

[0676] Step 5:

[0677] The server sends the search results to the terminal. Specifically, it sends the answers back to the terminal as an HTTP response. In this step, the answers retrieved from the database are sent to the terminal as output data.

[0678] Step 6:

[0679] The terminal displays the received response to the user. The response content is displayed on the user's terminal screen. For example, it might say, "Current sale items are A, B, and C."

[0680] Step 7:

[0681] The server stores user inquiries and their corresponding answers in a database. This improves the accuracy of automated responses to future inquiries. In this step, the input data (inquiry content) and output data (answer) are stored in the database as a single record.

[0682] Step 8:

[0683] The server trains machine learning models using the accumulated data. Specifically, it uses libraries such as TensorFlow and PyTorch to train the models and improve their accuracy. In this step, the historical data stored in the database is used as input data for the machine learning models, and the model parameters are updated.

[0684] Step 9:

[0685] The transport system responds to the customer based on the analyzed inquiry. For example, a robot guides the customer to the location where the product is located. In this step, the transport system physically operates based on the input data (analyzed inquiry).

[0686] Through the processing steps described above, this system can streamline customer service in physical stores and provide timely and appropriate information. Furthermore, machine learning using accumulated data continuously improves the system's response accuracy.

[0687] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0688] This invention is a system that efficiently receives user inquiries, provides appropriate answers, and optimizes responses by recognizing user emotions. This achieves improved operational efficiency and enhanced user experience. Specific embodiments for implementing this invention are described below.

[0689] System Configuration

[0690] The system consists of a user terminal, a server, a database, and an emotion engine. Users make queries using the terminal, the server receives and analyzes them, searches for relevant answers, and provides them to the user. The emotion engine also recognizes the user's emotions and provides appropriate responses accordingly.

[0691] The user enters an inquiry.

[0692] Users use their devices (e.g., PCs or smartphones) to input language-based inquiries into the system. This allows them to ask questions about product and service details, account setup instructions, and more.

[0693] The device sends the query to the server.

[0694] The terminal sends the entered query content to the server. The server receives this via the API endpoint and begins parsing the query.

[0695] The server parses the query.

[0696] The server uses a natural language processing (NLP) model to analyze the query. Specifically, it uses morphological and syntax analysis to extract key keywords and context to understand the intent of the query.

[0697] The server uses an emotion engine to recognize emotions.

[0698] The server uses an emotion engine to recognize emotions from the user's inquiry. For example, it uses text analysis to determine whether the user is angry, anxious, or calm.

[0699] The server searches the FAQ database.

[0700] The server searches the database for relevant answers based on the analyzed query. The database contains pre-configured FAQs and manuals, and the search engine uses this information to return the most suitable answer.

[0701] The server generates responses based on emotions.

[0702] The server adjusts its responses and tone to be more user-friendly based on the analysis results of the emotion engine. For example, if the user is angry, it will generate a response that includes a polite apology.

[0703] The server sends the response to the terminal.

[0704] The server sends the generated response to the device. The device displays this to the user as a chat window or notification. This allows the user to quickly obtain the appropriate information.

[0705] The server stores queries and answers in a database.

[0706] The server stores user inquiries and their corresponding responses in a database. This data is used to improve the accuracy of automated responses to future inquiries.

[0707] The server trains the machine learning model.

[0708] The server uses accumulated data to train a machine learning model. This model improves the response generation algorithm to handle new queries, resulting in more accurate automated responses.

[0709] Automated operation

[0710] Furthermore, if a user requests specific actions such as sending a replacement device, signing a new contract, or changing models, the server will call the appropriate API to automate these operations. For example, if a user requests "Please send a replacement device," the server will integrate with the customer management system and automatically handle the process of sending the replacement device.

[0711] Specific example

[0712] For example, if a user submits an inquiry such as "Please tell me the conditions for maintenance coverage," the server analyzes the inquiry and extracts relevant keywords such as "maintenance," "coverage," and "conditions." The server then uses an emotion engine to analyze the user's emotions and recognizes that the user is feeling anxious. Next, the server searches its database for an appropriate answer, such as "Maintenance services are only applicable within the product warranty period," and generates a response that includes a more considerate explanation that takes the user's feelings into account. This response is then sent to the terminal and displayed to the user. This allows the user to obtain the necessary information with peace of mind.

[0713] In this way, the system of the present invention efficiently processes user inquiries, automates appropriate responses and operations, and simultaneously improves operational efficiency and user experience by recognizing user emotions and optimizing responses.

[0714] The following describes the processing flow.

[0715] Step 1:

[0716] The user enters the inquiry. The user uses a device (for example, a PC or smartphone) to enter a question in text format, such as "Please tell me the conditions for applying maintenance."

[0717] Step 2:

[0718] The terminal sends the query to the server. The terminal sends the entered query text to the server via the API.

[0719] Step 3:

[0720] The server receives the query. The server receives the query text from the API endpoint and prepares to parse it.

[0721] Step 4:

[0722] The server analyzes the query using a natural language processing (NLP) model. The server performs morphological analysis and extracts key keywords (e.g., maintenance, application, condition).

[0723] Step 5:

[0724] The server uses an emotion engine to recognize the user's emotions. The server analyzes the query text using pattern matching and an emotion dictionary to determine, for example, that the user may be feeling "anxious."

[0725] Step 6:

[0726] The server searches the FAQ database. Based on the extracted keywords, the server searches the FAQ database for relevant answers.

[0727] Step 7:

[0728] The server selects the best answer. The server selects the most appropriate answer from among multiple found responses.

[0729] Step 8:

[0730] The server adjusts the response based on the emotion recognition results. Based on the results of the emotion engine, the server adjusts the response by adding, for example, a gentle tone or polite language.

[0731] Step 9:

[0732] The server generates the response. The server generates the optimal response text and prepares to send it to the terminal in text format.

[0733] Step 10:

[0734] The server sends the response to the device. The server sends the generated response text to the device via the API.

[0735] Step 11:

[0736] The device displays the response to the user. The device displays the received response to the user in a chat window or notification format. For example, it may display a polite explanation such as, "Maintenance service is only applicable within the product warranty period. Please contact us anytime if you have any concerns."

[0737] Step 12:

[0738] The server stores inquiries and answers in a database. The server records the details of the current inquiry and the corresponding answer in the database, and uses this information to handle future inquiries.

[0739] Step 13:

[0740] The server trains a machine learning model. The server periodically trains the machine learning model using accumulated data to improve the accuracy of automated responses.

[0741] Step 14:

[0742] The user requests a specific action. For example, the user might request a specific action through the chat window, such as "Please send a replacement device."

[0743] Step 15:

[0744] The terminal sends an operation request to the server. The terminal sends an operation request to the server via an API.

[0745] Step 16:

[0746] The server processes the request. The server calls the appropriate API (e.g., customer management system) and automatically performs the requested operation (e.g., arranging the shipment of a replacement unit).

[0747] Step 17:

[0748] The server sends the processing results to the terminal. A notification indicating that processing is complete is sent to the terminal.

[0749] Step 18:

[0750] The terminal displays the processing results to the user. The terminal displays the processing results to the user in a chat window or notification format. For example, it might display, "Your replacement device has been shipped. Please wait a while for its arrival."

[0751] Through the steps outlined above, the system efficiently processes user inquiries, automates appropriate responses and operations, and simultaneously improves operational efficiency and user experience by recognizing user emotions and optimizing responses.

[0752] (Example 2)

[0753] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0754] Traditional inquiry response systems often provided uniform answers without considering user emotions, failing to adequately improve user satisfaction. Furthermore, insufficient training data for machine learning models led to decreased accuracy in automated responses to new inquiries. Additionally, a lack of means to automate appropriate processing in response to specific user requests prevented sufficient operational efficiency.

[0755] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0756] In this invention, the server includes means for receiving inquiries from users, means for analyzing inquiries using a natural language processing model, means for searching for relevant answers from a pre-configured database based on the analyzed inquiries, means for providing the retrieved answers to the user, means for storing inquiries and answers in a database, means for training a machine learning model using the data stored in the database to improve responses to new inquiries, means for using an emotion analysis engine that recognizes emotions from the content of the user's inquiry, and means for adjusting the tone and content of the response based on the emotion analysis results. This enables appropriate responses that take into account the user's emotions, thereby increasing satisfaction, improving the accuracy of the machine learning model, and enabling quick and accurate responses to new inquiries. Furthermore, automated processing of specific operation requests can be realized, thus improving operational efficiency.

[0757] A "user" is an individual or organization that uses the system to make an inquiry.

[0758] An "inquiry" refers to a question or request that a user makes to a system.

[0759] A "terminal" refers to an electronic device (such as a PC or smartphone) that a user uses to enter an inquiry.

[0760] A "server" is a computer system that receives user inquiries, analyzes them, and provides appropriate answers.

[0761] A "natural language processing model" is an algorithm or software used to analyze, understand, and generate responses to human language.

[0762] "Analysis" is the process of examining the content of an inquiry and extracting its intent and meaning.

[0763] A "database" is a system for systematically storing inquiries, their answers, and other related information.

[0764] A "sentiment analysis engine" is software or an algorithm that recognizes emotions from user input text and provides the results.

[0765] A "machine learning model" is a collection of algorithms that learn from large amounts of data, extract patterns and knowledge, and then make predictions and classifications based on new data.

[0766] "Searching" is the process of finding specific information from a database.

[0767] "Tone" refers to the way responses are expressed and the writing style, which is adjusted to match the user's emotions.

[0768] An "API" is an interface that allows different software systems to communicate with each other.

[0769] "Automated processing" refers to operations or procedures that a system performs automatically with minimal human intervention.

[0770] This invention provides a system that efficiently receives user inquiries and provides appropriate responses, and further optimizes responses by recognizing user emotions. This achieves improved operational efficiency and enhanced user experience.

[0771] System Configuration

[0772] The system consists of user terminals, servers, a database, and an emotion analysis engine.

[0773] Hardware and software used

[0774] User terminal: A device such as a PC or smartphone used by the user to enter their inquiry.

[0775] Server: A server runs multiple software programs. Details are provided below.

[0776] Natural language processing models: "spaCy" and "NLTK" are used for morphological and syntax analysis.

[0777] Sentiment Analysis Engine: IBM Watson Tone Analyzer is used to analyze user emotions.

[0778] FAQ database search engine: Use "Elasticsearch" for searching.

[0779] Machine learning models: TensorFlow and PyTorch are used for data training and prediction.

[0780] Program processing flow

[0781] The user enters an inquiry.

[0782] Users enter inquiries into the system using their own devices (e.g., PCs or smartphones). For example, a user might enter a question such as, "Please tell me the conditions for applying maintenance."

[0783] The device sends the query to the server.

[0784] The terminal sends the entered query content to the server using an HTTP POST request. The request contains the query content and is sent to the specified API endpoint (e.g., / api / v1 / inquiries).

[0785] The server parses the query.

[0786] The server passes the received request to a natural language processing (NLP) model for analysis. For example, it uses the Python library "spaCy" to perform morphological and syntax analysis, extracting key keywords and context. To understand the intent of the query, the keywords are categorized into specific groups.

[0787] The server recognizes emotions using an emotion engine.

[0788] The server passes the analysis results to a sentiment analysis engine (e.g., IBM Watson Tone Analyzer) to recognize the user's emotions. For example, if an inquiry frequently uses language that includes anxiety or anger, the server will identify those emotions.

[0789] The server searches the database.

[0790] The server searches the FAQ database based on the analysis results. It uses Elasticsearch to quickly find answers related to the inquiry. For example, if a user inquires about maintenance services, the server retrieves information from the database such as "Maintenance services are only applicable within the product warranty period."

[0791] The server generates responses based on emotions.

[0792] The server uses a generative AI model (e.g., GPT-3) based on the sentiment analysis results to generate responses in an appropriate tone for the user. For example, if the user is feeling anxious, it might generate a response such as, "To alleviate your concerns, please rest assured that maintenance services are only applicable within the product warranty period."

[0793] The server sends the response to the terminal.

[0794] The server sends the generated response to the device and displays it to the user. The response is displayed on the user's device in a chat window or as a notification.

[0795] The server stores queries and answers in a database.

[0796] The server stores user inquiries and generated responses in a database. This information can be reused for future inquiries. MySQL and PostgreSQL are commonly used databases.

[0797] The server trains the machine learning model.

[0798] The server uses the accumulated data to train machine learning models. For example, it uses tools like TensorFlow or PyTorch to improve the model's performance and enable it to respond quickly and accurately to new queries.

[0799] Automated operation

[0800] The server automatically responds to specific user requests (e.g., shipping a replacement device, signing a new contract, changing models, etc.). For example, if a user requests "ship a replacement device," the server will integrate with a customer management system (e.g., Salesforce) and automatically initiate the process of shipping the replacement device.

[0801] Specific example

[0802] For example, if a user submits an inquiry such as "Please tell me the conditions for maintenance coverage," the server analyzes the inquiry and extracts relevant keywords such as "maintenance," "coverage," and "conditions." The server then uses a sentiment analysis engine to analyze the user's emotions and recognizes that the user is feeling anxious. Next, the server searches its database for an appropriate answer, such as "Maintenance services are only applicable within the product warranty period," and generates a response that includes a more considerate and detailed explanation that takes the user's feelings into account. This response is then sent to the terminal and displayed to the user. This allows the user to obtain the necessary information with peace of mind.

[0803] In this way, the system of the present invention efficiently processes user inquiries, automates appropriate responses and operations, and simultaneously improves operational efficiency and user experience by recognizing user emotions and optimizing responses.

[0804] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0805] Step 1:

[0806] The user enters an inquiry.

[0807] Users enter inquiries into the system using their own devices (PCs, smartphones, etc.). Specifically, users open a web browser or application and enter questions or requests in text format into input fields. For example, a user might enter the prompt message, "Please tell me the conditions for maintenance application."

[0808] Step 2:

[0809] The device sends the query to the server.

[0810] The terminal sends the entered query content to the server as an HTTP POST request. The input data (user's query text) is included in the request body. The request is sent to the specified API endpoint (e.g., / api / v1 / inquiries). The server receives this request and prepares to process the query content.

[0811] Step 3:

[0812] The server parses the query.

[0813] The server passes the received query content to a natural language processing (NLP) model for analysis. The query text is provided as input, and morphological and syntax analysis is performed. Specifically, the Python library "spaCy" is used to extract key keywords and context. The output provides the analyzed keywords and contextual information.

[0814] Step 4:

[0815] The server recognizes emotions using an emotion engine.

[0816] The server passes the parsed query to the sentiment analysis engine to recognize the user's emotions. The parsed query text is used as input, and the sentiment analysis engine (e.g., IBM Watson Tone Analyzer) analyzes the text. The output is data indicating the user's emotions (anger, anxiety, calmness, etc.).

[0817] Step 5:

[0818] The server searches the FAQ database.

[0819] The server searches the FAQ database for relevant answers based on the analysis results. The query keywords and contextual information are used as input, and a search engine (e.g., Elasticsearch) extracts the appropriate answers. The output is the relevant answers retrieved from the database.

[0820] Step 6:

[0821] The server generates responses based on emotions.

[0822] The server generates responses to the user in an appropriate tone based on the searched answers and sentiment analysis results. The inputs used are answers and sentiment data from a database. A generative AI model (e.g., GPT-3) generates text in an appropriate tone based on this input data. The output is a sentiment-sensitive response text. For example, a response such as "Please rest assured that maintenance services are only applicable within the product warranty period" might be generated.

[0823] Step 7:

[0824] The server sends the response to the terminal.

[0825] The server sends the generated response to the device. The generated response text is used as input and returned to the device as an HTTP response. The device receives this response and displays the response to the user. Specifically, it is displayed as a chat window on a web page or as a notification in an application.

[0826] Step 8:

[0827] The server stores queries and answers in a database.

[0828] The server stores user inquiries and generated responses in a database. Inquiry text and response text are used as input and stored in a database (e.g., MySQL or PostgreSQL). This data is accumulated in the database as output. This accumulated data helps improve future query responses.

[0829] Step 9:

[0830] The server trains the machine learning model.

[0831] The server uses accumulated data to train a machine learning model. Past query and response data is used as input, and the model is trained using a machine learning framework (e.g., TensorFlow or PyTorch). The output is a trained model capable of generating highly accurate responses.

[0832] Step 10:

[0833] The server automates the operation.

[0834] The server automatically responds to specific user requests (e.g., shipping a replacement device, signing a new contract, changing models, etc.). The user's request text is used as input, and the server calls the appropriate API to process it. For example, if a user enters "Please ship a replacement device," the server will integrate with the customer management system and automatically initiate the shipping process. The user receives a notification that the requested operation has been completed.

[0835] (Application Example 2)

[0836] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0837] Conventional inquiry handling systems struggle to respond to user inquiries quickly and appropriately, particularly lacking in responses that take emotions into consideration. Therefore, there is a need for improved user experience and operational efficiency. Furthermore, in security services, emotional recognition is especially crucial in handling emergency inquiries, requiring quick and appropriate responses. This often leads to challenges such as a lack of user confidence and delays in responses.

[0838] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0839] In this invention, the server includes means for recognizing emotions from the content of a user's inquiry, means for adjusting the tone of the response and generating an emotion-appropriate answer based on the recognized emotion, and means for training a machine learning model using data stored in the database to improve responses to new inquiries. This enables quick and appropriate responses that take into account the user's emotions, improving the user experience and operational efficiency.

[0840] "Means for receiving user inquiries" refers to a function that provides an interface for end users to make inquiries to the system via text or voice.

[0841] "Methods for analysis using natural language processing models" refer to functions that convert linguistic information contained in user inquiries into a format that a computer can understand, and then apply a model to analyze the intent and content of the inquiry.

[0842] "Means for searching for relevant answers" refers to a function that finds appropriate answers from a database based on the analyzed query content.

[0843] "Means of providing information to users" refers to functions that display search results in a user-friendly format and provide information to users.

[0844] "Means of recognizing emotions" refers to a function that analyzes the user's inquiry to determine their emotional state, such as anger, anxiety, or calmness.

[0845] "Means for generating emotionally responsive answers" refers to a function that creates more appropriate and friendly responses based on the recognized emotions of the user.

[0846] "Means for storing inquiries and answers in a database" refers to a function that records received inquiries and their responses in a database, which can then be used to handle future inquiries.

[0847] "Methods for training machine learning models" refers to the function of training models that improve the accuracy of automated responses using machine learning algorithms based on a large amount of query and answer data accumulated in a database.

[0848] This invention is a system that efficiently receives user inquiries, provides appropriate answers, and further recognizes user emotions to optimize responses. The system consists of a user terminal, a server, a database, and an emotion engine. Detailed embodiments of this system are described below.

[0849] System Configuration

[0850] Users input language-based queries to the system using devices such as PCs and smartphones. The device sends the entered query content to the server. The server receives this via an API endpoint and parses the query.

[0851] Hardware and software to be used

[0852] Hardware:

[0853] Smartphone or PC (device)

[0854] server

[0855] software:

[0856] Natural Language Processing (NLP) Model: Using the Python transformers library

[0857] Emotion recognition model: Uses the transformers library from Hugging Face.

[0858] Databases: Relational databases such as MySQL and PostgreSQL

[0859] Machine learning models: TensorFlow or PyTorch

[0860] Data processing and data calculation

[0861] Inquiry Analysis

[0862] The server analyzes user inquiries using a natural language processing model. Specifically, it performs morphological and syntax analysis to extract key keywords and their context.

[0863] emotion recognition

[0864] After analyzing the content of the inquiry, an emotion engine is used to recognize the user's emotions. Through text analysis, it is determined whether the user is angry, anxious, or calm.

[0865] Generating an answer

[0866] The server searches for relevant answers based on the analyzed query and the perceived emotion, and adjusts the tone of the response accordingly. For example, if the user is angry, it might include a polite apology.

[0867] Data storage and training of machine learning models

[0868] All inquiries and answers are stored in a database. This data is used to train machine learning models, improving the accuracy of responses to new inquiries.

[0869] Specific example

[0870] Specific inquiry scenarios and responses

[0871] User inquiry: "I heard gunshots nearby."

[0872] System response: "Please stay calm. Close windows and doors and move to a safe place immediately."

[0873] Examples of prompts for a generative AI model:

[0874] "When a user reports hearing gunshots nearby, analyze their emotions and generate a quick response to ensure their safety."

[0875] In this way, the system of the invention can efficiently and quickly process user inquiries and generate appropriate responses. This system optimizes responses to urgent user inquiries, particularly in the field of security services, and provides users with peace of mind.

[0876] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0877] Step 1:

[0878] The user enters their inquiry.

[0879] The user enters their inquiry in text format using a terminal. After input, the inquiry data is sent from the terminal to the server.

[0880] Input: Text data entered by the user on the device.

[0881] Output: Query data sent to the server

[0882] Step 2:

[0883] The server receives the query.

[0884] The server receives query data sent from the terminal via the API endpoint.

[0885] Input: Inquiry data sent from the terminal

[0886] Output: Text data received by the server

[0887] Step 3:

[0888] The server analyzes the query using a natural language processing model.

[0889] The server uses an NLP model (transformers library) to perform morphological and syntax analysis on the query content. It extracts key keywords and context to understand the intent of the query.

[0890] Input: Text data received by the server

[0891] Output: Intent and key keywords of the query

[0892] Step 4:

[0893] The server uses an emotion recognition model to recognize emotions.

[0894] The server uses an emotion recognition model (transformers library) to determine the user's emotions from the analyzed query content. For example, it can determine whether the user is angry, anxious, or calm.

[0895] Input: Intent and main keywords of the inquiry

[0896] Output: User's emotional state

[0897] Step 5:

[0898] The server searches the FAQ database.

[0899] The server searches the database for relevant answers based on the analyzed query content and sentiment state. If a relevant answer is found, it retrieves it.

[0900] Input: Inquiry intent, key keywords, user sentiment

[0901] Output: Related answers

[0902] Step 6:

[0903] The server generates responses based on emotions.

[0904] Based on the emotion recognition results, the server adjusts the tone to be user-friendly and generates an appropriate response. For example, if the user is angry, it will generate a response that includes a polite apology.

[0905] Input: Related answers, user's emotional state

[0906] Output: Emotionally adjusted response

[0907] Step 7:

[0908] The server sends the response to the terminal.

[0909] The server sends the generated response to the device. The device then displays this to the user as a chat window or notification.

[0910] Input: Emotionally adjusted response

[0911] Output: The answer displayed on the user's device.

[0912] Step 8:

[0913] The server stores inquiries and answers in a database.

[0914] The server records the received inquiry and the generated response in a database. This data will be used to improve the accuracy of future inquiry handling.

[0915] Input: Inquiry details, generated response

[0916] Output: Queries and answers stored in the database

[0917] Step 9:

[0918] The server trains the machine learning model.

[0919] The server uses the accumulated data to train a machine learning model. This model improves the response generation algorithm to handle new queries.

[0920] Input: Inquiries and answers stored in the database

[0921] Output: Trained machine learning model

[0922] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0923] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0924] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0925] [Third Embodiment]

[0926] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0927] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0928] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0929] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0930] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0931] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0932] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0933] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0934] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0935] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0936] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0937] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0938] This invention is a system that efficiently receives and analyzes user inquiries and provides appropriate responses. This achieves improved operational efficiency and cost reduction. Specific embodiments for implementing this invention are described below.

[0939] System Configuration

[0940] The system consists of user terminals, a server, and a database. Users make queries using the terminals, the server receives and analyzes them, searches for relevant answers, and provides them to the user.

[0941] The user enters an inquiry.

[0942] Users use their devices (e.g., PCs or smartphones) to input language-based inquiries into the system. This allows them to ask questions about seasonal products and services, how to set up an account, and more.

[0943] The device sends the query to the server.

[0944] The terminal sends the entered query content to the server. The server receives this via the API endpoint and begins parsing the query.

[0945] The server parses the query.

[0946] The server analyzes the query using a natural language processing (NLP) model. For example, the NLP model extracts key keywords and context from the text to understand the intent of the query. Techniques such as morphological analysis and syntax analysis are used in this step.

[0947] The server searches for relevant answers.

[0948] The server searches the database for relevant answers based on the analyzed query. The database contains pre-configured FAQs and manuals, and the search engine uses this information to return the most suitable answer.

[0949] The server sends the response to the terminal.

[0950] The server sends the search results to the device. The device displays this to the user as a chat window or notification. This allows the user to quickly obtain the appropriate information.

[0951] The server stores queries and answers in a database.

[0952] The server stores user inquiries and their corresponding responses in a database. This data is used to improve the accuracy of automated responses to future inquiries.

[0953] The server trains the machine learning model.

[0954] The server uses accumulated data to train a machine learning model. This model improves the response generation algorithm to handle new queries, resulting in more accurate automated responses.

[0955] Specific example

[0956] For example, if a user submits a query such as "Please tell me the conditions for applying maintenance," the server analyzes the query and extracts the relevant keywords "maintenance," "application," and "conditions." The server then searches its database for the appropriate answer, such as "Maintenance services are only applicable within the product warranty period," and sends it to the user's terminal. This allows the user to quickly obtain the necessary information.

[0957] Automated operation

[0958] Furthermore, if a user requests specific actions such as sending a replacement device, signing a new contract, or changing models, the server will call the appropriate API to automate these operations. For example, if a user requests "Please send a replacement device," the server will integrate with the customer management system and automatically handle the process of sending the replacement device.

[0959] In this way, the system of the present invention efficiently processes user inquiries and automates appropriate responses and operations, thereby achieving improved operational efficiency and cost reduction.

[0960] The following describes the processing flow.

[0961] Step 1:

[0962] The user enters the inquiry. The user uses a device (e.g., a PC or smartphone) to input questions or requests to the system in text format.

[0963] Step 2:

[0964] The device sends the query to the server. The device sends the entered query content to the server via the API endpoint.

[0965] Step 3:

[0966] The server receives the query. The server retrieves the text data received from the API endpoint.

[0967] Step 4:

[0968] The server analyzes the query. The server uses a natural language processing (NLP) model to analyze the query content. Specifically, it uses morphological analysis and syntax analysis to extract key keywords and context.

[0969] Step 5:

[0970] The server extracts keywords. It extracts important keywords, mainly nouns and verbs, from the query text.

[0971] Step 6:

[0972] The server searches the FAQ database. Based on the extracted keywords, the server searches the FAQ database for relevant answers.

[0973] Step 7:

[0974] The server selects the most appropriate answer. If multiple answer choices exist, the server selects the most relevant answer.

[0975] Step 8:

[0976] The server generates the answer. Based on the selected answer, it generates the final text to be provided to the user.

[0977] Step 9:

[0978] The server sends the response to the device. An API endpoint is used to send the generated response text to the device.

[0979] Step 10:

[0980] The device displays the response to the user. The device displays the received response to the user in a chat window or notification format.

[0981] Step 11:

[0982] The server stores queries and answers in a database. The query content and the corresponding answers are recorded in the database and used for future query processing.

[0983] Step 12:

[0984] The server trains a machine learning model. Using accumulated data, the machine learning model is periodically trained to improve the accuracy of automated responses to inquiries.

[0985] Step 13:

[0986] The user requests a specific action. Specific requests, such as sending a replacement device, signing a new contract, or changing models, are entered in text format.

[0987] Step 14:

[0988] The server processes the user's request. It calls the appropriate API and performs the specific action required for the request (e.g., arranging for a replacement device to be shipped).

[0989] Step 15:

[0990] The server sends the processing results to the terminal. A notification indicating that processing is complete is sent to the terminal.

[0991] Step 16:

[0992] The terminal displays the processing results to the user. The terminal displays the processing results to the user in a chat window or notification format.

[0993] Through the steps described above, the system can efficiently process user inquiries and automate appropriate responses and actions.

[0994] (Example 1)

[0995] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0996] Conventional inquiry handling systems suffered from long response times to user inquiries and low accuracy in providing appropriate answers. Furthermore, the need for numerous manual operations to find appropriate answers prevented sufficient efficiency improvements and cost reductions. The objective of this invention is to solve these problems and provide users with efficient and highly accurate responses.

[0997] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0998] In this invention, the server includes means for receiving inquiries from users, means for analyzing the inquiries using a natural language processing model, means for sending the input inquiries to the server via an API endpoint, means for searching for relevant answers from a pre-configured data storage based on the analyzed inquiries, means for providing the retrieved answers to the user, means for storing the inquiries and answers in the data storage, and means for training a computational model using the data stored in the data storage to improve the response to new inquiries. This enables a rapid and accurate response to user inquiries, efficient business operations, and cost reduction.

[0999] An "inquiry" is the act of a user inputting information they want to know or a problem they want to solve into a system using natural language.

[1000] A "natural language processing model" is a type of artificial intelligence that analyzes text written in natural language, understands its content and meaning, and processes it.

[1001] An "API endpoint" is part of an application programming interface and is a connection point for different software systems to communicate with each other.

[1002] A "server" is a computer system that provides services to other computers via a network.

[1003] "Data storage" refers to devices and systems for storing digital data, and includes databases and cloud storage.

[1004] "Analysis" is the act of breaking down input data into its components, understanding its content, and identifying its meaning and intent.

[1005] An "answer" refers to the information or solution provided in response to a user's inquiry.

[1006] A "machine learning model" is an algorithm that learns from data and performs tasks such as prediction and classification.

[1007] A "computational model" is a mathematical model used to perform a specific task through numerical calculations and data processing.

[1008] A "programming interface" is a set of definitions and protocols that enable different software programs to work together smoothly.

[1009] "Accumulation" is the act of collecting data and saving it.

[1010] This invention is a system that efficiently receives and analyzes user inquiries and provides appropriate responses. This achieves improved operational efficiency and cost reduction. Specific embodiments of this invention are described below.

[1011] System Configuration

[1012] This system consists of user terminals, a server, and data storage. Users use the terminals to make inquiries, the server receives and analyzes them, searches for relevant answers, and provides them to the user.

[1013] The user enters an inquiry.

[1014] Users use their devices (such as PCs or smartphones) to input natural language queries into the system. For example, they might ask, "I'd like to know about the product's warranty period."

[1015] The device sends the query to the server.

[1016] The terminal sends the entered query content as an HTTP POST request to the server's API endpoint. The server receives this and begins parsing the query.

[1017] The server parses the query.

[1018] The server uses an NLP (Natural Language Processing) model to analyze the query. This analysis utilizes NLP libraries such as spaCy and Hugging Face in Python. For example, morphological analysis is used to extract the keyword "product warranty period."

[1019] The server searches for relevant answers.

[1020] The server searches for relevant answers in data storage based on the extracted keywords. Data storage may include SQL Server or Elasticsearch. The server finds the best answer, retrieving information such as "The product warranty period is one year."

[1021] The server sends the response to the terminal.

[1022] The server sends the retrieved response to the device in JSON format. The device then displays this response to the user in a chat window or notification, allowing the user to quickly obtain the necessary information.

[1023] The server stores queries and answers in data storage.

[1024] The server records the query content and the corresponding response in a separate data storage. This data will be used to improve future query responses.

[1025] The server trains the machine learning model.

[1026] The server uses data stored in data storage to train machine learning models. For example, it uses Python's Scikit-learn or TensorFlow to improve the accuracy of the models. This enables sophisticated automated responses to new queries.

[1027] Automated operation

[1028] Furthermore, this system can handle specific user requests such as shipping replacement equipment, signing up for new contracts, or changing models. For example, if a user requests "Please ship a replacement unit," the server will call the appropriate programming interface and automatically execute the operation. This programming interface integrates with customer management systems, logistics systems, and other systems to automate necessary procedures.

[1029] Specific example

[1030] For example, if a user asks, "What are the conditions for maintenance?", the server analyzes the inquiry and extracts the keywords "maintenance," "applicability," and "conditions." The server then searches its database for the answer, "Maintenance services are applicable only within the product warranty period," and sends it to the user's terminal. This allows the user to quickly obtain the necessary information.

[1031] Thus, the present invention enables prompt and accurate responses to user inquiries, thereby improving operational efficiency and reducing costs.

[1032] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1033] Step 1:

[1034] The user enters an inquiry.

[1035] Users enter their inquiries in natural language into a text form using a device such as a PC or smartphone. Specifically, they might type "Please tell me about the product's warranty period" and click the submit button. The entered text is temporarily stored on the user's device.

[1036] Step 2:

[1037] The device sends the query to the server.

[1038] When the user clicks the submit button, the device sends the entered query content to the server as an HTTP POST request. For example, it sends a request with data in JSON format like the following:

[1039] json

[1040] {

[1041] "query": "Please tell me about the product's warranty period."

[1042] }

[1043] The server receives this request and temporarily holds the query details in memory.

[1044] Step 3:

[1045] The server parses the query.

[1046] The server inputs the received query into a natural language processing (NLP) model and begins analysis. Specifically, it uses a Python NLP library (e.g., spaCy) to perform morphological analysis and extract keywords and phrases from the text. For example, the keyword "product warranty period" might be extracted. The input is the received query, and the output is the extracted keywords.

[1047] Step 4:

[1048] The server searches for relevant answers.

[1049] The server searches for information in data storage based on keywords extracted from the analyzed query. It uses SQL Server, Elasticsearch, etc., to retrieve relevant answers. Specifically, it executes SQL queries like the following:

[1050] sql

[1051] SELECT answer FROM faq WHERE question LIKE '%warranty period%'

[1052] The input is the extracted keywords, and the output is the appropriate answer. This query yields the answer, "The product warranty period is one year."

[1053] Step 5:

[1054] The server sends the response to the terminal.

[1055] The server sends the retrieved response to the terminal as a JSON response. Specifically, it sends it in the following format:

[1056] json

[1057] {

[1058] "Answer": "The product warranty period is one year."

[1059] }

[1060] The terminal receives this data and displays it in an inquiry form or a dedicated window. The output is displayed on the user's terminal.

[1061] Step 6:

[1062] The user receives the response.

[1063] The user checks the answer displayed on their device. Specifically, they can see a message such as "The product warranty period is one year" and get the appropriate answer to their inquiry.

[1064] Step 7:

[1065] The server stores queries and answers in a database.

[1066] The server stores the query details and corresponding answers in a database. This data is used to improve future responses. Specifically, the data is stored as follows:

[1067] sql

[1068] INSERT INTO logs (query, answer) VALUES ('Please tell me about the product's warranty period', 'The product's warranty period is 1 year')

[1069] The input consists of the inquiry content and the response, and the output is the storage of the data in the database.

[1070] Step 8:

[1071] The server trains the machine learning model.

[1072] The server trains a machine learning model using historical query data stored in the database. Specifically, it uses Python's Scikit-learn and TensorFlow to improve the model's accuracy. This model is designed to increase the accuracy of generating answers to new queries. The input is historical query data, and the output is the updated machine learning model.

[1073] (Application Example 1)

[1074] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1075] In modern brick-and-mortar stores, it is necessary to respond quickly and accurately to a wide range of customer inquiries, but relying on manual responses makes efficient and cost-effective operation difficult. Furthermore, if customers cannot obtain information about products and services immediately, their willingness to purchase may decrease. In addition, manual processing is time-consuming when performing specific actions based on user inquiries, leading to decreased customer satisfaction. This invention aims to solve these problems and achieve efficiency and automation in customer service in physical stores.

[1076] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1077] In this invention, the server includes means for receiving inquiries from users, means for analyzing inquiries using a natural language processing model, and means for searching for relevant answers from a pre-configured database based on the analyzed inquiries. This enables the rapid provision of answers to customer inquiries. Furthermore, by including means for customer interaction in the transport device, it becomes possible to automate tasks such as guiding customers to product locations and assisting with purchase procedures. This streamlines customer service in physical stores and improves customer satisfaction.

[1078] definition statement

[1079] "Means of receiving user inquiries" refers to the function of receiving any inquiries that users input into the system via their terminals.

[1080] "Methods for analysis using natural language processing models" refers to a function that analyzes input queries using natural language processing technology, extracts key keywords and context, and understands the user's intent.

[1081] "Means of searching for relevant answers from a pre-configured database" refers to a function that searches for the most suitable answer from a pre-configured database based on the analyzed query content.

[1082] "Means of providing information to the user" refers to functions that notify or display the searched answers on the user's device.

[1083] "Means for storing inquiries and answers in a database" refers to a function that saves user inquiries and the corresponding answers in a database.

[1084] "Means of training machine learning models to improve responses to new inquiries" refers to a function that uses accumulated data to train machine learning models and improve the accuracy and effectiveness of responses to future inquiries.

[1085] "Means by which a transport device responds to customers" refers to a function in which a transport device or robot directly responds to customers based on the analyzed inquiry content.

[1086] "Means for automatically generating answers to new inquiries" refers to a function that uses machine learning models to automatically generate appropriate answers.

[1087] "Means of calling an appropriate application program interface to perform processing" refers to a function that automatically executes processes such as shipping replacement equipment, signing new contracts, and changing models by calling an appropriate application program interface in response to a user's operation request.

[1088] Modes for carrying out the invention

[1089] This invention is a system that efficiently receives and analyzes user inquiries and provides appropriate responses. This system is particularly aimed at improving the efficiency of customer service in physical stores. Specific embodiments for implementing this invention are described below.

[1090] System Configuration

[1091] The system consists of user terminals, a server, and a database. Users make queries using the terminals, the server receives and analyzes them, searches for relevant answers, and provides them to the user. Furthermore, a transport device responds to customers as needed.

[1092] The user enters an inquiry.

[1093] Users use a device (e.g., smartphone, tablet, or transport device) to input language-based inquiries into the system. This allows them to ask questions such as the location of products or detailed information about services.

[1094] The device sends the query to the server.

[1095] The terminal sends the entered query content to the server. The server receives this via the API endpoint and begins parsing the query.

[1096] The server parses the query.

[1097] The server analyzes the query content using a natural language processing (NLP) model. For example, the NLP model extracts key keywords and context from the text to understand the intent of the query. Techniques such as morphological analysis and syntax analysis are used in this step. Libraries such as spaCy are examples of NLP models that can be used.

[1098] The server searches for relevant answers.

[1099] The server searches the database for relevant answers based on the analyzed query. The database contains pre-configured frequently asked questions (FAQs) and store manuals, and the search engine uses this information to return the most appropriate answer. SQLite and MySQL are used as database management systems (DBMS).

[1100] The server sends the response to the terminal.

[1101] The server sends the search results to the device. The device displays this to the user as a chat window or notification. This allows the user to quickly obtain the appropriate information.

[1102] The server stores queries and answers in a database.

[1103] The server stores user inquiries and their corresponding responses in a database. This data is used to improve the accuracy of automated responses to future inquiries.

[1104] The server trains the machine learning model.

[1105] The server trains a machine learning model using the accumulated data. This model improves the response generation algorithm to handle new queries, resulting in more accurate automated responses. Machine learning libraries such as TensorFlow and PyTorch are used.

[1106] The conveying equipment responds to the customer.

[1107] A transport device (e.g., a robot) can guide customers within a store based on the analyzed inquiry. For example, if a customer asks, "Where is this product?", the device can guide the customer to the location of the product based on the analyzed results.

[1108] Specific example

[1109] For example, if a user submits a request such as "Please tell me which items are on sale," the server analyzes the request and extracts relevant keywords like "sale" and "items." The server then searches its database for an answer such as "The current sale items are A, B, and C" and sends it to the user's terminal. This allows the user to quickly obtain the information they need.

[1110] Example of a prompt

[1111] "Please analyze the following user inquiry and provide an appropriate answer: 'What items are on sale?'"

[1112] The above describes the embodiments for carrying out the present invention. This system enables efficient customer service in physical stores and improves customer satisfaction.

[1113] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1114] Program processing flow

[1115] Step 1:

[1116] The user enters a language-based inquiry using a terminal. The entered inquiry is captured as text data by the user's terminal's input interface. For example, the user enters the inquiry, "Please tell me what items are on sale."

[1117] Step 2:

[1118] The terminal sends the entered query content to the server. Here, the query data is sent to the server's API endpoint via an HTTP POST request. The input data is the text of the query.

[1119] Step 3:

[1120] The server analyzes the received query using a natural language processing (NLP) model. Specifically, it uses an NLP model (e.g., spaCy) to analyze the text data and extract the main keywords "sale" and "product." In this step, the text of the input data is converted into the main keywords.

[1121] Step 4:

[1122] The server searches for relevant answers from a pre-configured database based on the parsed query. For example, it might use an SQLite database to search for information on "sale items." In this step, it uses the entered keywords to execute a database query that contains the appropriate answers.

[1123] Step 5:

[1124] The server sends the search results to the terminal. Specifically, it sends the answers back to the terminal as an HTTP response. In this step, the answers retrieved from the database are sent to the terminal as output data.

[1125] Step 6:

[1126] The terminal displays the received response to the user. The response content is displayed on the user's terminal screen. For example, it might say, "Current sale items are A, B, and C."

[1127] Step 7:

[1128] The server stores user inquiries and their corresponding answers in a database. This improves the accuracy of automated responses to future inquiries. In this step, the input data (inquiry content) and output data (answer) are stored in the database as a single record.

[1129] Step 8:

[1130] The server trains machine learning models using the accumulated data. Specifically, it uses libraries such as TensorFlow and PyTorch to train the models and improve their accuracy. In this step, the historical data stored in the database is used as input data for the machine learning models, and the model parameters are updated.

[1131] Step 9:

[1132] The transport system responds to the customer based on the analyzed inquiry. For example, a robot guides the customer to the location where the product is located. In this step, the transport system physically operates based on the input data (analyzed inquiry).

[1133] Through the processing steps described above, this system can streamline customer service in physical stores and provide timely and appropriate information. Furthermore, machine learning using accumulated data continuously improves the system's response accuracy.

[1134] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1135] This invention is a system that efficiently receives user inquiries, provides appropriate answers, and optimizes responses by recognizing user emotions. This achieves improved operational efficiency and enhanced user experience. Specific embodiments for implementing this invention are described below.

[1136] System Configuration

[1137] The system consists of a user terminal, a server, a database, and an emotion engine. Users make queries using the terminal, the server receives and analyzes them, searches for relevant answers, and provides them to the user. The emotion engine also recognizes the user's emotions and provides appropriate responses accordingly.

[1138] The user enters an inquiry.

[1139] Users use their devices (e.g., PCs or smartphones) to input language-based inquiries into the system. This allows them to ask questions about product and service details, account setup instructions, and more.

[1140] The device sends the query to the server.

[1141] The terminal sends the entered query content to the server. The server receives this via the API endpoint and begins parsing the query.

[1142] The server parses the query.

[1143] The server uses a natural language processing (NLP) model to analyze the query. Specifically, it uses morphological and syntax analysis to extract key keywords and context to understand the intent of the query.

[1144] The server uses an emotion engine to recognize emotions.

[1145] The server uses an emotion engine to recognize emotions from the user's inquiry. For example, it uses text analysis to determine whether the user is angry, anxious, or calm.

[1146] The server searches the FAQ database.

[1147] The server searches the database for relevant answers based on the analyzed query. The database contains pre-configured FAQs and manuals, and the search engine uses this information to return the most suitable answer.

[1148] The server generates responses based on emotions.

[1149] The server adjusts its responses and tone to be more user-friendly based on the analysis results of the emotion engine. For example, if the user is angry, it will generate a response that includes a polite apology.

[1150] The server sends the response to the terminal.

[1151] The server sends the generated response to the device. The device displays this to the user as a chat window or notification. This allows the user to quickly obtain the appropriate information.

[1152] The server stores queries and answers in a database.

[1153] The server stores user inquiries and their corresponding responses in a database. This data is used to improve the accuracy of automated responses to future inquiries.

[1154] The server trains the machine learning model.

[1155] The server uses accumulated data to train a machine learning model. This model improves the response generation algorithm to handle new queries, resulting in more accurate automated responses.

[1156] Automated operation

[1157] Furthermore, if a user requests specific actions such as sending a replacement device, signing a new contract, or changing models, the server will call the appropriate API to automate these operations. For example, if a user requests "Please send a replacement device," the server will integrate with the customer management system and automatically handle the process of sending the replacement device.

[1158] Specific example

[1159] For example, if a user submits an inquiry such as "Please tell me the conditions for maintenance coverage," the server analyzes the inquiry and extracts relevant keywords such as "maintenance," "coverage," and "conditions." The server then uses an emotion engine to analyze the user's emotions and recognizes that the user is feeling anxious. Next, the server searches its database for an appropriate answer, such as "Maintenance services are only applicable within the product warranty period," and generates a response that includes a more considerate explanation that takes the user's feelings into account. This response is then sent to the terminal and displayed to the user. This allows the user to obtain the necessary information with peace of mind.

[1160] In this way, the system of the present invention efficiently processes user inquiries, automates appropriate responses and operations, and simultaneously improves operational efficiency and user experience by recognizing user emotions and optimizing responses.

[1161] The following describes the processing flow.

[1162] Step 1:

[1163] The user enters the inquiry. The user uses a device (for example, a PC or smartphone) to enter a question in text format, such as "Please tell me the conditions for applying maintenance."

[1164] Step 2:

[1165] The terminal sends the query to the server. The terminal sends the entered query text to the server via the API.

[1166] Step 3:

[1167] The server receives the query. The server receives the query text from the API endpoint and prepares to parse it.

[1168] Step 4:

[1169] The server analyzes the query using a natural language processing (NLP) model. The server performs morphological analysis and extracts key keywords (e.g., maintenance, application, condition).

[1170] Step 5:

[1171] The server uses an emotion engine to recognize the user's emotions. The server analyzes the query text using pattern matching and an emotion dictionary to determine, for example, that the user may be feeling "anxious."

[1172] Step 6:

[1173] The server searches the FAQ database. Based on the extracted keywords, the server searches the FAQ database for relevant answers.

[1174] Step 7:

[1175] The server selects the best answer. The server selects the most appropriate answer from among multiple found responses.

[1176] Step 8:

[1177] The server adjusts the response based on the emotion recognition results. Based on the results of the emotion engine, the server adjusts the response by adding, for example, a gentle tone or polite language.

[1178] Step 9:

[1179] The server generates the response. The server generates the optimal response text and prepares to send it to the terminal in text format.

[1180] Step 10:

[1181] The server sends the response to the device. The server sends the generated response text to the device via the API.

[1182] Step 11:

[1183] The device displays the response to the user. The device displays the received response to the user in a chat window or notification format. For example, it may display a polite explanation such as, "Maintenance service is only applicable within the product warranty period. Please contact us anytime if you have any concerns."

[1184] Step 12:

[1185] The server stores inquiries and answers in a database. The server records the details of the current inquiry and the corresponding answer in the database, and uses this information to handle future inquiries.

[1186] Step 13:

[1187] The server trains a machine learning model. The server periodically trains the machine learning model using accumulated data to improve the accuracy of automated responses.

[1188] Step 14:

[1189] The user requests a specific action. For example, the user might request a specific action through the chat window, such as "Please send a replacement device."

[1190] Step 15:

[1191] The terminal sends an operation request to the server. The terminal sends an operation request to the server via an API.

[1192] Step 16:

[1193] The server processes the request. The server calls the appropriate API (e.g., customer management system) and automatically performs the requested operation (e.g., arranging the shipment of a replacement unit).

[1194] Step 17:

[1195] The server sends the processing results to the terminal. A notification indicating that processing is complete is sent to the terminal.

[1196] Step 18:

[1197] The terminal displays the processing results to the user. The terminal displays the processing results to the user in a chat window or notification format. For example, it might display, "Your replacement device has been shipped. Please wait a while for its arrival."

[1198] Through the steps outlined above, the system efficiently processes user inquiries, automates appropriate responses and operations, and simultaneously improves operational efficiency and user experience by recognizing user emotions and optimizing responses.

[1199] (Example 2)

[1200] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1201] Traditional inquiry response systems often provided uniform answers without considering user emotions, failing to adequately improve user satisfaction. Furthermore, insufficient training data for machine learning models led to decreased accuracy in automated responses to new inquiries. Additionally, a lack of means to automate appropriate processing in response to specific user requests prevented sufficient operational efficiency.

[1202] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1203] In this invention, the server includes means for receiving inquiries from users, means for analyzing inquiries using a natural language processing model, means for searching for relevant answers from a pre-configured database based on the analyzed inquiries, means for providing the retrieved answers to the user, means for storing inquiries and answers in a database, means for training a machine learning model using the data stored in the database to improve responses to new inquiries, means for using an emotion analysis engine that recognizes emotions from the content of the user's inquiry, and means for adjusting the tone and content of the response based on the emotion analysis results. This enables appropriate responses that take into account the user's emotions, thereby increasing satisfaction, improving the accuracy of the machine learning model, and enabling quick and accurate responses to new inquiries. Furthermore, automated processing of specific operation requests can be realized, thus improving operational efficiency.

[1204] A "user" is an individual or organization that uses the system to make an inquiry.

[1205] An "inquiry" refers to a question or request that a user makes to a system.

[1206] A "terminal" refers to an electronic device (such as a PC or smartphone) that a user uses to enter an inquiry.

[1207] A "server" is a computer system that receives user inquiries, analyzes them, and provides appropriate answers.

[1208] A "natural language processing model" is an algorithm or software used to analyze, understand, and generate responses to human language.

[1209] "Analysis" is the process of examining the content of an inquiry and extracting its intent and meaning.

[1210] A "database" is a system for systematically storing inquiries, their answers, and other related information.

[1211] A "sentiment analysis engine" is software or an algorithm that recognizes emotions from user input text and provides the results.

[1212] A "machine learning model" is a collection of algorithms that learn from large amounts of data, extract patterns and knowledge, and then make predictions and classifications based on new data.

[1213] "Searching" is the process of finding specific information from a database.

[1214] "Tone" refers to the way responses are expressed and the writing style, which is adjusted to match the user's emotions.

[1215] An "API" is an interface that allows different software systems to communicate with each other.

[1216] "Automated processing" refers to operations or procedures that a system performs automatically with minimal human intervention.

[1217] This invention provides a system that efficiently receives user inquiries and provides appropriate responses, and further optimizes responses by recognizing user emotions. This achieves improved operational efficiency and enhanced user experience.

[1218] System Configuration

[1219] The system consists of user terminals, servers, a database, and an emotion analysis engine.

[1220] Hardware and software used

[1221] User terminal: A device such as a PC or smartphone used by the user to enter their inquiry.

[1222] Server: A server runs multiple software programs. Details are provided below.

[1223] Natural language processing models: "spaCy" and "NLTK" are used for morphological and syntax analysis.

[1224] Sentiment Analysis Engine: IBM Watson Tone Analyzer is used to analyze user emotions.

[1225] FAQ database search engine: Use "Elasticsearch" for searching.

[1226] Machine learning models: TensorFlow and PyTorch are used for data training and prediction.

[1227] Program processing flow

[1228] The user enters an inquiry.

[1229] Users enter inquiries into the system using their own devices (e.g., PCs or smartphones). For example, a user might enter a question such as, "Please tell me the conditions for applying maintenance."

[1230] The device sends the query to the server.

[1231] The terminal sends the entered query content to the server using an HTTP POST request. The request contains the query content and is sent to the specified API endpoint (e.g., / api / v1 / inquiries).

[1232] The server parses the query.

[1233] The server passes the received request to a natural language processing (NLP) model for analysis. For example, it uses the Python library "spaCy" to perform morphological and syntax analysis, extracting key keywords and context. To understand the intent of the query, the keywords are categorized into specific groups.

[1234] The server recognizes emotions using an emotion engine.

[1235] The server passes the analysis results to a sentiment analysis engine (e.g., IBM Watson Tone Analyzer) to recognize the user's emotions. For example, if an inquiry frequently uses language that includes anxiety or anger, the server will identify those emotions.

[1236] The server searches the database.

[1237] The server searches the FAQ database based on the analysis results. It uses Elasticsearch to quickly find answers related to the inquiry. For example, if a user inquires about maintenance services, the server retrieves information from the database such as "Maintenance services are only applicable within the product warranty period."

[1238] The server generates responses based on emotions.

[1239] The server uses a generative AI model (e.g., GPT-3) based on the sentiment analysis results to generate responses in an appropriate tone for the user. For example, if the user is feeling anxious, it might generate a response such as, "To alleviate your concerns, please rest assured that maintenance services are only applicable within the product warranty period."

[1240] The server sends the response to the terminal.

[1241] The server sends the generated response to the device and displays it to the user. The response is displayed on the user's device in a chat window or as a notification.

[1242] The server stores queries and answers in a database.

[1243] The server stores user inquiries and generated responses in a database. This information can be reused for future inquiries. MySQL and PostgreSQL are commonly used databases.

[1244] The server trains the machine learning model.

[1245] The server uses the accumulated data to train machine learning models. For example, it uses tools like TensorFlow or PyTorch to improve the model's performance and enable it to respond quickly and accurately to new queries.

[1246] Automated operation

[1247] The server automatically responds to specific user requests (e.g., shipping a replacement device, signing a new contract, changing models, etc.). For example, if a user requests "ship a replacement device," the server will integrate with a customer management system (e.g., Salesforce) and automatically initiate the process of shipping the replacement device.

[1248] Specific example

[1249] For example, if a user submits an inquiry such as "Please tell me the conditions for maintenance coverage," the server analyzes the inquiry and extracts relevant keywords such as "maintenance," "coverage," and "conditions." The server then uses a sentiment analysis engine to analyze the user's emotions and recognizes that the user is feeling anxious. Next, the server searches its database for an appropriate answer, such as "Maintenance services are only applicable within the product warranty period," and generates a response that includes a more considerate and detailed explanation that takes the user's feelings into account. This response is then sent to the terminal and displayed to the user. This allows the user to obtain the necessary information with peace of mind.

[1250] In this way, the system of the present invention efficiently processes user inquiries, automates appropriate responses and operations, and simultaneously improves operational efficiency and user experience by recognizing user emotions and optimizing responses.

[1251] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1252] Step 1:

[1253] The user enters an inquiry.

[1254] Users enter inquiries into the system using their own devices (PCs, smartphones, etc.). Specifically, users open a web browser or application and enter questions or requests in text format into input fields. For example, a user might enter the prompt message, "Please tell me the conditions for maintenance application."

[1255] Step 2:

[1256] The device sends the query to the server.

[1257] The terminal sends the entered query content to the server as an HTTP POST request. The input data (user's query text) is included in the request body. The request is sent to the specified API endpoint (e.g., / api / v1 / inquiries). The server receives this request and prepares to process the query content.

[1258] Step 3:

[1259] The server parses the query.

[1260] The server passes the received query content to a natural language processing (NLP) model for analysis. The query text is provided as input, and morphological and syntax analysis is performed. Specifically, the Python library "spaCy" is used to extract key keywords and context. The output provides the analyzed keywords and contextual information.

[1261] Step 4:

[1262] The server recognizes emotions using an emotion engine.

[1263] The server passes the parsed query to the sentiment analysis engine to recognize the user's emotions. The parsed query text is used as input, and the sentiment analysis engine (e.g., IBM Watson Tone Analyzer) analyzes the text. The output is data indicating the user's emotions (anger, anxiety, calmness, etc.).

[1264] Step 5:

[1265] The server searches the FAQ database.

[1266] The server searches the FAQ database for relevant answers based on the analysis results. The query keywords and contextual information are used as input, and a search engine (e.g., Elasticsearch) extracts the appropriate answers. The output is the relevant answers retrieved from the database.

[1267] Step 6:

[1268] The server generates responses based on emotions.

[1269] The server generates responses to the user in an appropriate tone based on the searched answers and sentiment analysis results. The inputs used are answers and sentiment data from a database. A generative AI model (e.g., GPT-3) generates text in an appropriate tone based on this input data. The output is a sentiment-sensitive response text. For example, a response such as "Please rest assured that maintenance services are only applicable within the product warranty period" might be generated.

[1270] Step 7:

[1271] The server sends the response to the terminal.

[1272] The server sends the generated response to the device. The generated response text is used as input and returned to the device as an HTTP response. The device receives this response and displays the response to the user. Specifically, it is displayed as a chat window on a web page or as a notification in an application.

[1273] Step 8:

[1274] The server stores queries and answers in a database.

[1275] The server stores user inquiries and generated responses in a database. Inquiry text and response text are used as input and stored in a database (e.g., MySQL or PostgreSQL). This data is accumulated in the database as output. This accumulated data helps improve future query responses.

[1276] Step 9:

[1277] The server trains the machine learning model.

[1278] The server uses accumulated data to train a machine learning model. Past query and response data is used as input, and the model is trained using a machine learning framework (e.g., TensorFlow or PyTorch). The output is a trained model capable of generating highly accurate responses.

[1279] Step 10:

[1280] The server automates the operation.

[1281] The server automatically responds to specific user requests (e.g., shipping a replacement device, signing a new contract, changing models, etc.). The user's request text is used as input, and the server calls the appropriate API to process it. For example, if a user enters "Please ship a replacement device," the server will integrate with the customer management system and automatically initiate the shipping process. The user receives a notification that the requested operation has been completed.

[1282] (Application Example 2)

[1283] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1284] Conventional inquiry handling systems struggle to respond to user inquiries quickly and appropriately, particularly lacking in responses that take emotions into consideration. Therefore, there is a need for improved user experience and operational efficiency. Furthermore, in security services, emotional recognition is especially crucial in handling emergency inquiries, requiring quick and appropriate responses. This often leads to challenges such as a lack of user confidence and delays in responses.

[1285] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[1286] In this invention, the server includes means for recognizing emotions from the content of a user's inquiry, means for adjusting the tone of the response and generating an emotion-appropriate answer based on the recognized emotion, and means for training a machine learning model using data stored in the database to improve responses to new inquiries. This enables quick and appropriate responses that take into account the user's emotions, improving the user experience and operational efficiency.

[1287] "Means for receiving user inquiries" refers to a function that provides an interface for end users to make inquiries to the system via text or voice.

[1288] "Methods for analysis using natural language processing models" refer to functions that convert linguistic information contained in user inquiries into a format that a computer can understand, and then apply a model to analyze the intent and content of the inquiry.

[1289] "Means for searching for relevant answers" refers to a function that finds appropriate answers from a database based on the analyzed query content.

[1290] "Means of providing information to users" refers to functions that display search results in a user-friendly format and provide information to users.

[1291] "Means of recognizing emotions" refers to a function that analyzes the user's inquiry to determine their emotional state, such as anger, anxiety, or calmness.

[1292] "Means for generating emotionally responsive answers" refers to a function that creates more appropriate and friendly responses based on the recognized emotions of the user.

[1293] "Means for storing inquiries and answers in a database" refers to a function that records received inquiries and their responses in a database, which can then be used to handle future inquiries.

[1294] "Methods for training machine learning models" refers to the function of training models that improve the accuracy of automated responses using machine learning algorithms based on a large amount of query and answer data accumulated in a database.

[1295] This invention is a system that efficiently receives user inquiries, provides appropriate answers, and further recognizes user emotions to optimize responses. The system consists of a user terminal, a server, a database, and an emotion engine. Detailed embodiments of this system are described below.

[1296] System Configuration

[1297] Users input language-based queries to the system using devices such as PCs and smartphones. The device sends the entered query content to the server. The server receives this via an API endpoint and parses the query.

[1298] Hardware and software to be used

[1299] Hardware:

[1300] Smartphone or PC (device)

[1301] server

[1302] software:

[1303] Natural Language Processing (NLP) Model: Using the Python transformers library

[1304] Emotion recognition model: Uses the transformers library from Hugging Face.

[1305] Databases: Relational databases such as MySQL and PostgreSQL

[1306] Machine learning models: TensorFlow or PyTorch

[1307] Data processing and data calculation

[1308] Inquiry Analysis

[1309] The server analyzes user inquiries using a natural language processing model. Specifically, it performs morphological and syntax analysis to extract key keywords and their context.

[1310] emotion recognition

[1311] After analyzing the content of the inquiry, an emotion engine is used to recognize the user's emotions. Through text analysis, it is determined whether the user is angry, anxious, or calm.

[1312] Generating an answer

[1313] The server searches for relevant answers based on the analyzed query and the perceived emotion, and adjusts the tone of the response accordingly. For example, if the user is angry, it might include a polite apology.

[1314] Data storage and training of machine learning models

[1315] All inquiries and answers are stored in a database. This data is used to train machine learning models, improving the accuracy of responses to new inquiries.

[1316] Specific example

[1317] Specific inquiry scenarios and responses

[1318] User inquiry: "I heard gunshots nearby."

[1319] System response: "Please stay calm. Close windows and doors and move to a safe place immediately."

[1320] Examples of prompts for a generative AI model:

[1321] "When a user reports hearing gunshots nearby, analyze their emotions and generate a quick response to ensure their safety."

[1322] In this way, the system of the invention can efficiently and quickly process user inquiries and generate appropriate responses. This system optimizes responses to urgent user inquiries, particularly in the field of security services, and provides users with peace of mind.

[1323] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1324] Step 1:

[1325] The user enters their inquiry.

[1326] The user enters their inquiry in text format using a terminal. After input, the inquiry data is sent from the terminal to the server.

[1327] Input: Text data entered by the user on the device.

[1328] Output: Query data sent to the server

[1329] Step 2:

[1330] The server receives the query.

[1331] The server receives query data sent from the terminal via the API endpoint.

[1332] Input: Inquiry data sent from the terminal

[1333] Output: Text data received by the server

[1334] Step 3:

[1335] The server analyzes the query using a natural language processing model.

[1336] The server uses an NLP model (transformers library) to perform morphological and syntax analysis on the query content. It extracts key keywords and context to understand the intent of the query.

[1337] Input: Text data received by the server

[1338] Output: Intent and key keywords of the query

[1339] Step 4:

[1340] The server uses an emotion recognition model to recognize emotions.

[1341] The server uses an emotion recognition model (transformers library) to determine the user's emotions from the analyzed query content. For example, it can determine whether the user is angry, anxious, or calm.

[1342] Input: Intent and main keywords of the inquiry

[1343] Output: User's emotional state

[1344] Step 5:

[1345] The server searches the FAQ database.

[1346] The server searches the database for relevant answers based on the analyzed query content and sentiment state. If a relevant answer is found, it retrieves it.

[1347] Input: Inquiry intent, key keywords, user sentiment

[1348] Output: Related answers

[1349] Step 6:

[1350] The server generates responses based on emotions.

[1351] Based on the emotion recognition results, the server adjusts the tone to be user-friendly and generates an appropriate response. For example, if the user is angry, it will generate a response that includes a polite apology.

[1352] Input: Related answers, user's emotional state

[1353] Output: Emotionally adjusted response

[1354] Step 7:

[1355] The server sends the response to the terminal.

[1356] The server sends the generated response to the device. The device then displays this to the user as a chat window or notification.

[1357] Input: Emotionally adjusted response

[1358] Output: The answer displayed on the user's device.

[1359] Step 8:

[1360] The server stores inquiries and answers in a database.

[1361] The server records the received inquiry and the generated response in a database. This data will be used to improve the accuracy of future inquiry handling.

[1362] Input: Inquiry details, generated response

[1363] Output: Queries and answers stored in the database

[1364] Step 9:

[1365] The server trains the machine learning model.

[1366] The server uses the accumulated data to train a machine learning model. This model improves the response generation algorithm to handle new queries.

[1367] Input: Inquiries and answers stored in the database

[1368] Output: Trained machine learning model

[1369] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1370] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1371] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[1372] [Fourth Embodiment]

[1373] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1374] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1375] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1376] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[1377] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[1378] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[1379] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[1380] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[1381] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[1382] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1383] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1384] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[1385] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1386] This invention is a system that efficiently receives and analyzes user inquiries and provides appropriate responses. This achieves improved operational efficiency and cost reduction. Specific embodiments for implementing this invention are described below.

[1387] System Configuration

[1388] The system consists of user terminals, a server, and a database. Users make queries using the terminals, the server receives and analyzes them, searches for relevant answers, and provides them to the user.

[1389] The user enters an inquiry.

[1390] Users use their devices (e.g., PCs or smartphones) to input language-based inquiries into the system. This allows them to ask questions about seasonal products and services, how to set up an account, and more.

[1391] The device sends the query to the server.

[1392] The terminal sends the entered query content to the server. The server receives this via the API endpoint and begins parsing the query.

[1393] The server parses the query.

[1394] The server analyzes the query using a natural language processing (NLP) model. For example, the NLP model extracts key keywords and context from the text to understand the intent of the query. Techniques such as morphological analysis and syntax analysis are used in this step.

[1395] The server searches for relevant answers.

[1396] The server searches the database for relevant answers based on the analyzed query. The database contains pre-configured FAQs and manuals, and the search engine uses this information to return the most suitable answer.

[1397] The server sends the response to the terminal.

[1398] The server sends the search results to the device. The device displays this to the user as a chat window or notification. This allows the user to quickly obtain the appropriate information.

[1399] The server stores queries and answers in a database.

[1400] The server stores user inquiries and their corresponding responses in a database. This data is used to improve the accuracy of automated responses to future inquiries.

[1401] The server trains the machine learning model.

[1402] The server uses accumulated data to train a machine learning model. This model improves the response generation algorithm to handle new queries, resulting in more accurate automated responses.

[1403] Specific example

[1404] For example, if a user submits a query such as "Please tell me the conditions for applying maintenance," the server analyzes the query and extracts the relevant keywords "maintenance," "application," and "conditions." The server then searches its database for the appropriate answer, such as "Maintenance services are only applicable within the product warranty period," and sends it to the user's terminal. This allows the user to quickly obtain the necessary information.

[1405] Automated operation

[1406] Furthermore, if a user requests specific actions such as sending a replacement device, signing a new contract, or changing models, the server will call the appropriate API to automate these operations. For example, if a user requests "Please send a replacement device," the server will integrate with the customer management system and automatically handle the process of sending the replacement device.

[1407] In this way, the system of the present invention efficiently processes user inquiries and automates appropriate responses and operations, thereby achieving improved operational efficiency and cost reduction.

[1408] The following describes the processing flow.

[1409] Step 1:

[1410] The user enters the inquiry. The user uses a device (e.g., a PC or smartphone) to input questions or requests to the system in text format.

[1411] Step 2:

[1412] The device sends the query to the server. The device sends the entered query content to the server via the API endpoint.

[1413] Step 3:

[1414] The server receives the query. The server retrieves the text data received from the API endpoint.

[1415] Step 4:

[1416] The server analyzes the query. The server uses a natural language processing (NLP) model to analyze the query content. Specifically, it uses morphological analysis and syntax analysis to extract key keywords and context.

[1417] Step 5:

[1418] The server extracts keywords. It extracts important keywords, mainly nouns and verbs, from the query text.

[1419] Step 6:

[1420] The server searches the FAQ database. Based on the extracted keywords, the server searches the FAQ database for relevant answers.

[1421] Step 7:

[1422] The server selects the most appropriate answer. If multiple answer choices exist, the server selects the most relevant answer.

[1423] Step 8:

[1424] The server generates the answer. Based on the selected answer, it generates the final text to be provided to the user.

[1425] Step 9:

[1426] The server sends the response to the device. An API endpoint is used to send the generated response text to the device.

[1427] Step 10:

[1428] The device displays the response to the user. The device displays the received response to the user in a chat window or notification format.

[1429] Step 11:

[1430] The server stores queries and answers in a database. The query content and the corresponding answers are recorded in the database and used for future query processing.

[1431] Step 12:

[1432] The server trains a machine learning model. Using accumulated data, the machine learning model is periodically trained to improve the accuracy of automated responses to inquiries.

[1433] Step 13:

[1434] The user requests a specific action. Specific requests, such as sending a replacement device, signing a new contract, or changing models, are entered in text format.

[1435] Step 14:

[1436] The server processes the user's request. It calls the appropriate API and performs the specific action required for the request (e.g., arranging for a replacement device to be shipped).

[1437] Step 15:

[1438] The server sends the processing results to the terminal. A notification indicating that processing is complete is sent to the terminal.

[1439] Step 16:

[1440] The terminal displays the processing results to the user. The terminal displays the processing results to the user in a chat window or notification format.

[1441] Through the steps described above, the system can efficiently process user inquiries and automate appropriate responses and actions.

[1442] (Example 1)

[1443] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1444] Conventional inquiry handling systems suffered from long response times to user inquiries and low accuracy in providing appropriate answers. Furthermore, the need for numerous manual operations to find appropriate answers prevented sufficient efficiency improvements and cost reductions. The objective of this invention is to solve these problems and provide users with efficient and highly accurate responses.

[1445] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1446] In this invention, the server includes means for receiving inquiries from users, means for analyzing the inquiries using a natural language processing model, means for sending the input inquiries to the server via an API endpoint, means for searching for relevant answers from a pre-configured data storage based on the analyzed inquiries, means for providing the retrieved answers to the user, means for storing the inquiries and answers in the data storage, and means for training a computational model using the data stored in the data storage to improve the response to new inquiries. This enables a rapid and accurate response to user inquiries, efficient business operations, and cost reduction.

[1447] An "inquiry" is the act of a user inputting information they want to know or a problem they want to solve into a system using natural language.

[1448] A "natural language processing model" is a type of artificial intelligence that analyzes text written in natural language, understands its content and meaning, and processes it.

[1449] An "API endpoint" is part of an application programming interface and is a connection point for different software systems to communicate with each other.

[1450] A "server" is a computer system that provides services to other computers via a network.

[1451] "Data storage" refers to devices and systems for storing digital data, and includes databases and cloud storage.

[1452] "Analysis" is the act of breaking down input data into its components, understanding its content, and identifying its meaning and intent.

[1453] An "answer" refers to the information or solution provided in response to a user's inquiry.

[1454] A "machine learning model" is an algorithm that learns from data and performs tasks such as prediction and classification.

[1455] A "computational model" is a mathematical model used to perform a specific task through numerical calculations and data processing.

[1456] A "programming interface" is a set of definitions and protocols that enable different software programs to work together smoothly.

[1457] "Accumulation" is the act of collecting data and saving it.

[1458] This invention is a system that efficiently receives and analyzes user inquiries and provides appropriate responses. This achieves improved operational efficiency and cost reduction. Specific embodiments of this invention are described below.

[1459] System Configuration

[1460] This system consists of user terminals, a server, and data storage. Users use the terminals to make inquiries, the server receives and analyzes them, searches for relevant answers, and provides them to the user.

[1461] The user enters an inquiry.

[1462] Users use their devices (such as PCs or smartphones) to input natural language queries into the system. For example, they might ask, "I'd like to know about the product's warranty period."

[1463] The device sends the query to the server.

[1464] The terminal sends the entered query content as an HTTP POST request to the server's API endpoint. The server receives this and begins parsing the query.

[1465] The server parses the query.

[1466] The server uses an NLP (Natural Language Processing) model to analyze the query. This analysis utilizes NLP libraries such as spaCy and Hugging Face in Python. For example, morphological analysis is used to extract the keyword "product warranty period."

[1467] The server searches for relevant answers.

[1468] The server searches for relevant answers in data storage based on the extracted keywords. Data storage may include SQL Server or Elasticsearch. The server finds the best answer, retrieving information such as "The product warranty period is one year."

[1469] The server sends the response to the terminal.

[1470] The server sends the retrieved response to the device in JSON format. The device then displays this response to the user in a chat window or notification, allowing the user to quickly obtain the necessary information.

[1471] The server stores queries and answers in data storage.

[1472] The server records the query content and the corresponding response in a separate data storage. This data will be used to improve future query responses.

[1473] The server trains the machine learning model.

[1474] The server uses data stored in data storage to train machine learning models. For example, it uses Python's Scikit-learn or TensorFlow to improve the accuracy of the models. This enables sophisticated automated responses to new queries.

[1475] Automated operation

[1476] Furthermore, this system can handle specific user requests such as shipping replacement equipment, signing up for new contracts, or changing models. For example, if a user requests "Please ship a replacement unit," the server will call the appropriate programming interface and automatically execute the operation. This programming interface integrates with customer management systems, logistics systems, and other systems to automate necessary procedures.

[1477] Specific example

[1478] For example, if a user asks, "What are the conditions for maintenance?", the server analyzes the inquiry and extracts the keywords "maintenance," "applicability," and "conditions." The server then searches its database for the answer, "Maintenance services are applicable only within the product warranty period," and sends it to the user's terminal. This allows the user to quickly obtain the necessary information.

[1479] Thus, the present invention enables prompt and accurate responses to user inquiries, thereby improving operational efficiency and reducing costs.

[1480] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1481] Step 1:

[1482] The user enters an inquiry.

[1483] Users enter their inquiries in natural language into a text form using a device such as a PC or smartphone. Specifically, they might type "Please tell me about the product's warranty period" and click the submit button. The entered text is temporarily stored on the user's device.

[1484] Step 2:

[1485] The device sends the query to the server.

[1486] When the user clicks the submit button, the device sends the entered query content to the server as an HTTP POST request. For example, it sends a request with data in JSON format like the following:

[1487] json

[1488] {

[1489] "query": "Please tell me about the product's warranty period."

[1490] }

[1491] The server receives this request and temporarily holds the query details in memory.

[1492] Step 3:

[1493] The server parses the query.

[1494] The server inputs the received query into a natural language processing (NLP) model and begins analysis. Specifically, it uses a Python NLP library (e.g., spaCy) to perform morphological analysis and extract keywords and phrases from the text. For example, the keyword "product warranty period" might be extracted. The input is the received query, and the output is the extracted keywords.

[1495] Step 4:

[1496] The server searches for relevant answers.

[1497] The server searches for information in data storage based on keywords extracted from the analyzed query. It uses SQL Server, Elasticsearch, etc., to retrieve relevant answers. Specifically, it executes SQL queries like the following:

[1498] sql

[1499] SELECT answer FROM faq WHERE question LIKE '%warranty period%'

[1500] The input is the extracted keywords, and the output is the appropriate answer. This query yields the answer, "The product warranty period is one year."

[1501] Step 5:

[1502] The server sends the response to the terminal.

[1503] The server sends the retrieved response to the terminal as a JSON response. Specifically, it sends it in the following format:

[1504] json

[1505] {

[1506] "Answer": "The product warranty period is one year."

[1507] }

[1508] The terminal receives this data and displays it in an inquiry form or a dedicated window. The output is displayed on the user's terminal.

[1509] Step 6:

[1510] The user receives the response.

[1511] The user checks the answer displayed on their device. Specifically, they can see a message such as "The product warranty period is one year" and get the appropriate answer to their inquiry.

[1512] Step 7:

[1513] The server stores queries and answers in a database.

[1514] The server stores the query details and corresponding answers in a database. This data is used to improve future responses. Specifically, the data is stored as follows:

[1515] sql

[1516] INSERT INTO logs (query, answer) VALUES ('Please tell me about the product's warranty period', 'The product's warranty period is 1 year')

[1517] The input consists of the inquiry content and the response, and the output is the storage of the data in the database.

[1518] Step 8:

[1519] The server trains the machine learning model.

[1520] The server trains a machine learning model using historical query data stored in the database. Specifically, it uses Python's Scikit-learn and TensorFlow to improve the model's accuracy. This model is designed to increase the accuracy of generating answers to new queries. The input is historical query data, and the output is the updated machine learning model.

[1521] (Application Example 1)

[1522] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1523] In modern brick-and-mortar stores, it is necessary to respond quickly and accurately to a wide range of customer inquiries, but relying on manual responses makes efficient and cost-effective operation difficult. Furthermore, if customers cannot obtain information about products and services immediately, their willingness to purchase may decrease. In addition, manual processing is time-consuming when performing specific actions based on user inquiries, leading to decreased customer satisfaction. This invention aims to solve these problems and achieve efficiency and automation in customer service in physical stores.

[1524] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1525] In this invention, the server includes means for receiving inquiries from users, means for analyzing inquiries using a natural language processing model, and means for searching for relevant answers from a pre-configured database based on the analyzed inquiries. This enables the rapid provision of answers to customer inquiries. Furthermore, by including means for customer interaction in the transport device, it becomes possible to automate tasks such as guiding customers to product locations and assisting with purchase procedures. This streamlines customer service in physical stores and improves customer satisfaction.

[1526] definition statement

[1527] "Means of receiving user inquiries" refers to the function of receiving any inquiries that users input into the system via their terminals.

[1528] "Methods for analysis using natural language processing models" refers to a function that analyzes input queries using natural language processing technology, extracts key keywords and context, and understands the user's intent.

[1529] "Means of searching for relevant answers from a pre-configured database" refers to a function that searches for the most suitable answer from a pre-configured database based on the analyzed query content.

[1530] "Means of providing information to the user" refers to functions that notify or display the searched answers on the user's device.

[1531] "Means for storing inquiries and answers in a database" refers to a function that saves user inquiries and the corresponding answers in a database.

[1532] "Means of training machine learning models to improve responses to new inquiries" refers to a function that uses accumulated data to train machine learning models and improve the accuracy and effectiveness of responses to future inquiries.

[1533] "Means by which a transport device responds to customers" refers to a function in which a transport device or robot directly responds to customers based on the analyzed inquiry content.

[1534] "Means for automatically generating answers to new inquiries" refers to a function that uses machine learning models to automatically generate appropriate answers.

[1535] "Means of calling an appropriate application program interface to perform processing" refers to a function that automatically executes processes such as shipping replacement equipment, signing new contracts, and changing models by calling an appropriate application program interface in response to a user's operation request.

[1536] Modes for carrying out the invention

[1537] This invention is a system that efficiently receives and analyzes user inquiries and provides appropriate responses. This system is particularly aimed at improving the efficiency of customer service in physical stores. Specific embodiments for implementing this invention are described below.

[1538] System Configuration

[1539] The system consists of user terminals, a server, and a database. Users make queries using the terminals, the server receives and analyzes them, searches for relevant answers, and provides them to the user. Furthermore, a transport device responds to customers as needed.

[1540] The user enters an inquiry.

[1541] Users use a device (e.g., smartphone, tablet, or transport device) to input language-based inquiries into the system. This allows them to ask questions such as the location of products or detailed information about services.

[1542] The device sends the query to the server.

[1543] The terminal sends the entered query content to the server. The server receives this via the API endpoint and begins parsing the query.

[1544] The server parses the query.

[1545] The server analyzes the query content using a natural language processing (NLP) model. For example, the NLP model extracts key keywords and context from the text to understand the intent of the query. Techniques such as morphological analysis and syntax analysis are used in this step. Libraries such as spaCy are examples of NLP models that can be used.

[1546] The server searches for relevant answers.

[1547] The server searches the database for relevant answers based on the analyzed query. The database contains pre-configured frequently asked questions (FAQs) and store manuals, and the search engine uses this information to return the most appropriate answer. SQLite and MySQL are used as database management systems (DBMS).

[1548] The server sends the response to the terminal.

[1549] The server sends the search results to the device. The device displays this to the user as a chat window or notification. This allows the user to quickly obtain the appropriate information.

[1550] The server stores queries and answers in a database.

[1551] The server stores user inquiries and their corresponding responses in a database. This data is used to improve the accuracy of automated responses to future inquiries.

[1552] The server trains the machine learning model.

[1553] The server trains a machine learning model using the accumulated data. This model improves the response generation algorithm to handle new queries, resulting in more accurate automated responses. Machine learning libraries such as TensorFlow and PyTorch are used.

[1554] The conveying equipment responds to the customer.

[1555] A transport device (e.g., a robot) can guide customers within a store based on the analyzed inquiry. For example, if a customer asks, "Where is this product?", the device can guide the customer to the location where the product is located based on the analyzed results.

[1556] Specific example

[1557] For example, if a user submits a request such as "Please tell me which items are on sale," the server analyzes the request and extracts relevant keywords like "sale" and "item." The server then searches its database for an answer such as "The current sale items are A, B, and C" and sends it to the user's terminal. This allows the user to quickly obtain the information they need.

[1558] Example of a prompt

[1559] "Please analyze the following user inquiry and provide an appropriate answer: 'What items are on sale?'"

[1560] The above describes the embodiments for carrying out the present invention. This system enables efficient customer service in physical stores and improves customer satisfaction.

[1561] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1562] Program processing flow

[1563] Step 1:

[1564] The user enters a language-based inquiry using a terminal. The entered inquiry is captured as text data by the user's terminal's input interface. For example, the user enters the inquiry, "Please tell me what items are on sale."

[1565] Step 2:

[1566] The terminal sends the entered query content to the server. Here, the query data is sent to the server's API endpoint via an HTTP POST request. The input data is the text of the query.

[1567] Step 3:

[1568] The server analyzes the received query using a natural language processing (NLP) model. Specifically, it uses an NLP model (e.g., spaCy) to analyze the text data and extract the main keywords "sale" and "product." In this step, the text of the input data is converted into the main keywords.

[1569] Step 4:

[1570] The server searches for relevant answers from a pre-configured database based on the parsed query. For example, it might use an SQLite database to search for information on "sale items." In this step, it uses the entered keywords to execute a database query that contains the appropriate answers.

[1571] Step 5:

[1572] The server sends the search results to the terminal. Specifically, it sends the answers back to the terminal as an HTTP response. In this step, the answers retrieved from the database are sent to the terminal as output data.

[1573] Step 6:

[1574] The terminal displays the received response to the user. The response content is displayed on the user's terminal screen. For example, it might say, "Current sale items are A, B, and C."

[1575] Step 7:

[1576] The server stores user inquiries and their corresponding answers in a database. This improves the accuracy of automated responses to future inquiries. In this step, the input data (inquiry content) and output data (answer) are stored in the database as a single record.

[1577] Step 8:

[1578] The server trains machine learning models using the accumulated data. Specifically, it uses libraries such as TensorFlow and PyTorch to train the models and improve their accuracy. In this step, the historical data stored in the database is used as input data for the machine learning models, and the model parameters are updated.

[1579] Step 9:

[1580] The transport system responds to the customer based on the analyzed inquiry. For example, a robot guides the customer to the location where the product is located. In this step, the transport system physically operates based on the input data (analyzed inquiry).

[1581] Through the processing steps described above, this system can streamline customer service in physical stores and provide timely and appropriate information. Furthermore, machine learning using accumulated data continuously improves the system's response accuracy.

[1582] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1583] This invention is a system that efficiently receives user inquiries, provides appropriate answers, and optimizes responses by recognizing user emotions. This achieves improved operational efficiency and enhanced user experience. Specific embodiments for implementing this invention are described below.

[1584] System Configuration

[1585] The system consists of a user terminal, a server, a database, and an emotion engine. Users make queries using the terminal, the server receives and analyzes them, searches for relevant answers, and provides them to the user. The emotion engine also recognizes the user's emotions and provides appropriate responses accordingly.

[1586] The user enters an inquiry.

[1587] Users use their devices (e.g., PCs or smartphones) to input language-based inquiries into the system. This allows them to ask questions about product and service details, account setup instructions, and more.

[1588] The device sends the query to the server.

[1589] The terminal sends the entered query content to the server. The server receives this via the API endpoint and begins parsing the query.

[1590] The server parses the query.

[1591] The server uses a natural language processing (NLP) model to analyze the query. Specifically, it uses morphological and syntax analysis to extract key keywords and context to understand the intent of the query.

[1592] The server uses an emotion engine to recognize emotions.

[1593] The server uses an emotion engine to recognize emotions from the user's inquiry. For example, it uses text analysis to determine whether the user is angry, anxious, or calm.

[1594] The server searches the FAQ database.

[1595] The server searches the database for relevant answers based on the analyzed query. The database contains pre-configured FAQs and manuals, and the search engine uses this information to return the most suitable answer.

[1596] The server generates responses based on emotions.

[1597] The server adjusts its responses and tone to be more user-friendly based on the analysis results of the emotion engine. For example, if the user is angry, it will generate a response that includes a polite apology.

[1598] The server sends the response to the terminal.

[1599] The server sends the generated response to the device. The device displays this to the user as a chat window or notification. This allows the user to quickly obtain the appropriate information.

[1600] The server stores queries and answers in a database.

[1601] The server stores user inquiries and their corresponding responses in a database. This data is used to improve the accuracy of automated responses to future inquiries.

[1602] The server trains the machine learning model.

[1603] The server uses accumulated data to train a machine learning model. This model improves the response generation algorithm to handle new queries, resulting in more accurate automated responses.

[1604] Automated operation

[1605] Furthermore, if a user requests specific actions such as sending a replacement device, signing a new contract, or changing models, the server will call the appropriate API to automate these operations. For example, if a user requests "Please send a replacement device," the server will integrate with the customer management system and automatically handle the process of sending the replacement device.

[1606] Specific example

[1607] For example, if a user submits an inquiry such as "Please tell me the conditions for maintenance coverage," the server analyzes the inquiry and extracts relevant keywords such as "maintenance," "coverage," and "conditions." The server then uses an emotion engine to analyze the user's emotions and recognizes that the user is feeling anxious. Next, the server searches its database for an appropriate answer, such as "Maintenance services are only applicable within the product warranty period," and generates a response that includes a more considerate explanation that takes the user's feelings into account. This response is then sent to the terminal and displayed to the user. This allows the user to obtain the necessary information with peace of mind.

[1608] In this way, the system of the present invention efficiently processes user inquiries, automates appropriate responses and operations, and simultaneously improves operational efficiency and user experience by recognizing user emotions and optimizing responses.

[1609] The following describes the processing flow.

[1610] Step 1:

[1611] The user enters the inquiry. The user uses a device (for example, a PC or smartphone) to enter a question in text format, such as "Please tell me the conditions for applying maintenance."

[1612] Step 2:

[1613] The terminal sends the query to the server. The terminal sends the entered query text to the server via the API.

[1614] Step 3:

[1615] The server receives the query. The server receives the query text from the API endpoint and prepares to parse it.

[1616] Step 4:

[1617] The server analyzes the query using a natural language processing (NLP) model. The server performs morphological analysis and extracts key keywords (e.g., maintenance, application, condition).

[1618] Step 5:

[1619] The server uses an emotion engine to recognize the user's emotions. The server analyzes the query text using pattern matching and an emotion dictionary to determine, for example, that the user may be feeling "anxious."

[1620] Step 6:

[1621] The server searches the FAQ database. Based on the extracted keywords, the server searches the FAQ database for relevant answers.

[1622] Step 7:

[1623] The server selects the best answer. The server selects the most appropriate answer from among multiple found responses.

[1624] Step 8:

[1625] The server adjusts the response based on the emotion recognition results. Based on the results of the emotion engine, the server adjusts the response by adding, for example, a gentle tone or polite language.

[1626] Step 9:

[1627] The server generates the response. The server generates the optimal response text and prepares to send it to the terminal in text format.

[1628] Step 10:

[1629] The server sends the response to the device. The server sends the generated response text to the device via the API.

[1630] Step 11:

[1631] The device displays the response to the user. The device displays the received response to the user in a chat window or notification format. For example, it may display a polite explanation such as, "Maintenance service is only applicable within the product warranty period. Please contact us anytime if you have any concerns."

[1632] Step 12:

[1633] The server stores inquiries and answers in a database. The server records the details of the current inquiry and the corresponding answer in the database, and uses this information to handle future inquiries.

[1634] Step 13:

[1635] The server trains a machine learning model. The server periodically trains the machine learning model using accumulated data to improve the accuracy of automated responses.

[1636] Step 14:

[1637] The user requests a specific action. For example, the user might request a specific action through the chat window, such as "Please send a replacement device."

[1638] Step 15:

[1639] The terminal sends an operation request to the server. The terminal sends an operation request to the server via an API.

[1640] Step 16:

[1641] The server processes the request. The server calls the appropriate API (e.g., customer management system) and automatically performs the requested operation (e.g., arranging the shipment of a replacement unit).

[1642] Step 17:

[1643] The server sends the processing results to the terminal. A notification indicating that processing is complete is sent to the terminal.

[1644] Step 18:

[1645] The terminal displays the processing results to the user. The terminal displays the processing results to the user in a chat window or notification format. For example, it might display, "Your replacement device has been shipped. Please wait a while for its arrival."

[1646] Through the steps outlined above, the system efficiently processes user inquiries, automates appropriate responses and operations, and simultaneously improves operational efficiency and user experience by recognizing user emotions and optimizing responses.

[1647] (Example 2)

[1648] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1649] Traditional inquiry response systems often provided uniform answers without considering user emotions, failing to adequately improve user satisfaction. Furthermore, insufficient training data for machine learning models led to decreased accuracy in automated responses to new inquiries. Additionally, a lack of means to automate appropriate processing in response to specific user requests prevented sufficient operational efficiency.

[1650] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1651] In this invention, the server includes means for receiving inquiries from users, means for analyzing inquiries using a natural language processing model, means for searching for relevant answers from a pre-configured database based on the analyzed inquiries, means for providing the retrieved answers to the user, means for storing inquiries and answers in a database, means for training a machine learning model using the data stored in the database to improve responses to new inquiries, means for using an emotion analysis engine that recognizes emotions from the content of the user's inquiry, and means for adjusting the tone and content of the response based on the emotion analysis results. This enables appropriate responses that take into account the user's emotions, thereby increasing satisfaction, improving the accuracy of the machine learning model, and enabling quick and accurate responses to new inquiries. Furthermore, automated processing of specific operation requests can be realized, thus improving operational efficiency.

[1652] A "user" is an individual or organization that uses the system to make an inquiry.

[1653] An "inquiry" refers to a question or request that a user makes to a system.

[1654] A "terminal" refers to an electronic device (such as a PC or smartphone) that a user uses to enter an inquiry.

[1655] A "server" is a computer system that receives user inquiries, analyzes them, and provides appropriate answers.

[1656] A "natural language processing model" is an algorithm or software used to analyze, understand, and generate responses to human language.

[1657] "Analysis" is the process of examining the content of an inquiry and extracting its intent and meaning.

[1658] A "database" is a system for systematically storing inquiries, their answers, and other related information.

[1659] A "sentiment analysis engine" is software or an algorithm that recognizes emotions from user input text and provides the results.

[1660] A "machine learning model" is a collection of algorithms that learn from large amounts of data, extract patterns and knowledge, and then make predictions and classifications based on new data.

[1661] "Searching" is the process of finding specific information from a database.

[1662] "Tone" refers to the way responses are expressed and the writing style, which is adjusted to match the user's emotions.

[1663] An "API" is an interface that allows different software systems to communicate with each other.

[1664] "Automated processing" refers to operations or procedures that a system performs automatically with minimal human intervention.

[1665] This invention provides a system that efficiently receives user inquiries and provides appropriate responses, and further optimizes responses by recognizing user emotions. This achieves improved operational efficiency and enhanced user experience.

[1666] System Configuration

[1667] The system consists of user terminals, servers, a database, and an emotion analysis engine.

[1668] Hardware and software used

[1669] User terminal: A device such as a PC or smartphone used by the user to enter their inquiry.

[1670] Server: A server runs multiple software programs. Details are provided below.

[1671] Natural language processing models: "spaCy" and "NLTK" are used for morphological and syntax analysis.

[1672] Sentiment Analysis Engine: IBM Watson Tone Analyzer is used to analyze user emotions.

[1673] FAQ database search engine: Use "Elasticsearch" for searching.

[1674] Machine learning models: TensorFlow and PyTorch are used for data training and prediction.

[1675] Program processing flow

[1676] The user enters an inquiry.

[1677] Users enter inquiries into the system using their own devices (e.g., PCs or smartphones). For example, a user might enter a question such as, "Please tell me the conditions for applying maintenance."

[1678] The device sends the query to the server.

[1679] The terminal sends the entered query content to the server using an HTTP POST request. The request contains the query content and is sent to the specified API endpoint (e.g., / api / v1 / inquiries).

[1680] The server parses the query.

[1681] The server passes the received request to a natural language processing (NLP) model for analysis. For example, it uses the Python library "spaCy" to perform morphological and syntax analysis, extracting key keywords and context. To understand the intent of the query, the keywords are categorized into specific groups.

[1682] The server recognizes emotions using an emotion engine.

[1683] The server passes the analysis results to a sentiment analysis engine (e.g., IBM Watson Tone Analyzer) to recognize the user's emotions. For example, if an inquiry frequently uses language that includes anxiety or anger, the server will identify those emotions.

[1684] The server searches the database.

[1685] The server searches the FAQ database based on the analysis results. It uses Elasticsearch to quickly find answers related to the inquiry. For example, if a user inquires about maintenance services, the server retrieves information from the database such as "Maintenance services are only applicable within the product warranty period."

[1686] The server generates responses based on emotions.

[1687] The server uses a generative AI model (e.g., GPT-3) based on the sentiment analysis results to generate responses in an appropriate tone for the user. For example, if the user is feeling anxious, it might generate a response such as, "To alleviate your concerns, please rest assured that maintenance services are only applicable within the product warranty period."

[1688] The server sends the response to the terminal.

[1689] The server sends the generated response to the device and displays it to the user. The response is displayed on the user's device in a chat window or as a notification.

[1690] The server stores queries and answers in a database.

[1691] The server stores user inquiries and generated responses in a database. This information can be reused for future inquiries. MySQL and PostgreSQL are commonly used databases.

[1692] The server trains the machine learning model.

[1693] The server uses the accumulated data to train machine learning models. For example, it uses tools like TensorFlow or PyTorch to improve the model's performance and enable it to respond quickly and accurately to new queries.

[1694] Automated operation

[1695] The server automatically responds to specific user requests (e.g., shipping a replacement device, signing a new contract, changing models, etc.). For example, if a user requests "ship a replacement device," the server will integrate with a customer management system (e.g., Salesforce) and automatically initiate the process of shipping the replacement device.

[1696] Specific example

[1697] For example, if a user submits an inquiry such as "Please tell me the conditions for maintenance coverage," the server analyzes the inquiry and extracts relevant keywords such as "maintenance," "coverage," and "conditions." The server then uses a sentiment analysis engine to analyze the user's emotions and recognizes that the user is feeling anxious. Next, the server searches its database for an appropriate answer, such as "Maintenance services are only applicable within the product warranty period," and generates a response that includes a more considerate and detailed explanation that takes the user's feelings into account. This response is then sent to the terminal and displayed to the user. This allows the user to obtain the necessary information with peace of mind.

[1698] In this way, the system of the present invention efficiently processes user inquiries, automates appropriate responses and operations, and simultaneously improves operational efficiency and user experience by recognizing user emotions and optimizing responses.

[1699] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1700] Step 1:

[1701] The user enters an inquiry.

[1702] Users enter inquiries into the system using their own devices (PCs, smartphones, etc.). Specifically, users open a web browser or application and enter questions or requests in text format into input fields. For example, a user might enter the prompt message, "Please tell me the conditions for maintenance application."

[1703] Step 2:

[1704] The device sends the query to the server.

[1705] The terminal sends the entered query content to the server as an HTTP POST request. The input data (user's query text) is included in the request body. The request is sent to the specified API endpoint (e.g., / api / v1 / inquiries). The server receives this request and prepares to process the query content.

[1706] Step 3:

[1707] The server parses the query.

[1708] The server passes the received query content to a natural language processing (NLP) model for analysis. The query text is provided as input, and morphological and syntax analysis is performed. Specifically, the Python library "spaCy" is used to extract key keywords and context. The output provides the analyzed keywords and contextual information.

[1709] Step 4:

[1710] The server recognizes emotions using an emotion engine.

[1711] The server passes the parsed query to the sentiment analysis engine to recognize the user's emotions. The parsed query text is used as input, and the sentiment analysis engine (e.g., IBM Watson Tone Analyzer) analyzes the text. The output is data indicating the user's emotions (anger, anxiety, calmness, etc.).

[1712] Step 5:

[1713] The server searches the FAQ database.

[1714] The server searches the FAQ database for relevant answers based on the analysis results. The query keywords and contextual information are used as input, and a search engine (e.g., Elasticsearch) extracts the appropriate answers. The output is the relevant answers retrieved from the database.

[1715] Step 6:

[1716] The server generates responses based on emotions.

[1717] The server generates responses to the user in an appropriate tone based on the searched answers and sentiment analysis results. The inputs used are answers and sentiment data from a database. A generative AI model (e.g., GPT-3) generates text in an appropriate tone based on this input data. The output is a sentiment-sensitive response text. For example, a response such as "Please rest assured that maintenance services are only applicable within the product warranty period" might be generated.

[1718] Step 7:

[1719] The server sends the response to the terminal.

[1720] The server sends the generated response to the device. The generated response text is used as input and returned to the device as an HTTP response. The device receives this response and displays the response to the user. Specifically, it is displayed as a chat window on a web page or as a notification in an application.

[1721] Step 8:

[1722] The server stores queries and answers in a database.

[1723] The server stores user inquiries and generated responses in a database. Inquiry text and response text are used as input and stored in a database (e.g., MySQL or PostgreSQL). This data is accumulated in the database as output. This accumulated data helps improve future query responses.

[1724] Step 9:

[1725] The server trains the machine learning model.

[1726] The server uses accumulated data to train a machine learning model. Past query and response data is used as input, and the model is trained using a machine learning framework (e.g., TensorFlow or PyTorch). The output is a trained model capable of generating highly accurate responses.

[1727] Step 10:

[1728] The server automates the operation.

[1729] The server automatically responds to specific user requests (e.g., shipping a replacement device, signing a new contract, changing models, etc.). The user's request text is used as input, and the server calls the appropriate API to process it. For example, if a user enters "Please ship a replacement device," the server will integrate with the customer management system and automatically initiate the shipping process. The user receives a notification that the requested operation has been completed.

[1730] (Application Example 2)

[1731] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1732] Conventional inquiry handling systems struggle to respond to user inquiries quickly and appropriately, particularly lacking in responses that take emotions into consideration. Therefore, there is a need for improved user experience and operational efficiency. Furthermore, in security services, emotional recognition is especially crucial in handling emergency inquiries, requiring quick and appropriate responses. This often leads to challenges such as a lack of user confidence and delays in responses.

[1733] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[1734] In this invention, the server includes means for recognizing emotions from the content of a user's inquiry, means for adjusting the tone of the response and generating an emotion-appropriate answer based on the recognized emotion, and means for training a machine learning model using data stored in the database to improve responses to new inquiries. This enables quick and appropriate responses that take into account the user's emotions, improving the user experience and operational efficiency.

[1735] "Means for receiving user inquiries" refers to a function that provides an interface for end users to make inquiries to the system via text or voice.

[1736] "Methods for analysis using natural language processing models" refer to functions that convert linguistic information contained in user inquiries into a format that a computer can understand, and then apply a model to analyze the intent and content of the inquiry.

[1737] "Means for searching for relevant answers" refers to a function that finds appropriate answers from a database based on the analyzed query content.

[1738] "Means of providing information to users" refers to functions that display search results in a user-friendly format and provide information to users.

[1739] "Means of recognizing emotions" refers to a function that analyzes the user's inquiry to determine their emotional state, such as anger, anxiety, or calmness.

[1740] "Means for generating emotionally responsive answers" refers to a function that creates more appropriate and friendly responses based on the recognized emotions of the user.

[1741] "Means for storing inquiries and answers in a database" refers to a function that records received inquiries and their responses in a database, which can then be used to handle future inquiries.

[1742] "Methods for training machine learning models" refers to the function of training models that improve the accuracy of automated responses using machine learning algorithms based on a large amount of query and answer data accumulated in a database.

[1743] This invention is a system that efficiently receives user inquiries, provides appropriate answers, and further recognizes user emotions to optimize responses. The system consists of a user terminal, a server, a database, and an emotion engine. Detailed embodiments of this system are described below.

[1744] System Configuration

[1745] Users input language-based queries to the system using devices such as PCs and smartphones. The device sends the entered query content to the server. The server receives this via an API endpoint and parses the query.

[1746] Hardware and software to be used

[1747] Hardware:

[1748] Smartphone or PC (device)

[1749] server

[1750] software:

[1751] Natural Language Processing (NLP) Model: Using the Python transformers library

[1752] Emotion recognition model: Uses the transformers library from Hugging Face.

[1753] Databases: Relational databases such as MySQL and PostgreSQL

[1754] Machine learning models: TensorFlow or PyTorch

[1755] Data processing and data calculation

[1756] Inquiry Analysis

[1757] The server analyzes user inquiries using a natural language processing model. Specifically, it performs morphological and syntax analysis to extract key keywords and their context.

[1758] emotion recognition

[1759] After analyzing the content of the inquiry, an emotion engine is used to recognize the user's emotions. Through text analysis, it is determined whether the user is angry, anxious, or calm.

[1760] Generating an answer

[1761] The server searches for relevant answers based on the analyzed query and the perceived emotion, and adjusts the tone of the response accordingly. For example, if the user is angry, it might include a polite apology.

[1762] Data storage and training of machine learning models

[1763] All inquiries and answers are stored in a database. This data is used to train machine learning models, improving the accuracy of responses to new inquiries.

[1764] Specific example

[1765] Specific inquiry scenarios and responses

[1766] User inquiry: "I heard gunshots nearby."

[1767] System response: "Please stay calm. Close windows and doors and move to a safe place immediately."

[1768] Examples of prompts for a generative AI model:

[1769] "When a user reports hearing gunshots nearby, analyze their emotions and generate a quick response to ensure their safety."

[1770] In this way, the system of the invention can efficiently and quickly process user inquiries and generate appropriate responses. This system optimizes responses to urgent user inquiries, particularly in the field of security services, and provides users with peace of mind.

[1771] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1772] Step 1:

[1773] The user enters their inquiry.

[1774] The user enters their inquiry in text format using a terminal. After input, the inquiry data is sent from the terminal to the server.

[1775] Input: Text data entered by the user on the device.

[1776] Output: Query data sent to the server

[1777] Step 2:

[1778] The server receives the query.

[1779] The server receives query data sent from the terminal via the API endpoint.

[1780] Input: Inquiry data sent from the terminal

[1781] Output: Text data received by the server

[1782] Step 3:

[1783] The server analyzes the query using a natural language processing model.

[1784] The server uses an NLP model (transformers library) to perform morphological and syntax analysis on the query content. It extracts key keywords and context to understand the intent of the query.

[1785] Input: Text data received by the server

[1786] Output: Intent and key keywords of the query

[1787] Step 4:

[1788] The server uses an emotion recognition model to recognize emotions.

[1789] The server uses an emotion recognition model (transformers library) to determine the user's emotions from the analyzed query content. For example, it can determine whether the user is angry, anxious, or calm.

[1790] Input: Intent and main keywords of the inquiry

[1791] Output: User's emotional state

[1792] Step 5:

[1793] The server searches the FAQ database.

[1794] The server searches the database for relevant answers based on the analyzed query content and sentiment state. If a relevant answer is found, it retrieves it.

[1795] Input: Inquiry intent, key keywords, user sentiment

[1796] Output: Related answers

[1797] Step 6:

[1798] The server generates responses based on emotions.

[1799] Based on the emotion recognition results, the server adjusts the tone to be user-friendly and generates an appropriate response. For example, if the user is angry, it will generate a response that includes a polite apology.

[1800] Input: Related answers, user's emotional state

[1801] Output: Emotionally adjusted response

[1802] Step 7:

[1803] The server sends the response to the terminal.

[1804] The server sends the generated response to the device. The device then displays this to the user as a chat window or notification.

[1805] Input: Emotionally adjusted response

[1806] Output: The answer displayed on the user's device.

[1807] Step 8:

[1808] The server stores inquiries and answers in a database.

[1809] The server records the received inquiry and the generated response in a database. This data will be used to improve the accuracy of future inquiry handling.

[1810] Input: Inquiry details, generated response

[1811] Output: Queries and answers stored in the database

[1812] Step 9:

[1813] The server trains the machine learning model.

[1814] The server uses the accumulated data to train a machine learning model. This model improves the response generation algorithm to handle new queries.

[1815] Input: Inquiries and answers stored in the database

[1816] Output: Trained machine learning model

[1817] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1818] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1819] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[1820] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1821] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[1822] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[1823] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[1824] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[1825] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[1826] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[1827] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[1828] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[1829] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[1830] 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.

[1831] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[1832] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[1833] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of 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). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[1834] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[1835] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[1836] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[1837] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted as being incorporated by reference.

[1838] The following is further disclosed regarding the embodiments described above.

[1839] (Claim 1)

[1840] A means of receiving inquiries from users,

[1841] A means for analyzing the aforementioned query using a natural language processing model,

[1842] A means for searching for relevant answers from a pre-configured database based on the analyzed query,

[1843] Means for providing the searched answers to the user,

[1844] A means for storing the aforementioned inquiry and the aforementioned answer in a database,

[1845] A system that includes means for training a machine learning model using data stored in the aforementioned database to improve responses to new inquiries.

[1846] (Claim 2)

[1847] The system according to claim 1, further comprising means for the machine learning model to automatically generate answers to new inquiries.

[1848] (Claim 3)

[1849] The system according to claim 1, further comprising means by which a user requests specific operations such as sending a replacement device, signing a new contract, or changing models, and a server calls an appropriate API to perform the operation.

[1850] "Example 1"

[1851] (Claim 1)

[1852] A means of receiving inquiries from users,

[1853] A means for analyzing the aforementioned query using a natural language processing model,

[1854] A means of sending the entered query to the server via the API endpoint,

[1855] A means for searching for relevant answers from a pre-configured data storage based on the analyzed query,

[1856] Means for providing the searched answers to the user,

[1857] A means for storing the aforementioned inquiry and the aforementioned answer in data storage,

[1858] A system including means for training a computational model using data stored in the aforementioned data storage to improve responses to new queries.

[1859] (Claim 2)

[1860] The system according to claim 1, further comprising means for the computation model to automatically generate answers to new queries.

[1861] (Claim 3)

[1862] The system according to claim 1, further comprising means by which a user requests a specific operation such as the shipment of a replacement device, a new contract, or a model change, and a server calls an appropriate programming interface to perform the operation.

[1863] "Application Example 1"

[1864] Claims for a New Invention

[1865] (Claim 1)

[1866] A means of receiving inquiries from users,

[1867] A means for analyzing the aforementioned query using a natural language processing model,

[1868] A means for searching for relevant answers from a pre-configured database based on the analyzed query,

[1869] Means for providing the searched answers to the user,

[1870] A means for storing the aforementioned inquiry and the aforementioned answer in a database,

[1871] A means of training a machine learning model using data accumulated in the aforementioned database to improve the response to new inquiries,

[1872] A system including means by which a transport device responds to a customer based on the analyzed inquiry.

[1873] (Claim 2)

[1874] The system according to claim 1, further comprising means for the machine learning model to automatically generate answers to new inquiries.

[1875] (Claim 3)

[1876] The system according to claim 1, further comprising means by which a user requests a specific operation such as sending a replacement device, signing a new contract, or changing models, and a server calls an appropriate application program interface to perform the operation.

[1877] "Example 2 of combining an emotion engine"

[1878] (Claim 1)

[1879] A means of receiving inquiries from users,

[1880] A means for analyzing the aforementioned query using a natural language processing model,

[1881] A means for searching for relevant answers from a pre-configured database based on the analyzed query,

[1882] Means for providing the searched answers to the user,

[1883] A means for storing the aforementioned inquiry and the aforementioned answer in a database,

[1884] A means of training a machine learning model using data accumulated in the aforementioned database to improve the response to new inquiries,

[1885] A method using an emotion analysis engine that recognizes emotions from the content of user inquiries,

[1886] A system including means for adjusting the tone and content of responses based on the aforementioned sentiment analysis results.

[1887] (Claim 2)

[1888] The system according to claim 1, further comprising means for the machine learning model to automatically generate answers to new inquiries.

[1889] (Claim 3)

[1890] The system according to claim 1, further comprising means by which a user requests specific operations such as sending a replacement device, signing a new contract, or changing models, and a server calls an appropriate API to perform the operation.

[1891] "Application example 2 when combining with an emotional engine"

[1892] (Claim 1)

[1893] A means of receiving inquiries from users,

[1894] A means for analyzing the aforementioned query using a natural language processing model,

[1895] A means for searching for relevant answers from a pre-configured database based on the analyzed query,

[1896] Means for providing the searched answers to the user,

[1897] A means of recognizing emotions from the content of user inquiries,

[1898] A means for adjusting the tone of the response and generating an emotionally appropriate response based on the recognized emotion,

[1899] A means for storing the aforementioned inquiry and the aforementioned answer in a database,

[1900] A system that includes means for training a machine learning model using data stored in the aforementioned database to improve responses to new inquiries.

[1901] (Claim 2)

[1902] The system according to claim 1, further comprising means for the machine learning model to automatically generate answers to new inquiries.

[1903] (Claim 3)

[1904] The system according to claim 1, further comprising means by which a user requests specific operations such as sending a replacement device, signing a new contract, or changing models, and a server calls an appropriate API to perform the operation. [Explanation of Symbols]

[1905] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of receiving inquiries from users, A means for analyzing the aforementioned query using a natural language processing model, A means for searching for relevant answers from a pre-configured database based on the analyzed query, Means for providing the searched answers to the user, A means for storing the aforementioned inquiry and the aforementioned answer in a database, A system that includes means for training a machine learning model using data stored in the aforementioned database to improve responses to new inquiries.

2. The system according to claim 1, further comprising means for the machine learning model to automatically generate answers to new inquiries.

3. The system according to claim 1, further comprising means by which a user requests specific operations such as sending a replacement device, signing a new contract, or changing models, and a server calls an appropriate API to perform the operation.

Images