system

Abstract

We provide the system. [Solution] A means for receiving questions entered by the user, A natural language processing means for analyzing received questions and searching for relevant information, A means of obtaining information from a knowledge base based on the analysis results, A means for generating a response to provide the acquired information to the user, Means of providing information to users, A system that includes this.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor and includes 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 as a response 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 conventional knowledge-based system, it has been difficult to immediately provide appropriate information when a user asks questions such as "What is ○○?" or "How should I do it?". Furthermore, when quick response or escalation to a responsible department is required for such questions, notification to the responsible department and execution of actions are complicated and cannot be processed efficiently. Also, when using voice input by a user, it has been difficult to accurately convert the voice into text, lead to an appropriate answer, or execute necessary actions.

Means for Solving the Problems

[0005] The present invention provides a system that includes means for receiving questions entered by a user, natural language processing means for analyzing the received questions and searching for relevant information, means for obtaining information from a knowledge base based on the analysis results, response generation means for providing the obtained information to the user, and means for providing information to the user. Furthermore, by adding action execution means that perform notifications and escalations based on the user's questions, it enables a rapid response to the relevant department. In addition, by including speech recognition means that convert voice input into text, accurate information provision and response from voice input are realized.

[0006] A "user" is the entity that uses the system to input questions.

[0007] A "question" is input data used to inquire about information, procedures, or steps from a system.

[0008] A "means of receiving" refers to the component that allows the system to receive questions from the user.

[0009] "Natural language processing means" refers to technologies that analyze questions entered by users and understand their intent.

[0010] "Analysis results" refer to data that encapsulates the meaning and intent of a user's question, as analyzed using natural language processing techniques.

[0011] A "knowledge base" is a database that contains information such as internal and external company regulations, procedures, steps, guidelines, and training.

[0012] "Means of acquisition" refers to the component that searches for and extracts relevant information from a knowledge base based on the analysis results.

[0013] A "response generation method" is a technology that generates answers to be provided to the user based on acquired information.

[0014] The "means for providing information" is a component for displaying the generated answer to the user or reading it aloud.

[0015] A "notification" is a message for transmitting necessary information and actions based on the user's question to specific departments or persons in charge.

[0016] "Escalation" is a process of transferring a problem to a more highly capable responsible department when the user's question or problem exceeds a specific scope or authority.

[0017] The "action execution means" is a technology for the system to execute specific actions such as notifications and escalations based on the user's question.

[0018] The "voice recognition means" is a technology for converting a question input by the user in voice into text data.

[0019] "Text" is character data generated by voice recognition means or user manual input.

Brief Description of Drawings

[0020] [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.[[ID=3�]] [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.

Mode for Carrying Out the Invention

[0021] 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.

[0022] First, the terms used in the following description will be explained.

[0023] 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).

[0024] 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.

[0025] 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.

[0026] 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).

[0027] 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."

[0028] [First Embodiment]

[0029] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0030] 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.

[0031] 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).

[0032] 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.

[0033] 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.

[0034] 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.

[0035] 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.

[0036] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0037] 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.

[0038] 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.

[0039] 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.

[0040] 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".

[0041] This invention is a system that analyzes questions entered by a user and quickly provides the relevant information. Its specific form is described below.

[0042] User Interface (UI)

[0043] Terminal: Provides an interface for users to query information. This includes text boxes for text input and voice input buttons. In the case of voice input, a speech recognition mechanism is built in to convert speech into text.

[0044] Natural Language Processing (NLP) Engine

[0045] Server: It has a natural language processing engine that analyzes text data received from users and understands their intent. This engine consists of the following technologies:

[0046] Tokenization: Breaking down a text into individual words or phrases.

[0047] Part-of-speech tagging: Identifying the part of speech of each word.

[0048] Semantic analysis: Analyzes the meaning of the entire text to identify the user's intent.

[0049] Knowledge Base

[0050] Server: It has a large database for searching relevant information based on user intent. This database includes internal and external regulations, procedures, steps, guidelines, training, etc.

[0051] Response generation engine

[0052] Server: There is an engine that retrieves information from a knowledge base based on the results analyzed by the NLP engine and generates appropriate responses for the user. The response generation engine has the following functions:

[0053] Template Selection: Choose the appropriate template.

[0054] Information Embedding: Embed the retrieved information into the template.

[0055] Action execution engine

[0056] Server: Includes an engine that takes notification or escalation actions in response to specific inquiries. For example, if a user's inquiry is about unpaid wages, it sends a notification to the HR department.

[0057] Information provision means

[0058] Terminal: Displays response messages sent from the server to the user. These may be displayed as text or read aloud.

[0059] Specific example

[0060] Example 1: Inquiry about leave application procedures

[0061] 1. User: Types the following text: "How do I apply for leave?"

[0062] 2. Terminal: Sends user input to the server.

[0063] 3. Server: The server uses a natural language processing engine to analyze the question and identify information related to "leave application procedures."

[0064] 4. Server: Retrieves relevant information from the knowledge base and generates a response stating, "To submit a leave request, follow these steps: 1. Log in to the HR portal. 2. Click the 'Leave Request' tab. 3. Enter the required information and click the submit button."

[0065] 5. Terminal: Displays the generated response to the user.

[0066] Example 2: Inquiry about unpaid wages

[0067] 1. User: "My salary hasn't been paid, who should I ask about it?" (Voice input)

[0068] 2. Terminal: Converts speech to text and sends it to the server.

[0069] 3. Server: Uses a natural language processing engine to analyze the question and identify the "contact information for inquiries regarding unpaid wages."

[0070] 4. Server: Retrieves HR department contact information from the knowledge base and generates a response stating, "Inquiries regarding unpaid wages will be escalated to the HR department contact."

[0071] 5. Server: Sends an escalation notification to the HR department.

[0072] 6. Terminal: Display the generated response to the user and confirm that the notification has been sent.

[0073] This invention provides a system that, by combining these functions, can instantly provide appropriate information in response to a variety of questions entered by the user and quickly process the necessary actions.

[0074] The following describes the processing flow.

[0075] Detailed program processing flow

[0076] Example 1: Inquiry about leave application procedures

[0077] Step 1:

[0078] The user types the text, "How do I apply for leave?"

[0079] Step 2:

[0080] The device receives user input. If the user provides voice input, it is first converted to text by the speech recognition engine.

[0081] Step 3:

[0082] The terminal sends the received text data to the server.

[0083] Step 4:

[0084] The server's natural language processing engine analyzes the user's text. First, it tokenizes the text, then tags it with parts of speech, and finally performs semantic analysis to determine the user's intent.

[0085] Step 5:

[0086] Based on the analysis results, the server queries the knowledge base for information regarding "leave application procedures."

[0087] Step 6:

[0088] The server retrieves the relevant information from the knowledge base.

[0089] Step 7:

[0090] The server's response generation engine creates a reply message based on the information it has obtained. Specifically, it generates a message that says, "To request leave, follow these steps: 1. Log in to the HR portal. 2. Click the 'Leave Request' tab. 3. Enter the required information and click the submit button."

[0091] Step 8:

[0092] The server sends the generated message to the terminal.

[0093] Step 9:

[0094] The device displays the message to the user. In the case of voice output, the text is converted to speech and read aloud.

[0095] Example 2: Inquiry about unpaid wages

[0096] Step 1:

[0097] The user inputs a voice message saying, "My salary hasn't been paid, who should I ask about it?"

[0098] Step 2:

[0099] The device receives the audio and converts it to text using a speech recognition engine.

[0100] Step 3:

[0101] The terminal sends text data to the server.

[0102] Step 4:

[0103] The server's natural language processing engine analyzes the user's text. It processes the text in the following order: tokenization, part-of-speech tagging, and semantic analysis, identifying the user's intent as "an inquiry about unpaid wages."

[0104] Step 5:

[0105] The server executes queries against the knowledge base based on the analysis results.

[0106] Step 6:

[0107] The server retrieves information on "contact information for inquiries regarding unpaid wages" from its knowledge base. Here, it retrieves information on the HR department contact person.

[0108] Step 7:

[0109] The server's response generation engine creates a message stating, "Inquiries regarding unpaid wages will be escalated to the HR department."

[0110] Step 8:

[0111] The server's action execution engine generates an escalation notification and sends it to the HR department.

[0112] Step 9:

[0113] The server sends the generated message to the terminal.

[0114] Step 10:

[0115] The device displays messages to the user and reads them aloud if necessary.

[0116] These steps enable us to provide immediate and accurate answers to user inquiries and take necessary actions.

[0117] (Example 1)

[0118] 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."

[0119] In today's world, systems that allow users to obtain information quickly and accurately are crucial. However, many systems lack the ability to accurately understand user questions and fail to provide appropriate information. Furthermore, while notifications and escalations are necessary in certain situations, few systems automate these processes. Therefore, there is a need to develop systems that can efficiently respond to user questions and take necessary actions.

[0120] 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.

[0121] In this invention, the server includes means for receiving questions entered by the user, natural language processing means for analyzing the questions and searching for relevant information, means for obtaining information from a knowledge base based on the analysis results, response generation means for providing the obtained information to the user, and action execution means for performing notifications and escalations based on the user's questions. This makes it possible to provide information quickly and accurately in response to a wide range of user questions and to automatically perform necessary actions.

[0122] A "user" is an individual or group that uses the system to input questions and obtain information.

[0123] A "question" is a text or voice message that a user enters to ask the system for information.

[0124] A "terminal" refers to a device (such as a PC, smartphone, or tablet) that a user uses to input questions and receive responses from a server.

[0125] A "server" is a central processing unit that analyzes questions received from users and searches for, retrieves, and provides the necessary information.

[0126] "Natural language processing methods" refer to technologies for analyzing text data received from users and understanding its intent. Specifically, this includes tokenization, part-of-speech tagging, and semantic analysis.

[0127] A "tokenization method" is a technique that breaks down a question into words or phrases.

[0128] A "part-of-speech tagging method" is a technique for identifying the part of speech of each word.

[0129] "Semantic analysis means" refers to technologies that analyze the meaning of an entire text and identify the user's intent.

[0130] A "knowledge base" is a database that stores information to provide appropriate answers to user questions.

[0131] A "response generation method" is a technology that generates answers to users based on information obtained from a knowledge base. Specifically, it includes template selection and information embedding.

[0132] "Template selection method" refers to the technique of selecting the appropriate response template.

[0133] "Information embedding means" refers to a technology that generates a final response by embedding the necessary information into a selected template.

[0134] An "action execution method" is a technology that automatically performs notifications or escalations in response to specific inquiries.

[0135] "Voice recognition means" refers to technology that converts voice input from a user into text.

[0136] A "notification" is a message that informs specific recipients of important information.

[0137] "Escalation" refers to the procedure of transferring a specific problem or inquiry to a senior officer or another department.

[0138] This invention relates to a system that analyzes user-inputted questions and quickly provides relevant information. This system provides information quickly and accurately in response to a wide range of user questions and automatically performs necessary actions.

[0139] User Interface (UI)

[0140] The terminal provides an interface for users to query information. The terminal includes text boxes for text input and voice input buttons, providing means for users to input questions. In the case of voice input, speech recognition technology is used to convert speech into text. Specifically, devices such as PCs, smartphones, and tablets are used.

[0141] Natural Language Processing (NLP) Engine

[0142] The server has a natural language processing engine that analyzes text data received from users and understands their intent. This engine uses tools such as Google® Cloud Natural Language API and Amazon Comprehend to perform the following processes:

[0143] Tokenization: Breaking down a question into words or phrases.

[0144] Part-of-speech tagging: Identifying the part of speech of each word.

[0145] Semantic analysis: Analyzes the meaning of the entire question to identify the user's intent.

[0146] Knowledge Base

[0147] The server has a database for searching for relevant information based on the user's intent. This database contains information such as internal and external regulations, procedures, steps, guidelines, and training. Specific examples of databases used include MongoDB and MySQL®.

[0148] Response generation engine

[0149] The server has an engine that retrieves information from a knowledge base based on the results analyzed by the NLP engine and generates appropriate responses for the user. This response generation engine has the following functions:

[0150] Template Selection: Choose the appropriate template.

[0151] Information Embedding: Embed retrieved information into the template. Specifically, use a template engine such as Mustache or Handlebars.

[0152] Action execution engine

[0153] The server includes an engine that performs notification and escalation actions in response to specific inquiries. For example, if there is an inquiry about unpaid wages, it will send a notification to the HR department. Notifications may use the Slack API or an SMTP server for sending emails.

[0154] Information provision means

[0155] The terminal displays the response message sent from the server to the user. The display method is either text display or text-to-speech.

[0156] Specific example

[0157] Example 1: Inquiry about leave application procedures

[0158] 1. The user types the text, "How do I apply for leave?"

[0159] 2. The terminal sends user input to the server.

[0160] 3. The server uses a natural language processing engine to analyze the question and identify information related to "leave application procedures."

[0161] 4. The server retrieves the relevant information from the knowledge base and generates a response similar to the following: "To submit a leave request, follow these steps: 1. Log in to the HR portal. 2. Click the 'Leave Request' tab. 3. Enter the required information and click the submit button."

[0162] 5. The terminal displays this response to the user.

[0163] Example 2: Inquiry about unpaid wages

[0164] 1. The user voice-inputs, "My salary hasn't been paid, who should I ask about it?"

[0165] 2. The device converts the audio to text using the Google Speech-to-Text API and sends it to the server.

[0166] 3. The server analyzes the question using a natural language processing engine and identifies the "contact information for inquiries regarding unpaid wages."

[0167] 4. The server retrieves information about the HR department contact person from its knowledge base and generates a response stating, "Inquiries regarding unpaid wages will be escalated to the HR department contact person."

[0168] 5. The server uses the Slack API to send an escalation notification to the HR department.

[0169] 6. The device displays the generated response to the user and confirms that the notification has been sent.

[0170] Example of a prompt:

[0171] "Could you please explain the procedure for requesting leave?"

[0172] "Please tell me how to deal with unpaid wages."

[0173] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0174] Step 1:

[0175] The user enters a question into the terminal's interface.

[0176] Input methods include text input or voice input.

[0177] In terms of specific actions, the user either types "How do I go about applying for leave?" into the text box or presses the voice input button to speak the question.

[0178] The entered text or audio will be sent to the next step.

[0179] Step 2:

[0180] The terminal sends the user's text input or voice to the server.

[0181] In the case of voice input, the device uses speech recognition to convert the speech into text.

[0182] Input: Text or audio data entered by the user.

[0183] Data processing: Speech-to-text conversion using speech recognition technology.

[0184] Output: User questions in text format

[0185] Specifically, the process involves converting speech to text using APIs such as the Google Speech-to-Text API, and then sending the converted text to a server over the network.

[0186] Step 3:

[0187] The server analyzes the received text data using a natural language processing (NLP) engine.

[0188] The NLP engine performs tokenization, part-of-speech tagging, and semantic analysis.

[0189] Input: A text-formatted question sent from the device.

[0190] Data processing:

[0191] Tokenization: Breaking down a question into words or phrases.

[0192] Part-of-speech tagging: Identify the part of speech of each token.

[0193] Semantic analysis: Interpret the meaning of the entire question and identify the user's intent.

[0194] Output: Structured data indicating user intent

[0195] Specifically, the system uses Google Cloud Natural Language API and Amazon Comprehend to analyze the questions.

[0196] Step 4:

[0197] The server searches for relevant information from its knowledge base based on the results analyzed by the NLP engine.

[0198] Input: Structured data indicating user intent

[0199] Data processing: Executing queries against knowledge bases (such as MongoDB or MySQL)

[0200] Output: Appropriate information in response to the user's question

[0201] Specifically, the process involves executing a database query to retrieve information regarding "leave application procedures."

[0202] Step 5:

[0203] The server's response generation engine generates an appropriate answer based on the information it has received.

[0204] Input: Information obtained from a knowledge base

[0205] Data processing:

[0206] Template Selection: Select the appropriate response template.

[0207] Information Embedding: Embed the retrieved information into the template.

[0208] Output: Final response message to the user

[0209] Specifically, the system generates response messages using template engines such as Mustache or Handlebars.

[0210] Step 6:

[0211] The server performs the necessary actions (notifications or escalation) in response to specific queries.

[0212] Input: User's questions and analysis results

[0213] Data processing: Creating and sending notification messages

[0214] Output: Escalation notifications and reports

[0215] Specifically, the system uses the Slack API to send escalation notifications to the HR department.

[0216] Step 7:

[0217] The terminal displays the response sent from the server to the user.

[0218] The display method is either text display or text-to-speech.

[0219] Input: Response message from the server

[0220] Data processing: Display in a user-friendly format.

[0221] Output: Response message displayed to the user

[0222] Specifically, the user interface will display the following message: "To request leave, follow these steps: 1. Log in to the HR portal. 2. Click the 'Leave Request' tab. 3. Enter the required information and click the submit button."

[0223] (Application Example 1)

[0224] 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."

[0225] Modern users demand interactive and personalized information. However, traditional systems have struggled to effectively analyze user input and quickly deliver the most relevant content to individual users. Furthermore, the rapid increase in video content has led to problems with users spending a lot of time finding the right videos. This could potentially lead to decreased user satisfaction and engagement.

[0226] 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.

[0227] In this invention, the server includes means for receiving a question entered by the user, means for analyzing the received question and searching for relevant information, means for obtaining information from a knowledge base based on the analysis results, means for generating a response to provide the obtained information to the user, and means for providing the user with personalized video recommendations. This makes it possible to quickly and efficiently recommend relevant video content based on the question entered by the user, thereby improving the user experience.

[0228] "Means for receiving user-entered questions" refers to an interface that allows users to input questions into the system via text boxes or voice input and send those questions to the server in digital format.

[0229] "Natural language processing means for searching for relevant information" refers to technology that analyzes text data received from users to understand their intent and search for appropriate information or content.

[0230] A "knowledge base" is a database that stores a large amount of related information and data, and is used to quickly provide appropriate information in response to user questions.

[0231] A "response generation means" is a means for generating an appropriate answer to provide to the user based on the analysis results and information obtained from a knowledge base.

[0232] "Means of providing personalized video recommendations to users" refers to methods for selecting and presenting the most suitable video content for each individual user based on their questions and past behavioral data.

[0233] "Action execution means for sending notifications and escalating" refers to means for taking appropriate actions, such as sending notifications to relevant departments or personnel, based on the content of a question from a specific user.

[0234] "Voice recognition means" refers to a technology for converting questions entered by a user via voice into text format, and is a means for analyzing voice signals and converting their content into character data.

[0235] Modes for carrying out the invention

[0236] The present invention is a system that personalizedly recommends relevant information and video content based on questions entered by the user. Specific embodiments for carrying out the present invention are described below.

[0237] Hardware and software to use

[0238] hardware

[0239] Smartphone: A device used by users to input questions and receive responses.

[0240] Server: A device that performs data processing and manages the knowledge base.

[0241] software

[0242] Natural Language Processing Engine (NLP): For example, the Google Cloud Natural Language API. It analyzes user text and voice input.

[0243] Video recommendation algorithms: For example, APIs from video streaming services (YouTube® API) or proprietary recommendation systems. These recommend appropriate videos to users.

[0244] Database: For example, MySQL or MongoDB. Used as a knowledge base.

[0245] Data processing and data calculation

[0246] smartphone

[0247] The smartphone provides a text box or voice input button for the user to enter a question. This input is sent to the server via an API. In the case of voice input, a speech recognition mechanism is used to convert the speech to text.

[0248] server

[0249] The server will perform the following actions:

[0250] Natural language processing engines analyze text data received from users to understand their intent. This includes tokenization, part-of-speech tagging, and semantic analysis.

[0251] The knowledge base searches for appropriate information and video content based on the results analyzed by the natural language processing engine.

[0252] The response generation engine generates appropriate answers for the user based on information retrieved from a knowledge base. This includes video links and related information.

[0253] The video recommendation algorithm provides personalized video recommendations based on the user's past viewing history and feedback.

[0254] Smartphone (response display)

[0255] The smartphone displays the response received from the server to the user. A text-to-speech function is also provided as needed.

[0256] Specific examples and prompt statements

[0257] Example 1: Desire for knowledge

[0258] The user types "I want to learn more about artificial intelligence." The system responds as follows:

[0259] User: "I want to learn more about artificial intelligence."

[0260] system:

[0261] I am searching for information about artificial intelligence...

[0262] I recommend this video:

[0263] 1. Fundamentals of Artificial Intelligence

[0264] 2. Latest AI technology

[0265] 3. The Future of AI

[0266] Specific example 2: The desire for entertainment

[0267] The user types "Tell me about some popular movies lately." The system responds as follows:

[0268] User: "Tell me about some popular movies lately."

[0269] system:

[0270] I'm searching for recently popular movies...

[0271] These movies are popular:

[0272] 1. Trailer for the movie 'XYZ'

[0273] 2. Review of the movie 'ABC'

[0274] 3. Interview about the film '123'

[0275] By combining these components, this invention provides users with instantly appropriate information and video content in response to a variety of questions they enter, thereby improving the user experience.

[0276] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0277] Step 1:

[0278] The user enters a question. The user enters the question using the text box or voice input button on their smartphone. The smartphone receives the entered text data (or voice data) and sends it to the server. The user's question is given as input and sent to the server as output.

[0279] Step 2:

[0280] The server receives input data from the user. The server receives the input text data or voice data, and in the case of voice data, converts it into text data using voice recognition means. Text data (or converted text data) is provided as input, and text data for analysis is obtained as output.

[0281] Step 3:

[0282] The server activates a natural language processing engine to analyze the text data. The natural language processing engine uses technologies such as tokenization, part-of-speech tagging, and semantic analysis to understand the intention of the user's question. Text data is provided as input, and the result of intention analysis is obtained as output.

[0283] Step 4:

[0284] The server obtains relevant information from the knowledge base. Based on the analysis result, the server searches for and obtains relevant information and content from the knowledge base (database). The result of intention analysis is provided as input, and relevant information is obtained as output.

[0285] Step 5:

[0286] The server uses a video recommendation algorithm to select personalized videos. The video recommendation algorithm considers the user's past viewing history and feedback, and calculates data for recommending the optimal videos. The analysis result and the user's history data are provided as input, and a list of recommended videos is obtained as output.

[0287] Step 6:

[0288] The server uses a response generation engine to generate a response to provide to the user. The response generation engine generates a response to the user based on the acquired information and the recommended video list. Relevant information and the recommended video list are given as input, and the final response message is obtained as output.

[0289] Step 7:

[0290] The server sends a response message to the smartphone. The final response message is sent from the server to the smartphone. The final response message is given as input, and the output is displayed on the smartphone.

[0291] Step 8:

[0292] The smartphone displays the response message to the user. The smartphone displays the response message received from the server to the user. It also provides a voice reading function if necessary. The final response message is given as input, and it is displayed to the user as output.

[0293] 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.

[0294] This invention incorporates a function that recognizes the user's emotions and adjusts the response accordingly, in addition to analyzing user-inputted questions and providing relevant information quickly. Its specific form is described below.

[0295] User Interface (UI)

[0296] Terminal: Provides an interface for the user to inquire information. It includes a text box for text input, a voice input button, etc. In the case of voice input, a voice recognition means is incorporated to convert voice into text.

[0297] Natural Language Processing (NLP) Engine

[0298] Server: Analyzes the text data received from the user and has a natural language processing engine for understanding the intention. This engine is composed of the following technologies:

[0299] Tokenization: Decomposes the text into words and phrases.

[0300] Part-of-Speech Tagging: Identifies the part of speech of each word.

[0301] Semantic Analysis: Analyzes the meaning of the whole text to identify the user's intention.

[0302] Knowledge Base

[0303] Server: Has a large-scale database for retrieving relevant information based on the user's intention. This database includes regulations, procedures, guidelines, training, etc. inside and outside the company. <00009​​​​​​​​​​​​​​​​​​​​

[0309] Server: It has an emotion engine for recognizing emotions from user input. This engine analyzes the degree of emotion from the user's text and voice input and identifies emotions such as joy, anger, and sadness.

[0310] Action execution engine

[0311] Server: Includes an engine that takes notification or escalation actions in response to specific inquiries. For example, if a user's emotions exceed a threshold for anger or sadness, it sends a notification to a specific department.

[0312] Information provision means

[0313] Terminal: Displays response messages sent from the server to the user. These may be displayed as text or read aloud.

[0314] Specific example

[0315] Example 1: Inquiry about leave application procedures

[0316] 1. User: Types the following text: "How do I apply for leave?"

[0317] 2. Terminal: Sends user input to the server.

[0318] 3. Server: The server uses a natural language processing engine to analyze the question and identify information related to "leave application procedures."

[0319] 4. Server: Retrieves relevant information from the knowledge base and generates a response stating, "To submit a leave request, follow these steps: 1. Log in to the HR portal. 2. Click the 'Leave Request' tab. 3. Enter the required information and click the submit button."

[0320] 5. Server: The emotion engine analyzes the user's emotions. Since it is determined to be a normal question, it determines that no special action is required.

[0321] 6. Terminal: Displays the generated response to the user.

[0322] Example 2: Inquiry about unpaid wages

[0323] 1. User: "My salary hasn't been paid, who should I ask about it?" (Voice input)

[0324] 2. Terminal: Converts speech to text and sends it to the server.

[0325] 3. Server: Uses a natural language processing engine to analyze the question and identify the "contact information for inquiries regarding unpaid wages."

[0326] 4. Server: Retrieves HR department contact information from the knowledge base and generates a response stating, "Inquiries regarding unpaid wages will be escalated to the HR department contact."

[0327] 5. Server: The emotion engine analyzes the user's emotions. If the user is showing strong anger or sadness, the action execution engine sends an escalation notification to the HR department.

[0328] 6. Terminal: Display the generated response to the user and confirm that the notification has been sent.

[0329] By combining these functions, the present invention provides a system that can immediately provide appropriate information in response to user inquiries, respond in a way that takes user emotions into consideration, and quickly execute necessary actions.

[0330] The following describes the processing flow.

[0331] Detailed program processing flow

[0332] Example 1: Inquiry about leave application procedures

[0333] Step 1:

[0334] The user types the text, "How do I apply for leave?"

[0335] Step 2:

[0336] The device receives user input. If the user provides voice input, the speech recognition engine first converts the speech into text.

[0337] Step 3:

[0338] The terminal sends the received text data to the server.

[0339] Step 4:

[0340] The server's natural language processing engine analyzes the user's text. First, it tokenizes the text, then tags it with parts of speech, and finally performs semantic analysis to determine the user's intent.

[0341] Step 5:

[0342] Based on the analysis results, the server queries the knowledge base for information regarding "leave application procedures."

[0343] Step 6:

[0344] The server retrieves the relevant information from the knowledge base.

[0345] Step 7:

[0346] The server's response generation engine creates a reply message based on the information it has obtained. Specifically, it generates a message that says, "To request leave, follow these steps: 1. Log in to the HR portal. 2. Click the 'Leave Request' tab. 3. Enter the required information and click the submit button."

[0347] Step 8:

[0348] The server's emotion engine analyzes the user's input text and determines the degree of emotion. In this case, it recognizes it as a normal question and determines that there is no special emotional response.

[0349] Step 9:

[0350] The server sends the generated message to the terminal.

[0351] Step 10:

[0352] The device displays the message to the user. In the case of voice output, the text is converted to speech and read aloud.

[0353] Example 2: Inquiry about unpaid wages

[0354] Step 1:

[0355] The user inputs a voice message saying, "My salary hasn't been paid, who should I ask about it?"

[0356] Step 2:

[0357] The device receives the audio and converts it to text using a speech recognition engine.

[0358] Step 3:

[0359] The terminal sends text data to the server.

[0360] Step 4:

[0361] The server's natural language processing engine analyzes the user's text. It processes the text in the following order: tokenization, part-of-speech tagging, and semantic analysis, identifying the user's intent as "an inquiry about unpaid wages."

[0362] Step 5:

[0363] The server executes queries against the knowledge base based on the analysis results.

[0364] Step 6:

[0365] The server retrieves information on "contact information for inquiries regarding unpaid wages" from its knowledge base. Here, it retrieves information on the HR department contact person.

[0366] Step 7:

[0367] The server's response generation engine creates a message stating, "Inquiries regarding unpaid wages will be escalated to the HR department."

[0368] Step 8:

[0369] The server's emotion engine analyzes the user's input text to determine the degree of emotion. If the user indicates strong anger or sadness, the action execution engine sends an escalation notification to the HR department.

[0370] Step 9:

[0371] The server generates an escalation notification and sends it to the HR department.

[0372] Step 10:

[0373] The server sends the generated message to the terminal.

[0374] Step 11:

[0375] The device displays the message to the user and confirms that the notification has been sent. In the case of voice output, it converts the text to speech and reads it aloud.

[0376] In this way, we can provide appropriate information in response to user questions, respond to their emotions, and quickly take necessary actions.

[0377] (Example 2)

[0378] 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 device 14 as the "terminal".

[0379] Traditional question-answering systems only provide information in response to user questions, failing to adequately improve user satisfaction by not considering user emotions or escalating problems. Furthermore, the processing and analysis of voice-input questions were sometimes inefficient. This made it difficult for users to obtain quick and appropriate answers, and particularly led to delays in responses when users were emotionally agitated.

[0380] The identification processing performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving a question entered by the user, natural language processing means for analyzing the received question and searching for relevant information, means for obtaining information from a database based on the analysis results, response generation means for providing the obtained information to the user, and emotion recognition means for analyzing emotions from the user's input. This makes it possible to provide quick and appropriate information in response to the user's question, and to automatically perform responses that take the user's emotions into consideration and necessary escalations.

[0381] A "user" refers to an individual or group that enters a question into the system.

[0382] "Means of receiving questions" refers to a device or software module that has the functionality to receive text or audio input from a user.

[0383] "Natural language processing means" refers to all technologies used to analyze received questions and understand the user's intent.

[0384] "Means of obtaining information from a database" refers to a device or software module that has the function of searching for and extracting relevant information from a database based on an analyzed query.

[0385] "Response generation means" refers to a device or software module that has the function of creating a response to be provided to the user based on acquired information.

[0386] "Emotion recognition means" refers to a device or software module that has the function of identifying emotions from user input and analyzing their degree.

[0387] "Action execution means" refers to a device or software module that has the functionality to perform notifications or escalations based on the detected user's emotions.

[0388] "Speech recognition means" refers to a device or software module that has the function of analyzing data input as speech and converting its content into text format.

[0389] This invention is a system that analyzes user-inputted questions and quickly provides relevant information. Furthermore, it incorporates a function to recognize the user's emotions and adjust its response accordingly. An embodiment of this system is described in detail below.

[0390] User Interface (UI)

[0391] Terminal: Provides an interface for users to query information. This interface includes text boxes for text input and buttons for voice input. In the case of voice input, a speech recognition mechanism is built in to convert the voice data into text.

[0392] Natural Language Processing (NLP) Engine

[0393] Server: It has a natural language processing engine that analyzes text data sent by the user and understands the user's intent. This engine consists of the following technologies:

[0394] Tokenization: Breaking down a text into individual words or phrases.

[0395] Part-of-speech tagging: Identifying the part of speech of each word.

[0396] Semantic analysis: Analyzes the meaning of the entire text to identify the user's intent.

[0397] Knowledge Base

[0398] Server: It has a large database for searching relevant information based on user intent. This database contains information such as internal and external regulations, procedures, steps, guidance, and training.

[0399] Response generation engine

[0400] Server: There is an engine that retrieves information from a knowledge base based on the results analyzed by the NLP engine and generates appropriate responses for the user. This response generation engine has the following functions:

[0401] Template Selection: Select the appropriate template.

[0402] Information Embedding: Embed the retrieved information into the template.

[0403] Emotional Engine

[0404] Server: It has an emotion engine for recognizing emotions from user input. This engine analyzes the degree of emotion from the user's text and voice input and identifies emotions such as joy, anger, and sadness.

[0405] Action execution engine

[0406] Server: Includes an engine that takes notification or escalation actions in response to specific inquiries. For example, if a user's emotions exceed a threshold for anger or sadness, it sends a notification to a specific department.

[0407] Information provision means

[0408] Terminal: Displays the response message sent from the server to the user. This response may be displayed in text format or read aloud.

[0409] Specific example

[0410] Example 1: Inquiry about leave application procedures

[0411] 1. User: Types the following text: "How do I apply for leave?"

[0412] 2. Terminal: Sends user input to the server.

[0413] 3. Server: The NLP engine analyzes the question and identifies information related to "leave application procedures".

[0414] 4. Server: Retrieves relevant information from the knowledge base and generates a response stating, "To submit a leave request, follow these steps: 1. Log in to the portal. 2. Click the 'Leave Request' tab. 3. Enter the required information and click the submit button."

[0415] 5. Server: The emotion engine analyzes the user's emotions. Since it is determined to be a normal question, it determines that no special action is required.

[0416] 6. Terminal: Displays the generated response to the user.

[0417] Example 2: Inquiry about unpaid wages

[0418] 1. User: "My salary hasn't been paid. Who should I contact?" (Voice input)

[0419] 2. Terminal: Converts speech to text and sends it to the server.

[0420] 3. Server: The NLP engine analyzes the question and identifies the "contact point for inquiries regarding unpaid wages."

[0421] 4. Server: Retrieves information on the HR department contact person from the knowledge base and generates a response stating, "Please direct inquiries regarding unpaid wages to the HR department contact person."

[0422] 5. Server: The emotion engine analyzes the user's emotions. If the user is showing strong anger or sadness, the action execution engine sends an escalation notification to the HR department.

[0423] 6. Terminal: Display the generated response to the user and confirm that the notification has been sent.

[0424] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0425] Step 1:

[0426] User: The user enters their question through the interface. For example, they might type "Please tell me about the company's benefits" into the text box. Alternatively, using voice input, they might say into the microphone, "My salary hasn't been paid, who should I contact?" The input is sent to the terminal as text or voice data.

[0427] Step 2:

[0428] Terminal: Receives user input, and if the input is voice data, converts the voice data into text data using speech recognition. This converted text data, or the original text data, is sent to the server. The input is the original user's question, and the output is text data.

[0429] Step 3:

[0430] Server: Receives text data and analyzes it using a natural language processing (NLP) engine. Specifically, it first tokenizes the input sentence, breaking it down into words and phrases. Then, it performs part-of-speech tagging to identify the grammatical role of each word. Next, it performs semantic analysis to analyze the meaning of the entire sentence and identify the user's intent. The input is text data, and the output is analyzed intent data.

[0431] Step 4:

[0432] Server: Based on intent data analyzed by the NLP engine, it searches for relevant information from a knowledge base. This knowledge base includes internal and external regulations, procedures, steps, and guidelines. It executes database queries within the knowledge base to retrieve appropriate information. The input is the analyzed intent data, and the output is the retrieved information.

[0433] Step 5:

[0434] Server: Uses a response generation engine to generate a response to provide to the user based on the information obtained. First, it selects an appropriate template, and then it embeds the obtained information into the template to form the final response. The input is the information obtained, and the output is the generated response text.

[0435] Step 6:

[0436] Server: Uses an emotion engine to analyze the user's emotions from the generated response and the user's input text. The emotion engine analyzes specific words, phrases, and tones in the user's text to identify emotions such as joy, anger, and sadness. The input is the user's input text and the generated response, and the output is emotion data.

[0437] Step 7:

[0438] Server: Based on the emotion data identified by the emotion engine, the action execution engine performs specific actions. For example, if a user's emotion exceeds the threshold for anger or sadness, a notification is sent to a specific department. The input is emotion data, and the output is an escalation notification.

[0439] Step 8:

[0440] Terminal: Receives response messages sent from the server and displays them to the user. This response is displayed in text format or read aloud. Specifically, text is displayed on the terminal's display, or audio is played through the speaker. The input is the response from the server, and the output is the display or audio output to the user.

[0441] (Application Example 2)

[0442] 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".

[0443] In autonomous vehicles, there is a need for a system that can respond quickly and appropriately to passengers' questions, and furthermore, respond flexibly according to the passengers' emotional state. However, conventional systems have had difficulty realizing responses that take passengers' emotions into account or escalation processing based on specific emotional states. As a result, passenger dissatisfaction and problems may not be properly resolved. To solve these problems, this invention provides a new system that combines natural language processing and sentiment analysis.

[0444] 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.

[0445] In this invention, the server includes means for receiving user input, means for natural language processing for analyzing user input, means for obtaining information from a knowledge base based on the analysis results, means for generating a response that provides the obtained information to the user, means for analyzing emotions from user input, and means for executing necessary actions based on the analyzed emotions. This makes it possible to answer user questions quickly and appropriately, and to respond flexibly based on the user's emotional state.

[0446] "Means for receiving user-entered questions" refers to an interface for receiving questions entered by the user in text or voice.

[0447] "Natural language processing means" are technical means for analyzing received text data and understanding its intent.

[0448] "Means of obtaining information from a knowledge base" refers to a function that extracts necessary information from relevant databases based on the analysis results.

[0449] A "response generation means" is an engine that generates answers to be provided to the user based on the acquired information.

[0450] An "emotion analysis tool" is a system that identifies emotions from user input text or voice and analyzes their emotional state.

[0451] An "action execution mechanism" is a function that performs a specific action (such as notification or escalation) based on the analyzed emotional state.

[0452] "Speech recognition means" refers to technical means for converting speech input into text data.

[0453] "Means of providing information to the user" refers to an interface for presenting the generated response to the user in text or audio format.

[0454] The present invention is implemented as a system for responding quickly and appropriately to passenger questions in autonomous vehicles. Specifically, it has the function of analyzing questions entered by the user, obtaining appropriate information from a knowledge base, providing the generated response to the user, and also analyzing the user's emotional state and taking necessary actions based on that analysis.

[0455] Hardware and software configuration

[0456] This system consists of the following hardware and software.

[0457] hardware

[0458] Server: Provides the primary computing resources for backend processing.

[0459] Terminal: Infotainment system within an autonomous vehicle. Includes user interfaces such as touchscreens and voice recognition microphones.

[0460] software

[0461] Natural Language Processing Engine (NLP Engine): Uses the Hugging Face natural language processing model. It has the capability to analyze questions using the transformers library.

[0462] Emotion analysis engine: Also using Hugging Face's pipeline, it identifies the user's emotional state.

[0463] Knowledge base: A large-scale database containing vehicle functions, operational information, etc.

[0464] Response generation engine: Generates responses to the user based on the acquired information.

[0465] Action Execution Engine: Based on sentiment analysis results, it executes actions (notifications or escalations) based on specific emotional states.

[0466] System operation

[0467] 1. Receiving and analyzing user input

[0468] The user enters a question via voice or text through the vehicle's infotainment system. In the case of voice input, a speech recognition system converts it into text. The converted text is sent to a server, where a natural language processing engine analyzes the intent of the question.

[0469] 2. Information acquisition from knowledge bases

[0470] Based on the results analyzed by the natural language processing engine, the server searches for and retrieves relevant information from its knowledge base. This information covers a wide range of topics, including vehicle operation methods and route information.

[0471] 3. Response generation and sentiment analysis

[0472] Based on the acquired information, the response generation engine generates an appropriate answer. Simultaneously, the sentiment analysis engine analyzes the user's emotional state from their questions and other inputs, and determines the analysis result.

[0473] 4. Performing a specific action

[0474] If the emotion analysis results are "negative," for example, if anger or frustration is strong, the action execution engine will send a notification to a specific department or service center. This notification is a measure to ensure that appropriate action is taken quickly.

[0475] Specific example

[0476] Example question: "Please tell me how to use the air conditioner."

[0477] Analysis results and response: The NLP engine analyzes the question, retrieves information about "how to use the air conditioner" from the knowledge base, and generates the response, "To operate the air conditioner, tap the 'Air Conditioner' tab on the central touchscreen."

[0478] Sentiment Analysis and Action: The sentiment analysis engine determines the user's emotion to be "normal." In this case, no special action is required.

[0479] Example question: "I want to change the background music in my car, how do I do that?"

[0480] Analysis results and response: The NLP engine analyzes the question, retrieves information about "how to change background music" from the knowledge base, and generates the response: "Click the menu icon at the top of the central touchscreen -> 'Settings' -> 'Audio Settings' -> Select from 'Playlist'."

[0481] Thus, the present invention provides a system for autonomous vehicles that can respond quickly to passengers' questions and further provide flexible responses according to their emotional state.

[0482] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0483] Step 1:

[0484] The terminal receives user input.

[0485] The user enters their question into the vehicle's infotainment system via voice or text.

[0486] In the case of voice input, a speech recognition system converts the speech into text and generates text data. The entered text is then sent from the terminal to the server.

[0487] Step 2:

[0488] The server analyzes the question.

[0489] The server uses a natural language processing engine to analyze the received text data and identify the intent of the input. This process involves tokenization, part-of-speech tagging, and semantic analysis to understand the structure and content of the user's question. The input is the user's question text, and the output is the intent of the question and important keywords.

[0490] Step 3:

[0491] The server retrieves information from the knowledge base.

[0492] The server searches for relevant information from a knowledge base based on the analysis results of the natural language processing engine. The knowledge base includes vehicle operation instructions and route information, and appropriate data is extracted. The input is the analysis result, and the output is the data of the relevant information.

[0493] Step 4:

[0494] The server generates a response.

[0495] The server uses a response generation engine to generate answers for the user based on the acquired information. The generated answers are created by embedding information into a template. The input is information acquired from the knowledge base, and the output is the answer text provided to the user.

[0496] Step 5:

[0497] The server performs sentiment analysis.

[0498] The server uses an emotion analysis engine to identify emotions from the user's input text. Here, the emotional state the user was in when asking a question (e.g., joy, anger, sadness, etc.) is determined. The input is the user's question text, and the output is the result of the emotional state analysis.

[0499] Step 6:

[0500] The server performs the necessary actions.

[0501] The server performs actions based on the sentiment analysis results for specific emotional states. For example, if a user is highly angry, the action execution engine sends a notification to a specific department or person in charge. The input is the sentiment analysis result, and the output is the result of the action (such as sending a notification).

[0502] Step 7:

[0503] The device provides information to the user.

[0504] The response text generated by the server is sent to the terminal and presented to the user. It may be displayed in text or audio format. The input is the response text from the server, and the output is information provided to the passenger.

[0505] 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.

[0506] 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.

[0507] 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.

[0508] [Second Embodiment]

[0509] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0510] 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.

[0511] 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).

[0512] 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.

[0513] 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.

[0514] 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).

[0515] 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.

[0516] 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.

[0517] 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.

[0518] 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.

[0519] 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.

[0520] 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".

[0521] This invention is a system that analyzes questions entered by a user and quickly provides the relevant information. Its specific form is described below.

[0522] User Interface (UI)

[0523] Terminal: Provides an interface for users to query information. This includes text boxes for text input and voice input buttons. In the case of voice input, a speech recognition mechanism is built in to convert speech into text.

[0524] Natural Language Processing (NLP) Engine

[0525] Server: It has a natural language processing engine that analyzes text data received from users and understands their intent. This engine consists of the following technologies:

[0526] Tokenization: Breaking down a text into individual words or phrases.

[0527] Part-of-speech tagging: Identifying the part of speech of each word.

[0528] Semantic analysis: Analyzes the meaning of the entire text to identify the user's intent.

[0529] Knowledge Base

[0530] Server: It has a large database for searching relevant information based on user intent. This database includes internal and external regulations, procedures, steps, guidelines, training, etc.

[0531] Response generation engine

[0532] Server: There is an engine that retrieves information from a knowledge base based on the results analyzed by the NLP engine and generates appropriate responses for the user. The response generation engine has the following functions:

[0533] Template Selection: Choose the appropriate template.

[0534] Information Embedding: Embed the retrieved information into the template.

[0535] Action execution engine

[0536] Server: Includes an engine that takes notification or escalation actions in response to specific inquiries. For example, if a user's inquiry is about unpaid wages, it sends a notification to the HR department.

[0537] Information provision means

[0538] Terminal: Displays response messages sent from the server to the user. These may be displayed as text or read aloud.

[0539] Specific example

[0540] Example 1: Inquiry about leave application procedures

[0541] 1. User: Types the following text: "How do I apply for leave?"

[0542] 2. Terminal: Sends user input to the server.

[0543] 3. Server: The server uses a natural language processing engine to analyze the question and identify information related to "leave application procedures."

[0544] 4. Server: Retrieves relevant information from the knowledge base and generates a response stating, "To submit a leave request, follow these steps: 1. Log in to the HR portal. 2. Click the 'Leave Request' tab. 3. Enter the required information and click the submit button."

[0545] 5. Terminal: Displays the generated response to the user.

[0546] Example 2: Inquiry about unpaid wages

[0547] 1. User: "My salary hasn't been paid, who should I ask about it?" (Voice input)

[0548] 2. Terminal: Converts speech to text and sends it to the server.

[0549] 3. Server: Uses a natural language processing engine to analyze the question and identify the "contact information for inquiries regarding unpaid wages."

[0550] 4. Server: Retrieves HR department contact information from the knowledge base and generates a response stating, "Inquiries regarding unpaid wages will be escalated to the HR department contact."

[0551] 5. Server: Sends an escalation notification to the HR department.

[0552] 6. Terminal: Display the generated response to the user and confirm that the notification has been sent.

[0553] This invention provides a system that, by combining these functions, can instantly provide appropriate information in response to a variety of questions entered by the user and quickly process the necessary actions.

[0554] The following describes the processing flow.

[0555] Detailed program processing flow

[0556] Example 1: Inquiry about leave application procedures

[0557] Step 1:

[0558] The user types the text, "How do I apply for leave?"

[0559] Step 2:

[0560] The device receives user input. If the user provides voice input, it is first converted to text by the speech recognition engine.

[0561] Step 3:

[0562] The terminal sends the received text data to the server.

[0563] Step 4:

[0564] The server's natural language processing engine analyzes the user's text. First, it tokenizes the text, then tags it with parts of speech, and finally performs semantic analysis to determine the user's intent.

[0565] Step 5:

[0566] Based on the analysis results, the server queries the knowledge base for information regarding "leave application procedures."

[0567] Step 6:

[0568] The server retrieves the relevant information from the knowledge base.

[0569] Step 7:

[0570] The server's response generation engine creates a reply message based on the information it has obtained. Specifically, it generates a message that says, "To request leave, follow these steps: 1. Log in to the HR portal. 2. Click the 'Leave Request' tab. 3. Enter the required information and click the submit button."

[0571] Step 8:

[0572] The server sends the generated message to the terminal.

[0573] Step 9:

[0574] The device displays the message to the user. In the case of voice output, the text is converted to speech and read aloud.

[0575] Example 2: Inquiry about unpaid wages

[0576] Step 1:

[0577] The user inputs a voice message saying, "My salary hasn't been paid, who should I ask about it?"

[0578] Step 2:

[0579] The device receives the audio and converts it to text using a speech recognition engine.

[0580] Step 3:

[0581] The terminal sends text data to the server.

[0582] Step 4:

[0583] The server's natural language processing engine analyzes the user's text. It processes the text in the following order: tokenization, part-of-speech tagging, and semantic analysis, identifying the user's intent as "an inquiry about unpaid wages."

[0584] Step 5:

[0585] The server executes queries against the knowledge base based on the analysis results.

[0586] Step 6:

[0587] The server retrieves information on "contact information for inquiries regarding unpaid wages" from its knowledge base. Here, it retrieves information on the HR department contact person.

[0588] Step 7:

[0589] The server's response generation engine creates a message stating, "Inquiries regarding unpaid wages will be escalated to the HR department."

[0590] Step 8:

[0591] The server's action execution engine generates an escalation notification and sends it to the HR department.

[0592] Step 9:

[0593] The server sends the generated message to the terminal.

[0594] Step 10:

[0595] The device displays messages to the user and reads them aloud if necessary.

[0596] These steps enable us to provide immediate and accurate answers to user inquiries and take necessary actions.

[0597] (Example 1)

[0598] 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."

[0599] In today's world, systems that allow users to obtain information quickly and accurately are crucial. However, many systems lack the ability to accurately understand user questions and fail to provide appropriate information. Furthermore, while notifications and escalations are necessary in certain situations, few systems automate these processes. Therefore, there is a need to develop systems that can efficiently respond to user questions and take necessary actions.

[0600] 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.

[0601] In this invention, the server includes means for receiving questions entered by the user, natural language processing means for analyzing the questions and searching for relevant information, means for obtaining information from a knowledge base based on the analysis results, response generation means for providing the obtained information to the user, and action execution means for performing notifications and escalations based on the user's questions. This makes it possible to provide information quickly and accurately in response to a wide range of user questions and to automatically perform necessary actions.

[0602] A "user" is an individual or group that uses the system to input questions and obtain information.

[0603] A "question" is a text or voice message that a user enters to ask the system for information.

[0604] A "terminal" refers to a device (such as a PC, smartphone, or tablet) that a user uses to input questions and receive responses from a server.

[0605] A "server" is a central processing unit that analyzes questions received from users and searches for, retrieves, and provides the necessary information.

[0606] "Natural language processing methods" refer to technologies for analyzing text data received from users and understanding its intent. Specifically, this includes tokenization, part-of-speech tagging, and semantic analysis.

[0607] A "tokenization method" is a technique that breaks down a question into words or phrases.

[0608] A "part-of-speech tagging method" is a technique for identifying the part of speech of each word.

[0609] "Semantic analysis means" refers to technologies that analyze the meaning of an entire text and identify the user's intent.

[0610] A "knowledge base" is a database that stores information to provide appropriate answers to user questions.

[0611] A "response generation method" is a technology that generates answers to users based on information obtained from a knowledge base. Specifically, it includes template selection and information embedding.

[0612] "Template selection method" refers to the technique of selecting the appropriate response template.

[0613] "Information embedding means" refers to a technology that generates a final response by embedding the necessary information into a selected template.

[0614] An "action execution method" is a technology that automatically performs notifications or escalations in response to specific inquiries.

[0615] "Voice recognition means" refers to technology that converts voice input from a user into text.

[0616] A "notification" is a message that informs specific recipients of important information.

[0617] "Escalation" refers to the procedure of transferring a specific problem or inquiry to a senior officer or another department.

[0618] This invention relates to a system that analyzes user-inputted questions and quickly provides relevant information. This system provides information quickly and accurately in response to a wide range of user questions and automatically performs necessary actions.

[0619] User Interface (UI)

[0620] The terminal provides an interface for users to query information. The terminal includes text boxes for text input and voice input buttons, providing means for users to input questions. In the case of voice input, speech recognition technology is used to convert speech into text. Specifically, devices such as PCs, smartphones, and tablets are used.

[0621] Natural Language Processing (NLP) Engine

[0622] The server has a natural language processing engine that analyzes text data received from users and understands their intent. This engine uses tools such as Google Cloud Natural Language API and Amazon Comprehend to perform the following processes:

[0623] Tokenization: Breaking down a question into words or phrases.

[0624] Part-of-speech tagging: Identifying the part of speech of each word.

[0625] Semantic analysis: Analyzes the meaning of the entire question to identify the user's intent.

[0626] Knowledge Base

[0627] The server has a database for searching for relevant information based on the user's intent. This database contains information such as internal and external regulations, procedures, steps, guidelines, and training. Specific examples of databases used include MongoDB and MySQL.

[0628] Response generation engine

[0629] The server has an engine that retrieves information from a knowledge base based on the results analyzed by the NLP engine and generates appropriate responses for the user. This response generation engine has the following functions:

[0630] Template Selection: Choose the appropriate template.

[0631] Information Embedding: Embed retrieved information into the template. Specifically, use a template engine such as Mustache or Handlebars.

[0632] Action execution engine

[0633] The server includes an engine that performs notification and escalation actions in response to specific inquiries. For example, if there is an inquiry about unpaid wages, it will send a notification to the HR department. Notifications may use the Slack API or an SMTP server for sending emails.

[0634] Information provision means

[0635] The terminal displays the response message sent from the server to the user. The display method is either text display or text-to-speech.

[0636] Specific example

[0637] Example 1: Inquiry about leave application procedures

[0638] 1. The user types the text, "How do I apply for leave?"

[0639] 2. The terminal sends user input to the server.

[0640] 3. The server uses a natural language processing engine to analyze the question and identify information related to "leave application procedures."

[0641] 4. The server retrieves the relevant information from the knowledge base and generates a response similar to the following: "To submit a leave request, follow these steps: 1. Log in to the HR portal. 2. Click the 'Leave Request' tab. 3. Enter the required information and click the submit button."

[0642] 5. The terminal displays this response to the user.

[0643] Example 2: Inquiry about unpaid wages

[0644] 1. The user voice-inputs, "My salary hasn't been paid, who should I ask about it?"

[0645] 2. The device converts the audio to text using the Google Speech-to-Text API and sends it to the server.

[0646] 3. The server analyzes the question using a natural language processing engine and identifies the "contact information for inquiries regarding unpaid wages."

[0647] 4. The server retrieves information about the HR department contact person from its knowledge base and generates a response stating, "Inquiries regarding unpaid wages will be escalated to the HR department contact person."

[0648] 5. The server uses the Slack API to send an escalation notification to the HR department.

[0649] 6. The device displays the generated response to the user and confirms that the notification has been sent.

[0650] Example of a prompt:

[0651] "Could you please explain the procedure for requesting leave?"

[0652] "Please tell me how to deal with unpaid wages."

[0653] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0654] Step 1:

[0655] The user enters a question into the terminal's interface.

[0656] Input methods include text input or voice input.

[0657] In terms of specific actions, the user either types "How do I go about applying for leave?" into the text box or presses the voice input button to speak the question.

[0658] The entered text or audio will be sent to the next step.

[0659] Step 2:

[0660] The terminal sends the user's text input or voice to the server.

[0661] In the case of voice input, the device uses speech recognition to convert the speech into text.

[0662] Input: Text or audio data entered by the user.

[0663] Data processing: Speech-to-text conversion using speech recognition technology.

[0664] Output: User questions in text format

[0665] Specifically, the process involves converting speech to text using APIs such as the Google Speech-to-Text API, and then sending the converted text to a server over the network.

[0666] Step 3:

[0667] The server analyzes the received text data using a natural language processing (NLP) engine.

[0668] The NLP engine performs tokenization, part-of-speech tagging, and semantic analysis.

[0669] Input: A text-formatted question sent from the device.

[0670] Data processing:

[0671] Tokenization: Breaking down a question into words or phrases.

[0672] Part-of-speech tagging: Identify the part of speech of each token.

[0673] Semantic analysis: Interpret the meaning of the entire question and identify the user's intent.

[0674] Output: Structured data indicating user intent

[0675] Specifically, the system uses Google Cloud Natural Language API and Amazon Comprehend to analyze the questions.

[0676] Step 4:

[0677] The server searches for relevant information from its knowledge base based on the results analyzed by the NLP engine.

[0678] Input: Structured data indicating user intent

[0679] Data processing: Executing queries against knowledge bases (such as MongoDB or MySQL)

[0680] Output: Appropriate information in response to the user's question

[0681] Specifically, the process involves executing a database query to retrieve information regarding "leave application procedures."

[0682] Step 5:

[0683] The server's response generation engine generates an appropriate answer based on the information it has received.

[0684] Input: Information obtained from a knowledge base

[0685] Data processing:

[0686] Template Selection: Select the appropriate response template.

[0687] Information Embedding: Embed the retrieved information into the template.

[0688] Output: Final response message to the user

[0689] Specifically, the system generates response messages using template engines such as Mustache or Handlebars.

[0690] Step 6:

[0691] The server performs the necessary actions (notifications or escalation) in response to specific queries.

[0692] Input: User's questions and analysis results

[0693] Data processing: Creating and sending notification messages

[0694] Output: Escalation notifications and reports

[0695] Specifically, the system uses the Slack API to send escalation notifications to the HR department.

[0696] Step 7:

[0697] The terminal displays the response sent from the server to the user.

[0698] The display method is either text display or text-to-speech.

[0699] Input: Response message from the server

[0700] Data processing: Display in a user-friendly format.

[0701] Output: Response message displayed to the user

[0702] Specifically, the user interface will display the following message: "To request leave, follow these steps: 1. Log in to the HR portal. 2. Click the 'Leave Request' tab. 3. Enter the required information and click the submit button."

[0703] (Application Example 1)

[0704] 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."

[0705] Modern users demand interactive and personalized information. However, traditional systems have struggled to effectively analyze user input and quickly deliver the most relevant content to individual users. Furthermore, the rapid increase in video content has led to problems with users spending a lot of time finding the right videos. This could potentially lead to decreased user satisfaction and engagement.

[0706] 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.

[0707] In this invention, the server includes means for receiving a question entered by the user, means for analyzing the received question and searching for relevant information, means for obtaining information from a knowledge base based on the analysis results, means for generating a response to provide the obtained information to the user, and means for providing the user with personalized video recommendations. This makes it possible to quickly and efficiently recommend relevant video content based on the question entered by the user, thereby improving the user experience.

[0708] "Means for receiving user-entered questions" refers to an interface that allows users to input questions into the system via text boxes or voice input and send those questions to the server in digital format.

[0709] "Natural language processing means for searching for relevant information" refers to technology that analyzes text data received from users to understand their intent and search for appropriate information or content.

[0710] A "knowledge base" is a database that stores a large amount of related information and data, and is used to quickly provide appropriate information in response to user questions.

[0711] A "response generation means" is a means for generating an appropriate answer to provide to the user based on the analysis results and information obtained from a knowledge base.

[0712] "Means of providing personalized video recommendations to users" refers to methods for selecting and presenting the most suitable video content for each individual user based on their questions and past behavioral data.

[0713] "Action execution means for sending notifications and escalating" refers to means for taking appropriate actions, such as sending notifications to relevant departments or personnel, based on the content of a question from a specific user.

[0714] "Voice recognition means" refers to a technology for converting questions entered by a user via voice into text format, and is a means for analyzing voice signals and converting their content into character data.

[0715] Modes for carrying out the invention

[0716] The present invention is a system that personalizedly recommends relevant information and video content based on questions entered by the user. Specific embodiments for carrying out the present invention are described below.

[0717] Hardware and software to use

[0718] hardware

[0719] Smartphone: A device used by users to input questions and receive responses.

[0720] Server: A device that performs data processing and manages the knowledge base.

[0721] software

[0722] Natural Language Processing Engine (NLP): For example, the Google Cloud Natural Language API. It analyzes user text and voice input.

[0723] Video recommendation algorithms: For example, APIs from video streaming services (YouTube API) or proprietary recommendation systems. These recommend appropriate videos to users.

[0724] Database: For example, MySQL or MongoDB. Used as a knowledge base.

[0725] Data processing and data calculation

[0726] smartphone

[0727] The smartphone provides a text box or voice input button for the user to enter a question. This input is sent to the server via an API. In the case of voice input, a speech recognition mechanism is used to convert the speech to text.

[0728] server

[0729] The server will perform the following actions:

[0730] Natural language processing engines analyze text data received from users to understand their intent. This includes tokenization, part-of-speech tagging, and semantic analysis.

[0731] The knowledge base searches for appropriate information and video content based on the results analyzed by the natural language processing engine.

[0732] The response generation engine generates appropriate answers for the user based on information retrieved from a knowledge base. This includes video links and related information.

[0733] The video recommendation algorithm provides personalized video recommendations based on the user's past viewing history and feedback.

[0734] Smartphone (response display)

[0735] The smartphone displays the response received from the server to the user. A text-to-speech function is also provided as needed.

[0736] Specific examples and prompt statements

[0737] Example 1: Desire for knowledge

[0738] The user types "I want to learn more about artificial intelligence." The system responds as follows:

[0739] User: "I want to learn more about artificial intelligence."

[0740] system:

[0741] I am searching for information about artificial intelligence...

[0742] I recommend this video:

[0743] 1. Fundamentals of Artificial Intelligence

[0744] 2. Latest AI technology

[0745] 3. The Future of AI

[0746] Specific example 2: The desire for entertainment

[0747] The user types "Tell me about some popular movies lately." The system responds as follows:

[0748] User: "Tell me about some popular movies lately."

[0749] system:

[0750] I'm searching for recently popular movies...

[0751] These movies are popular:

[0752] 1. Trailer for the movie 'XYZ'

[0753] 2. Review of the movie 'ABC'

[0754] 3. Interview about the film '123'

[0755] By combining these components, this invention provides users with instantly appropriate information and video content in response to a variety of questions they enter, thereby improving the user experience.

[0756] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0757] Step 1:

[0758] The user enters a question. The user enters the question using the text box or voice input button on their smartphone. The smartphone receives the entered text data (or voice data) and sends it to the server. The user's question is given as input and sent to the server as output.

[0759] Step 2:

[0760] The server receives input data from the user. The server receives the input text data or audio data, and if it is audio data, it converts it to text data using speech recognition. Text data (or converted text data) is given as input, and text data for analysis is obtained as output.

[0761] Step 3:

[0762] The server starts up a natural language processing engine and analyzes the text data. The natural language processing engine uses techniques such as tokenization, part-of-speech tagging, and semantic analysis to understand the intent of the user's question. Text data is given as input, and the result of intent analysis is obtained as output.

[0763] Step 4:

[0764] The server retrieves relevant information from the knowledge base. Based on the analysis results, the server searches for and retrieves relevant information and content from the knowledge base (database). The result of intent analysis is given as input, and relevant information is obtained as output.

[0765] Step 5:

[0766] The server uses a video recommendation algorithm to select personalized videos. The video recommendation algorithm considers the user's past viewing history and feedback to calculate data for recommending the most suitable videos. The analysis results and the user's history data are given as input, and a list of recommended videos is obtained as output.

[0767] Step 6:

[0768] The server uses a response generation engine to generate a response to provide to the user. The response generation engine generates a response to the user based on the acquired information and the recommended video list. Relevant information and the recommended video list are given as input, and the final response message is obtained as output.

[0769] Step 7:

[0770] The server sends a response message to the smartphone. The final response message is sent from the server to the smartphone. The final response message is given as input, and the output is displayed on the smartphone.

[0771] Step 8:

[0772] The smartphone displays the response message to the user. The smartphone displays the response message received from the server to the user. It also provides a voice reading function if necessary. The final response message is given as input, and it is displayed to the user as output.

[0773] 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.

[0774] This invention incorporates a function that recognizes the user's emotions and adjusts the response accordingly, in addition to analyzing user-inputted questions and providing relevant information quickly. Its specific form is described below.

[0775] User Interface (UI)

[0776] Terminal: Provides an interface for users to query information. This includes text boxes for text input and voice input buttons. In the case of voice input, a speech recognition mechanism is built in to convert speech into text.

[0777] Natural Language Processing (NLP) Engine

[0778] Server: It has a natural language processing engine that analyzes text data received from users and understands their intent. This engine consists of the following technologies:

[0779] Tokenization: Breaking down a text into individual words or phrases.

[0780] Part-of-speech tagging: Identifying the part of speech of each word.

[0781] Semantic analysis: Analyzes the meaning of the entire text to identify the user's intent.

[0782] Knowledge Base

[0783] Server: It has a large database for searching relevant information based on user intent. This database includes internal and external regulations, procedures, steps, guidelines, training, etc.

[0784] Response generation engine

[0785] Server: There is an engine that retrieves information from a knowledge base based on the results analyzed by the NLP engine and generates appropriate responses for the user. The response generation engine has the following functions:

[0786] Template Selection: Choose the appropriate template.

[0787] Information Embedding: Embed the retrieved information into the template.

[0788] Emotional Engine

[0789] Server: It has an emotion engine for recognizing emotions from user input. This engine analyzes the degree of emotion from the user's text and voice input and identifies emotions such as joy, anger, and sadness.

[0790] Action execution engine

[0791] Server: Includes an engine that takes notification or escalation actions in response to specific inquiries. For example, if a user's emotions exceed a threshold for anger or sadness, it sends a notification to a specific department.

[0792] Information provision means

[0793] Terminal: Displays response messages sent from the server to the user. These may be displayed as text or read aloud.

[0794] Specific example

[0795] Example 1: Inquiry about leave application procedures

[0796] 1. User: Types the following text: "How do I apply for leave?"

[0797] 2. Terminal: Sends user input to the server.

[0798] 3. Server: The server uses a natural language processing engine to analyze the question and identify information related to "leave application procedures."

[0799] 4. Server: Retrieves relevant information from the knowledge base and generates a response stating, "To submit a leave request, follow these steps: 1. Log in to the HR portal. 2. Click the 'Leave Request' tab. 3. Enter the required information and click the submit button."

[0800] 5. Server: The emotion engine analyzes the user's emotions. Since it is determined to be a normal question, it determines that no special action is required.

[0801] 6. Terminal: Displays the generated response to the user.

[0802] Example 2: Inquiry about unpaid wages

[0803] 1. User: "My salary hasn't been paid, who should I ask about it?" (Voice input)

[0804] 2. Terminal: Converts speech to text and sends it to the server.

[0805] 3. Server: Uses a natural language processing engine to analyze the question and identify the "contact information for inquiries regarding unpaid wages."

[0806] 4. Server: Retrieves HR department contact information from the knowledge base and generates a response stating, "Inquiries regarding unpaid wages will be escalated to the HR department contact."

[0807] 5. Server: The emotion engine analyzes the user's emotions. If the user is showing strong anger or sadness, the action execution engine sends an escalation notification to the HR department.

[0808] 6. Terminal: Display the generated response to the user and confirm that the notification has been sent.

[0809] By combining these functions, the present invention provides a system that can immediately provide appropriate information in response to user inquiries, respond in a way that takes user emotions into consideration, and quickly execute necessary actions.

[0810] The following describes the processing flow.

[0811] Detailed program processing flow

[0812] Example 1: Inquiry about leave application procedures

[0813] Step 1:

[0814] The user types the text, "How do I apply for leave?"

[0815] Step 2:

[0816] The device receives user input. If the user provides voice input, the speech recognition engine first converts the speech into text.

[0817] Step 3:

[0818] The terminal sends the received text data to the server.

[0819] Step 4:

[0820] The server's natural language processing engine analyzes the user's text. First, it tokenizes the text, then tags it with parts of speech, and finally performs semantic analysis to determine the user's intent.

[0821] Step 5:

[0822] Based on the analysis results, the server queries the knowledge base for information regarding "leave application procedures."

[0823] Step 6:

[0824] The server retrieves the relevant information from the knowledge base.

[0825] Step 7:

[0826] The server's response generation engine creates a reply message based on the information it has obtained. Specifically, it generates a message that says, "To request leave, follow these steps: 1. Log in to the HR portal. 2. Click the 'Leave Request' tab. 3. Enter the required information and click the submit button."

[0827] Step 8:

[0828] The server's emotion engine analyzes the user's input text and determines the degree of emotion. In this case, it recognizes it as a normal question and determines that there is no special emotional response.

[0829] Step 9:

[0830] The server sends the generated message to the terminal.

[0831] Step 10:

[0832] The device displays the message to the user. In the case of voice output, the text is converted to speech and read aloud.

[0833] Example 2: Inquiry about unpaid wages

[0834] Step 1:

[0835] The user inputs a voice message saying, "My salary hasn't been paid, who should I ask about it?"

[0836] Step 2:

[0837] The device receives the audio and converts it to text using a speech recognition engine.

[0838] Step 3:

[0839] The terminal sends text data to the server.

[0840] Step 4:

[0841] The server's natural language processing engine analyzes the user's text. It processes the text in the following order: tokenization, part-of-speech tagging, and semantic analysis, identifying the user's intent as "an inquiry about unpaid wages."

[0842] Step 5:

[0843] The server executes queries against the knowledge base based on the analysis results.

[0844] Step 6:

[0845] The server retrieves information on "contact information for inquiries regarding unpaid wages" from its knowledge base. Here, it retrieves information on the HR department contact person.

[0846] Step 7:

[0847] The server's response generation engine creates a message stating, "Inquiries regarding unpaid wages will be escalated to the HR department."

[0848] Step 8:

[0849] The server's emotion engine analyzes the user's input text to determine the degree of emotion. If the user indicates strong anger or sadness, the action execution engine sends an escalation notification to the HR department.

[0850] Step 9:

[0851] The server generates an escalation notification and sends it to the HR department.

[0852] Step 10:

[0853] The server sends the generated message to the terminal.

[0854] Step 11:

[0855] The device displays the message to the user and confirms that the notification has been sent. In the case of voice output, it converts the text to speech and reads it aloud.

[0856] In this way, we can provide appropriate information in response to user questions, respond to their emotions, and quickly take necessary actions.

[0857] (Example 2)

[0858] 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".

[0859] Traditional question-answering systems only provide information in response to user questions, failing to adequately improve user satisfaction by not considering user emotions or escalating problems. Furthermore, the processing and analysis of voice-input questions were sometimes inefficient. This made it difficult for users to obtain quick and appropriate answers, and particularly led to delays in responses when users were emotionally agitated.

[0860] The identification processing performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving a question entered by the user, natural language processing means for analyzing the received question and searching for relevant information, means for obtaining information from a database based on the analysis results, response generation means for providing the obtained information to the user, and emotion recognition means for analyzing emotions from the user's input. This makes it possible to provide quick and appropriate information in response to the user's question, and to automatically perform responses that take the user's emotions into consideration and necessary escalations.

[0861] A "user" refers to an individual or group that enters a question into the system.

[0862] "Means of receiving questions" refers to a device or software module that has the functionality to receive text or audio input from a user.

[0863] "Natural language processing means" refers to all technologies used to analyze received questions and understand the user's intent.

[0864] "Means of obtaining information from a database" refers to a device or software module that has the function of searching for and extracting relevant information from a database based on an analyzed query.

[0865] "Response generation means" refers to a device or software module that has the function of creating a response to be provided to the user based on acquired information.

[0866] "Emotion recognition means" refers to a device or software module that has the function of identifying emotions from user input and analyzing their degree.

[0867] "Action execution means" refers to a device or software module that has the functionality to perform notifications or escalations based on the detected user's emotions.

[0868] "Speech recognition means" refers to a device or software module that has the function of analyzing data input as speech and converting its content into text format.

[0869] This invention is a system that analyzes user-inputted questions and quickly provides relevant information. Furthermore, it incorporates a function to recognize the user's emotions and adjust its response accordingly. An embodiment of this system is described in detail below.

[0870] User Interface (UI)

[0871] Terminal: Provides an interface for users to query information. This interface includes text boxes for text input and buttons for voice input. In the case of voice input, a speech recognition mechanism is built in to convert the voice data into text.

[0872] Natural Language Processing (NLP) Engine

[0873] Server: It has a natural language processing engine that analyzes text data sent by the user and understands the user's intent. This engine consists of the following technologies:

[0874] Tokenization: Breaking down a text into individual words or phrases.

[0875] Part-of-speech tagging: Identifying the part of speech of each word.

[0876] Semantic analysis: Analyzes the meaning of the entire text to identify the user's intent.

[0877] Knowledge Base

[0878] Server: It has a large database for searching relevant information based on user intent. This database contains information such as internal and external regulations, procedures, steps, guidance, and training.

[0879] Response generation engine

[0880] Server: There is an engine that retrieves information from a knowledge base based on the results analyzed by the NLP engine and generates appropriate responses for the user. This response generation engine has the following functions:

[0881] Template Selection: Select the appropriate template.

[0882] Information Embedding: Embed the retrieved information into the template.

[0883] Emotional Engine

[0884] Server: It has an emotion engine for recognizing emotions from user input. This engine analyzes the degree of emotion from the user's text and voice input and identifies emotions such as joy, anger, and sadness.

[0885] Action execution engine

[0886] Server: Includes an engine that takes notification or escalation actions in response to specific inquiries. For example, if a user's emotions exceed a threshold for anger or sadness, it sends a notification to a specific department.

[0887] Information provision means

[0888] Terminal: Displays the response message sent from the server to the user. This response may be displayed in text format or read aloud.

[0889] Specific example

[0890] Example 1: Inquiry about leave application procedures

[0891] 1. User: Types the following text: "How do I apply for leave?"

[0892] 2. Terminal: Sends user input to the server.

[0893] 3. Server: The NLP engine analyzes the question and identifies information related to "leave application procedures".

[0894] 4. Server: Retrieves relevant information from the knowledge base and generates a response stating, "To submit a leave request, follow these steps: 1. Log in to the portal. 2. Click the 'Leave Request' tab. 3. Enter the required information and click the submit button."

[0895] 5. Server: The emotion engine analyzes the user's emotions. Since it is determined to be a normal question, it determines that no special action is required.

[0896] 6. Terminal: Displays the generated response to the user.

[0897] Example 2: Inquiry about unpaid wages

[0898] 1. User: "My salary hasn't been paid. Who should I contact?" (Voice input)

[0899] 2. Terminal: Converts speech to text and sends it to the server.

[0900] 3. Server: The NLP engine analyzes the question and identifies the "contact point for inquiries regarding unpaid wages."

[0901] 4. Server: Retrieves information on the HR department contact person from the knowledge base and generates a response stating, "Please direct inquiries regarding unpaid wages to the HR department contact person."

[0902] 5. Server: The emotion engine analyzes the user's emotions. If the user is showing strong anger or sadness, the action execution engine sends an escalation notification to the HR department.

[0903] 6. Terminal: Display the generated response to the user and confirm that the notification has been sent.

[0904] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0905] Step 1:

[0906] User: The user enters their question through the interface. For example, they might type "Please tell me about the company's benefits" into the text box. Alternatively, using voice input, they might say into the microphone, "My salary hasn't been paid, who should I contact?" The input is sent to the terminal as text or voice data.

[0907] Step 2:

[0908] Terminal: Receives user input, and if the input is voice data, converts the voice data into text data using speech recognition. This converted text data, or the original text data, is sent to the server. The input is the original user's question, and the output is text data.

[0909] Step 3:

[0910] Server: Receives text data and analyzes it using a natural language processing (NLP) engine. Specifically, it first tokenizes the input sentence, breaking it down into words and phrases. Then, it performs part-of-speech tagging to identify the grammatical role of each word. Next, it performs semantic analysis to analyze the meaning of the entire sentence and identify the user's intent. The input is text data, and the output is analyzed intent data.

[0911] Step 4:

[0912] Server: Based on intent data analyzed by the NLP engine, it searches for relevant information from a knowledge base. This knowledge base includes internal and external regulations, procedures, steps, and guidelines. It executes database queries within the knowledge base to retrieve appropriate information. The input is the analyzed intent data, and the output is the retrieved information.

[0913] Step 5:

[0914] Server: Uses a response generation engine to generate a response to provide to the user based on the information obtained. First, it selects an appropriate template, and then it embeds the obtained information into the template to form the final response. The input is the information obtained, and the output is the generated response text.

[0915] Step 6:

[0916] Server: Uses an emotion engine to analyze the user's emotions from the generated response and the user's input text. The emotion engine analyzes specific words, phrases, and tones in the user's text to identify emotions such as joy, anger, and sadness. The input is the user's input text and the generated response, and the output is emotion data.

[0917] Step 7:

[0918] Server: Based on the emotion data identified by the emotion engine, the action execution engine performs specific actions. For example, if a user's emotion exceeds the threshold for anger or sadness, a notification is sent to a specific department. The input is emotion data, and the output is an escalation notification.

[0919] Step 8:

[0920] Terminal: Receives response messages sent from the server and displays them to the user. This response is displayed in text format or read aloud. Specifically, text is displayed on the terminal's display, or audio is played through the speaker. The input is the response from the server, and the output is the display or audio output to the user.

[0921] (Application Example 2)

[0922] 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."

[0923] In autonomous vehicles, there is a need for a system that can respond quickly and appropriately to passengers' questions, and furthermore, respond flexibly according to the passengers' emotional state. However, conventional systems have had difficulty realizing responses that take passengers' emotions into account or escalation processing based on specific emotional states. As a result, passenger dissatisfaction and problems may not be properly resolved. To solve these problems, this invention provides a new system that combines natural language processing and sentiment analysis.

[0924] 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.

[0925] In this invention, the server includes means for receiving user input, means for natural language processing for analyzing user input, means for obtaining information from a knowledge base based on the analysis results, means for generating a response that provides the obtained information to the user, means for analyzing emotions from user input, and means for executing necessary actions based on the analyzed emotions. This makes it possible to answer user questions quickly and appropriately, and to respond flexibly based on the user's emotional state.

[0926] "Means for receiving user-entered questions" refers to an interface for receiving questions entered by the user in text or voice.

[0927] "Natural language processing means" are technical means for analyzing received text data and understanding its intent.

[0928] "Means of obtaining information from a knowledge base" refers to a function that extracts necessary information from relevant databases based on the analysis results.

[0929] A "response generation means" is an engine that generates answers to be provided to the user based on the acquired information.

[0930] An "emotion analysis tool" is a system that identifies emotions from user input text or voice and analyzes their emotional state.

[0931] An "action execution mechanism" is a function that performs a specific action (such as notification or escalation) based on the analyzed emotional state.

[0932] "Speech recognition means" refers to technical means for converting speech input into text data.

[0933] "Means of providing information to the user" refers to an interface for presenting the generated response to the user in text or audio format.

[0934] The present invention is implemented as a system for responding quickly and appropriately to passenger questions in autonomous vehicles. Specifically, it has the function of analyzing questions entered by the user, obtaining appropriate information from a knowledge base, providing the generated response to the user, and also analyzing the user's emotional state and taking necessary actions based on that analysis.

[0935] Hardware and software configuration

[0936] This system consists of the following hardware and software.

[0937] hardware

[0938] Server: Provides the primary computing resources for backend processing.

[0939] Terminal: Infotainment system within an autonomous vehicle. Includes user interfaces such as touchscreens and voice recognition microphones.

[0940] software

[0941] Natural Language Processing Engine (NLP Engine): Uses the Hugging Face natural language processing model. It has the capability to analyze questions using the transformers library.

[0942] Emotion analysis engine: Also using Hugging Face's pipeline, it identifies the user's emotional state.

[0943] Knowledge base: A large-scale database containing vehicle functions, operational information, etc.

[0944] Response generation engine: Generates responses to the user based on the acquired information.

[0945] Action Execution Engine: Based on sentiment analysis results, it executes actions (notifications or escalations) based on specific emotional states.

[0946] System operation

[0947] 1. Receiving and analyzing user input

[0948] The user enters a question via voice or text through the vehicle's infotainment system. In the case of voice input, a speech recognition system converts it into text. The converted text is sent to a server, where a natural language processing engine analyzes the intent of the question.

[0949] 2. Information acquisition from knowledge bases

[0950] Based on the results analyzed by the natural language processing engine, the server searches for and retrieves relevant information from its knowledge base. This information covers a wide range of topics, including vehicle operation methods and route information.

[0951] 3. Response generation and sentiment analysis

[0952] Based on the acquired information, the response generation engine generates an appropriate answer. Simultaneously, the sentiment analysis engine analyzes the user's emotional state from their questions and other inputs, and determines the analysis result.

[0953] 4. Performing a specific action

[0954] If the emotion analysis results are "negative," for example, if anger or frustration is strong, the action execution engine will send a notification to a specific department or service center. This notification is a measure to ensure that appropriate action is taken quickly.

[0955] Specific example

[0956] Example question: "Please tell me how to use the air conditioner."

[0957] Analysis results and response: The NLP engine analyzes the question, retrieves information about "how to use the air conditioner" from the knowledge base, and generates the response, "To operate the air conditioner, tap the 'Air Conditioner' tab on the central touchscreen."

[0958] Sentiment Analysis and Action: The sentiment analysis engine determines the user's emotion to be "normal." In this case, no special action is required.

[0959] Example question: "I want to change the background music in my car, how do I do that?"

[0960] Analysis results and response: The NLP engine analyzes the question, retrieves information about "how to change background music" from the knowledge base, and generates the response: "Click the menu icon at the top of the central touchscreen -> 'Settings' -> 'Audio Settings' -> Select from 'Playlist'."

[0961] Thus, the present invention provides a system for autonomous vehicles that can respond quickly to passengers' questions and further provide flexible responses according to their emotional state.

[0962] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0963] Step 1:

[0964] The terminal receives user input.

[0965] The user enters their question into the vehicle's infotainment system via voice or text.

[0966] In the case of voice input, a speech recognition system converts the speech into text and generates text data. The entered text is then sent from the terminal to the server.

[0967] Step 2:

[0968] The server analyzes the question.

[0969] The server uses a natural language processing engine to analyze the received text data and identify the intent of the input. This process involves tokenization, part-of-speech tagging, and semantic analysis to understand the structure and content of the user's question. The input is the user's question text, and the output is the intent of the question and important keywords.

[0970] Step 3:

[0971] The server retrieves information from the knowledge base.

[0972] The server searches for relevant information from a knowledge base based on the analysis results of the natural language processing engine. The knowledge base includes vehicle operation instructions and route information, and appropriate data is extracted. The input is the analysis result, and the output is the data of the relevant information.

[0973] Step 4:

[0974] The server generates a response.

[0975] The server uses a response generation engine to generate answers for the user based on the acquired information. The generated answers are created by embedding information into a template. The input is information acquired from the knowledge base, and the output is the answer text provided to the user.

[0976] Step 5:

[0977] The server performs sentiment analysis.

[0978] The server uses an emotion analysis engine to identify emotions from the user's input text. Here, the emotional state the user was in when asking a question (e.g., joy, anger, sadness, etc.) is determined. The input is the user's question text, and the output is the result of the emotional state analysis.

[0979] Step 6:

[0980] The server performs the necessary actions.

[0981] The server performs actions based on the sentiment analysis results for specific emotional states. For example, if a user is highly angry, the action execution engine sends a notification to a specific department or person in charge. The input is the sentiment analysis result, and the output is the result of the action (such as sending a notification).

[0982] Step 7:

[0983] The device provides information to the user.

[0984] The response text generated by the server is sent to the terminal and presented to the user. It may be displayed in text or audio format. The input is the response text from the server, and the output is information provided to the passenger.

[0985] 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.

[0986] 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.

[0987] 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.

[0988] [Third Embodiment]

[0989] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0990] 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.

[0991] 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).

[0992] 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.

[0993] 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.

[0994] 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).

[0995] 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.

[0996] 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.

[0997] 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.

[0998] 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.

[0999] 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.

[1000] 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".

[1001] This invention is a system that analyzes questions entered by a user and quickly provides the relevant information. Its specific form is described below.

[1002] User Interface (UI)

[1003] Terminal: Provides an interface for users to query information. This includes text boxes for text input and voice input buttons. In the case of voice input, a speech recognition mechanism is built in to convert speech into text.

[1004] Natural Language Processing (NLP) Engine

[1005] Server: It has a natural language processing engine that analyzes text data received from users and understands their intent. This engine consists of the following technologies:

[1006] Tokenization: Breaking down a text into individual words or phrases.

[1007] Part-of-speech tagging: Identifying the part of speech of each word.

[1008] Semantic analysis: Analyzes the meaning of the entire text to identify the user's intent.

[1009] Knowledge Base

[1010] Server: It has a large database for searching relevant information based on user intent. This database includes internal and external regulations, procedures, steps, guidelines, training, etc.

[1011] Response generation engine

[1012] Server: There is an engine that retrieves information from a knowledge base based on the results analyzed by the NLP engine and generates appropriate responses for the user. The response generation engine has the following functions:

[1013] Template Selection: Choose the appropriate template.

[1014] Information Embedding: Embed the retrieved information into the template.

[1015] Action execution engine

[1016] Server: Includes an engine that takes notification or escalation actions in response to specific inquiries. For example, if a user's inquiry is about unpaid wages, it sends a notification to the HR department.

[1017] Information provision means

[1018] Terminal: Displays response messages sent from the server to the user. These may be displayed as text or read aloud.

[1019] Specific example

[1020] Example 1: Inquiry about leave application procedures

[1021] 1. User: Types the following text: "How do I apply for leave?"

[1022] 2. Terminal: Sends user input to the server.

[1023] 3. Server: The server uses a natural language processing engine to analyze the question and identify information related to "leave application procedures."

[1024] 4. Server: Retrieves relevant information from the knowledge base and generates a response stating, "To submit a leave request, follow these steps: 1. Log in to the HR portal. 2. Click the 'Leave Request' tab. 3. Enter the required information and click the submit button."

[1025] 5. Terminal: Displays the generated response to the user.

[1026] Example 2: Inquiry about unpaid wages

[1027] 1. User: "My salary hasn't been paid, who should I ask about it?" (Voice input)

[1028] 2. Terminal: Converts speech to text and sends it to the server.

[1029] 3. Server: Uses a natural language processing engine to analyze the question and identify the "contact information for inquiries regarding unpaid wages."

[1030] 4. Server: Retrieves HR department contact information from the knowledge base and generates a response stating, "Inquiries regarding unpaid wages will be escalated to the HR department contact."

[1031] 5. Server: Sends an escalation notification to the HR department.

[1032] 6. Terminal: Display the generated response to the user and confirm that the notification has been sent.

[1033] This invention provides a system that, by combining these functions, can instantly provide appropriate information in response to a variety of questions entered by the user and quickly process the necessary actions.

[1034] The following describes the processing flow.

[1035] Detailed program processing flow

[1036] Example 1: Inquiry about leave application procedures

[1037] Step 1:

[1038] The user types the text, "How do I apply for leave?"

[1039] Step 2:

[1040] The device receives user input. If the user provides voice input, it is first converted to text by the speech recognition engine.

[1041] Step 3:

[1042] The terminal sends the received text data to the server.

[1043] Step 4:

[1044] The server's natural language processing engine analyzes the user's text. First, it tokenizes the text, then tags it with parts of speech, and finally performs semantic analysis to determine the user's intent.

[1045] Step 5:

[1046] Based on the analysis results, the server queries the knowledge base for information regarding "leave application procedures."

[1047] Step 6:

[1048] The server retrieves the relevant information from the knowledge base.

[1049] Step 7:

[1050] The server's response generation engine creates a reply message based on the information it has obtained. Specifically, it generates a message that says, "To request leave, follow these steps: 1. Log in to the HR portal. 2. Click the 'Leave Request' tab. 3. Enter the required information and click the submit button."

[1051] Step 8:

[1052] The server sends the generated message to the terminal.

[1053] Step 9:

[1054] The device displays the message to the user. In the case of voice output, the text is converted to speech and read aloud.

[1055] Example 2: Inquiry about unpaid wages

[1056] Step 1:

[1057] The user inputs a voice message saying, "My salary hasn't been paid, who should I ask about it?"

[1058] Step 2:

[1059] The device receives the audio and converts it to text using a speech recognition engine.

[1060] Step 3:

[1061] The terminal sends text data to the server.

[1062] Step 4:

[1063] The server's natural language processing engine analyzes the user's text. It processes the text in the following order: tokenization, part-of-speech tagging, and semantic analysis, identifying the user's intent as "an inquiry about unpaid wages."

[1064] Step 5:

[1065] The server executes queries against the knowledge base based on the analysis results.

[1066] Step 6:

[1067] The server retrieves information on "contact information for inquiries regarding unpaid wages" from its knowledge base. Here, it retrieves information on the HR department contact person.

[1068] Step 7:

[1069] The server's response generation engine creates a message stating, "Inquiries regarding unpaid wages will be escalated to the HR department."

[1070] Step 8:

[1071] The server's action execution engine generates an escalation notification and sends it to the HR department.

[1072] Step 9:

[1073] The server sends the generated message to the terminal.

[1074] Step 10:

[1075] The device displays messages to the user and reads them aloud if necessary.

[1076] These steps enable us to provide immediate and accurate answers to user inquiries and take necessary actions.

[1077] (Example 1)

[1078] 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."

[1079] In today's world, systems that allow users to obtain information quickly and accurately are crucial. However, many systems lack the ability to accurately understand user questions and fail to provide appropriate information. Furthermore, while notifications and escalations are necessary in certain situations, few systems automate these processes. Therefore, there is a need to develop systems that can efficiently respond to user questions and take necessary actions.

[1080] 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.

[1081] In this invention, the server includes means for receiving questions entered by the user, natural language processing means for analyzing the questions and searching for relevant information, means for obtaining information from a knowledge base based on the analysis results, response generation means for providing the obtained information to the user, and action execution means for performing notifications and escalations based on the user's questions. This makes it possible to provide information quickly and accurately in response to a wide range of user questions and to automatically perform necessary actions.

[1082] A "user" is an individual or group that uses the system to input questions and obtain information.

[1083] A "question" is a text or voice message that a user enters to ask the system for information.

[1084] A "terminal" refers to a device (such as a PC, smartphone, or tablet) that a user uses to input questions and receive responses from a server.

[1085] A "server" is a central processing unit that analyzes questions received from users and searches for, retrieves, and provides the necessary information.

[1086] "Natural language processing methods" refer to technologies for analyzing text data received from users and understanding its intent. Specifically, this includes tokenization, part-of-speech tagging, and semantic analysis.

[1087] A "tokenization method" is a technique that breaks down a question into words or phrases.

[1088] A "part-of-speech tagging method" is a technique for identifying the part of speech of each word.

[1089] "Semantic analysis means" refers to technologies that analyze the meaning of an entire text and identify the user's intent.

[1090] A "knowledge base" is a database that stores information to provide appropriate answers to user questions.

[1091] A "response generation method" is a technology that generates answers to users based on information obtained from a knowledge base. Specifically, it includes template selection and information embedding.

[1092] "Template selection method" refers to the technique of selecting the appropriate response template.

[1093] "Information embedding means" refers to a technology that generates a final response by embedding the necessary information into a selected template.

[1094] An "action execution method" is a technology that automatically performs notifications or escalations in response to specific inquiries.

[1095] "Voice recognition means" refers to technology that converts voice input from a user into text.

[1096] A "notification" is a message that informs specific recipients of important information.

[1097] "Escalation" refers to the procedure of transferring a specific problem or inquiry to a senior officer or another department.

[1098] This invention relates to a system that analyzes user-inputted questions and quickly provides relevant information. This system provides information quickly and accurately in response to a wide range of user questions and automatically performs necessary actions.

[1099] User Interface (UI)

[1100] The terminal provides an interface for users to query information. The terminal includes text boxes for text input and voice input buttons, providing means for users to input questions. In the case of voice input, speech recognition technology is used to convert speech into text. Specifically, devices such as PCs, smartphones, and tablets are used.

[1101] Natural Language Processing (NLP) Engine

[1102] The server has a natural language processing engine that analyzes text data received from users and understands their intent. This engine uses tools such as Google Cloud Natural Language API and Amazon Comprehend to perform the following processes:

[1103] Tokenization: Breaking down a question into words or phrases.

[1104] Part-of-speech tagging: Identifying the part of speech of each word.

[1105] Semantic analysis: Analyzes the meaning of the entire question to identify the user's intent.

[1106] Knowledge Base

[1107] The server has a database for searching for relevant information based on the user's intent. This database contains information such as internal and external regulations, procedures, steps, guidelines, and training. Specific examples of databases used include MongoDB and MySQL.

[1108] Response generation engine

[1109] The server has an engine that retrieves information from a knowledge base based on the results analyzed by the NLP engine and generates appropriate responses for the user. This response generation engine has the following functions:

[1110] Template Selection: Choose the appropriate template.

[1111] Information Embedding: Embed retrieved information into the template. Specifically, use a template engine such as Mustache or Handlebars.

[1112] Action execution engine

[1113] The server includes an engine that performs notification and escalation actions in response to specific inquiries. For example, if there is an inquiry about unpaid wages, it will send a notification to the HR department. Notifications may use the Slack API or an SMTP server for sending emails.

[1114] Information provision means

[1115] The terminal displays the response message sent from the server to the user. The display method is either text display or text-to-speech.

[1116] Specific example

[1117] Example 1: Inquiry about leave application procedures

[1118] 1. The user types the text, "How do I apply for leave?"

[1119] 2. The terminal sends user input to the server.

[1120] 3. The server uses a natural language processing engine to analyze the question and identify information related to "leave application procedures."

[1121] 4. The server retrieves the relevant information from the knowledge base and generates a response similar to the following: "To submit a leave request, follow these steps: 1. Log in to the HR portal. 2. Click the 'Leave Request' tab. 3. Enter the required information and click the submit button."

[1122] 5. The terminal displays this response to the user.

[1123] Example 2: Inquiry about unpaid wages

[1124] 1. The user voice-inputs, "My salary hasn't been paid, who should I ask about it?"

[1125] 2. The device converts the audio to text using the Google Speech-to-Text API and sends it to the server.

[1126] 3. The server analyzes the question using a natural language processing engine and identifies the "contact information for inquiries regarding unpaid wages."

[1127] 4. The server retrieves information about the HR department contact person from its knowledge base and generates a response stating, "Inquiries regarding unpaid wages will be escalated to the HR department contact person."

[1128] 5. The server uses the Slack API to send an escalation notification to the HR department.

[1129] 6. The device displays the generated response to the user and confirms that the notification has been sent.

[1130] Example of a prompt:

[1131] "Could you please explain the procedure for requesting leave?"

[1132] "Please tell me how to deal with unpaid wages."

[1133] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1134] Step 1:

[1135] The user enters a question into the terminal's interface.

[1136] Input methods include text input or voice input.

[1137] In terms of specific actions, the user either types "How do I go about applying for leave?" into the text box or presses the voice input button to speak the question.

[1138] The entered text or audio will be sent to the next step.

[1139] Step 2:

[1140] The terminal sends the user's text input or voice to the server.

[1141] In the case of voice input, the device uses speech recognition to convert the speech into text.

[1142] Input: Text or audio data entered by the user.

[1143] Data processing: Speech-to-text conversion using speech recognition technology.

[1144] Output: User questions in text format

[1145] Specifically, the process involves converting speech to text using APIs such as the Google Speech-to-Text API, and then sending the converted text to a server over the network.

[1146] Step 3:

[1147] The server analyzes the received text data using a natural language processing (NLP) engine.

[1148] The NLP engine performs tokenization, part-of-speech tagging, and semantic analysis.

[1149] Input: A text-formatted question sent from the device.

[1150] Data processing:

[1151] Tokenization: Breaking down a question into words or phrases.

[1152] Part-of-speech tagging: Identify the part of speech of each token.

[1153] Semantic analysis: Interpret the meaning of the entire question and identify the user's intent.

[1154] Output: Structured data indicating user intent

[1155] Specifically, the system uses Google Cloud Natural Language API and Amazon Comprehend to analyze the questions.

[1156] Step 4:

[1157] The server searches for relevant information from its knowledge base based on the results analyzed by the NLP engine.

[1158] Input: Structured data indicating user intent

[1159] Data processing: Executing queries against knowledge bases (such as MongoDB or MySQL)

[1160] Output: Appropriate information in response to the user's question

[1161] Specifically, the process involves executing a database query to retrieve information regarding "leave application procedures."

[1162] Step 5:

[1163] The server's response generation engine generates an appropriate answer based on the information it has received.

[1164] Input: Information obtained from a knowledge base

[1165] Data processing:

[1166] Template Selection: Select the appropriate response template.

[1167] Information Embedding: Embed the retrieved information into the template.

[1168] Output: Final response message to the user

[1169] Specifically, the system generates response messages using template engines such as Mustache or Handlebars.

[1170] Step 6:

[1171] The server performs the necessary actions (notifications or escalation) in response to specific queries.

[1172] Input: User's questions and analysis results

[1173] Data processing: Creating and sending notification messages

[1174] Output: Escalation notifications and reports

[1175] Specifically, the system uses the Slack API to send escalation notifications to the HR department.

[1176] Step 7:

[1177] The terminal displays the response sent from the server to the user.

[1178] The display method is either text display or text-to-speech.

[1179] Input: Response message from the server

[1180] Data processing: Display in a user-friendly format.

[1181] Output: Response message displayed to the user

[1182] Specifically, the user interface will display the following message: "To request leave, follow these steps: 1. Log in to the HR portal. 2. Click the 'Leave Request' tab. 3. Enter the required information and click the submit button."

[1183] (Application Example 1)

[1184] 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."

[1185] Modern users demand interactive and personalized information. However, traditional systems have struggled to effectively analyze user input and quickly deliver the most relevant content to individual users. Furthermore, the rapid increase in video content has led to problems with users spending a lot of time finding the right videos. This could potentially lead to decreased user satisfaction and engagement.

[1186] 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.

[1187] In this invention, the server includes means for receiving a question entered by the user, means for analyzing the received question and searching for relevant information, means for obtaining information from a knowledge base based on the analysis results, means for generating a response to provide the obtained information to the user, and means for providing the user with personalized video recommendations. This makes it possible to quickly and efficiently recommend relevant video content based on the question entered by the user, thereby improving the user experience.

[1188] "Means for receiving user-entered questions" refers to an interface that allows users to input questions into the system via text boxes or voice input and send those questions to the server in digital format.

[1189] "Natural language processing means for searching for relevant information" refers to technology that analyzes text data received from users to understand their intent and search for appropriate information or content.

[1190] A "knowledge base" is a database that stores a large amount of related information and data, and is used to quickly provide appropriate information in response to user questions.

[1191] A "response generation means" is a means for generating an appropriate answer to provide to the user based on the analysis results and information obtained from a knowledge base.

[1192] "Means of providing personalized video recommendations to users" refers to methods for selecting and presenting the most suitable video content for each individual user based on their questions and past behavioral data.

[1193] "Action execution means for sending notifications and escalating" refers to means for taking appropriate actions, such as sending notifications to relevant departments or personnel, based on the content of a question from a specific user.

[1194] "Voice recognition means" refers to a technology for converting questions entered by a user via voice into text format, and is a means for analyzing voice signals and converting their content into character data.

[1195] Modes for carrying out the invention

[1196] The present invention is a system that personalizedly recommends relevant information and video content based on questions entered by the user. Specific embodiments for carrying out the present invention are described below.

[1197] Hardware and software to use

[1198] hardware

[1199] Smartphone: A device used by users to input questions and receive responses.

[1200] Server: A device that performs data processing and manages the knowledge base.

[1201] software

[1202] Natural Language Processing Engine (NLP): For example, the Google Cloud Natural Language API. It analyzes user text and voice input.

[1203] Video recommendation algorithms: For example, APIs from video streaming services (YouTube API) or proprietary recommendation systems. These recommend appropriate videos to users.

[1204] Database: For example, MySQL or MongoDB. Used as a knowledge base.

[1205] Data processing and data calculation

[1206] smartphone

[1207] The smartphone provides a text box or voice input button for the user to enter a question. This input is sent to the server via an API. In the case of voice input, a speech recognition mechanism is used to convert the speech to text.

[1208] server

[1209] The server will perform the following actions:

[1210] Natural language processing engines analyze text data received from users to understand their intent. This includes tokenization, part-of-speech tagging, and semantic analysis.

[1211] The knowledge base searches for appropriate information and video content based on the results analyzed by the natural language processing engine.

[1212] The response generation engine generates appropriate answers for the user based on information retrieved from a knowledge base. This includes video links and related information.

[1213] The video recommendation algorithm provides personalized video recommendations based on the user's past viewing history and feedback.

[1214] Smartphone (response display)

[1215] The smartphone displays the response received from the server to the user. A text-to-speech function is also provided as needed.

[1216] Specific examples and prompt statements

[1217] Example 1: Desire for knowledge

[1218] The user types "I want to learn more about artificial intelligence." The system responds as follows:

[1219] User: "I want to learn more about artificial intelligence."

[1220] system:

[1221] I am searching for information about artificial intelligence...

[1222] I recommend this video:

[1223] 1. Fundamentals of Artificial Intelligence

[1224] 2. Latest AI technology

[1225] 3. The Future of AI

[1226] Specific example 2: The desire for entertainment

[1227] The user types "Tell me about some popular movies lately." The system responds as follows:

[1228] User: "Tell me about some popular movies lately."

[1229] system:

[1230] I'm searching for recently popular movies...

[1231] These movies are popular:

[1232] 1. Trailer for the movie 'XYZ'

[1233] 2. Review of the movie 'ABC'

[1234] 3. Interview about the film '123'

[1235] By combining these components, this invention provides users with instantly appropriate information and video content in response to a variety of questions they enter, thereby improving the user experience.

[1236] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1237] Step 1:

[1238] The user enters a question. The user enters the question using the text box or voice input button on their smartphone. The smartphone receives the entered text data (or voice data) and sends it to the server. The user's question is given as input and sent to the server as output.

[1239] Step 2:

[1240] The server receives input data from the user. The server receives the input text data or audio data, and if it is audio data, it converts it to text data using speech recognition. Text data (or converted text data) is given as input, and text data for analysis is obtained as output.

[1241] Step 3:

[1242] The server starts up a natural language processing engine and analyzes the text data. The natural language processing engine uses techniques such as tokenization, part-of-speech tagging, and semantic analysis to understand the intent of the user's question. Text data is given as input, and the result of intent analysis is obtained as output.

[1243] Step 4:

[1244] The server retrieves relevant information from the knowledge base. Based on the analysis results, the server searches for and retrieves relevant information and content from the knowledge base (database). The result of intent analysis is given as input, and relevant information is obtained as output.

[1245] Step 5:

[1246] The server uses a video recommendation algorithm to select personalized videos. The video recommendation algorithm considers the user's past viewing history and feedback to calculate data for recommending the most suitable videos. The analysis results and the user's history data are given as input, and a list of recommended videos is obtained as output.

[1247] Step 6:

[1248] The server uses a response generation engine to generate a response to provide to the user. The response generation engine generates a response to the user based on the acquired information and the recommended video list. Relevant information and the recommended video list are given as input, and the final response message is obtained as output.

[1249] Step 7:

[1250] The server sends a response message to the smartphone. The final response message is sent from the server to the smartphone. The final response message is given as input, and the output is displayed on the smartphone.

[1251] Step 8:

[1252] The smartphone displays the response message to the user. The smartphone displays the response message received from the server to the user. It also provides a voice reading function if necessary. The final response message is given as input, and it is displayed to the user as output.

[1253] 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.

[1254] This invention incorporates a function that recognizes the user's emotions and adjusts the response accordingly, in addition to analyzing user-inputted questions and providing relevant information quickly. Its specific form is described below.

[1255] User Interface (UI)

[1256] Terminal: Provides an interface for users to query information. This includes text boxes for text input and voice input buttons. In the case of voice input, a speech recognition mechanism is built in to convert speech into text.

[1257] Natural Language Processing (NLP) Engine

[1258] Server: It has a natural language processing engine that analyzes text data received from users and understands their intent. This engine consists of the following technologies:

[1259] Tokenization: Breaking down a text into individual words or phrases.

[1260] Part-of-speech tagging: Identifying the part of speech of each word.

[1261] Semantic analysis: Analyzes the meaning of the entire text to identify the user's intent.

[1262] Knowledge Base

[1263] Server: It has a large database for searching relevant information based on user intent. This database includes internal and external regulations, procedures, steps, guidelines, training, etc.

[1264] Response generation engine

[1265] Server: There is an engine that retrieves information from a knowledge base based on the results analyzed by the NLP engine and generates appropriate responses for the user. The response generation engine has the following functions:

[1266] Template Selection: Choose the appropriate template.

[1267] Information Embedding: Embed the retrieved information into the template.

[1268] Emotional Engine

[1269] Server: It has an emotion engine for recognizing emotions from user input. This engine analyzes the degree of emotion from the user's text and voice input and identifies emotions such as joy, anger, and sadness.

[1270] Action execution engine

[1271] Server: Includes an engine that takes notification or escalation actions in response to specific inquiries. For example, if a user's emotions exceed a threshold for anger or sadness, it sends a notification to a specific department.

[1272] Information provision means

[1273] Terminal: Displays response messages sent from the server to the user. These may be displayed as text or read aloud.

[1274] Specific example

[1275] Example 1: Inquiry about leave application procedures

[1276] 1. User: Types the following text: "How do I apply for leave?"

[1277] 2. Terminal: Sends user input to the server.

[1278] 3. Server: The server uses a natural language processing engine to analyze the question and identify information related to "leave application procedures."

[1279] 4. Server: Retrieves relevant information from the knowledge base and generates a response stating, "To submit a leave request, follow these steps: 1. Log in to the HR portal. 2. Click the 'Leave Request' tab. 3. Enter the required information and click the submit button."

[1280] 5. Server: The emotion engine analyzes the user's emotions. Since it is determined to be a normal question, it determines that no special action is required.

[1281] 6. Terminal: Displays the generated response to the user.

[1282] Example 2: Inquiry about unpaid wages

[1283] 1. User: "My salary hasn't been paid, who should I ask about it?" (Voice input)

[1284] 2. Terminal: Converts speech to text and sends it to the server.

[1285] 3. Server: Uses a natural language processing engine to analyze the question and identify the "contact information for inquiries regarding unpaid wages."

[1286] 4. Server: Retrieves HR department contact information from the knowledge base and generates a response stating, "Inquiries regarding unpaid wages will be escalated to the HR department contact."

[1287] 5. Server: The emotion engine analyzes the user's emotions. If the user is showing strong anger or sadness, the action execution engine sends an escalation notification to the HR department.

[1288] 6. Terminal: Display the generated response to the user and confirm that the notification has been sent.

[1289] By combining these functions, the present invention provides a system that can immediately provide appropriate information in response to user inquiries, respond in a way that takes user emotions into consideration, and quickly execute necessary actions.

[1290] The following describes the processing flow.

[1291] Detailed program processing flow

[1292] Example 1: Inquiry about leave application procedures

[1293] Step 1:

[1294] The user types the text, "How do I apply for leave?"

[1295] Step 2:

[1296] The device receives user input. If the user provides voice input, the speech recognition engine first converts the speech into text.

[1297] Step 3:

[1298] The terminal sends the received text data to the server.

[1299] Step 4:

[1300] The server's natural language processing engine analyzes the user's text. First, it tokenizes the text, then tags it with parts of speech, and finally performs semantic analysis to determine the user's intent.

[1301] Step 5:

[1302] Based on the analysis results, the server queries the knowledge base for information regarding "leave application procedures."

[1303] Step 6:

[1304] The server retrieves the relevant information from the knowledge base.

[1305] Step 7:

[1306] The server's response generation engine creates a reply message based on the information it has obtained. Specifically, it generates a message that says, "To request leave, follow these steps: 1. Log in to the HR portal. 2. Click the 'Leave Request' tab. 3. Enter the required information and click the submit button."

[1307] Step 8:

[1308] The server's emotion engine analyzes the user's input text and determines the degree of emotion. In this case, it recognizes it as a normal question and determines that there is no special emotional response.

[1309] Step 9:

[1310] The server sends the generated message to the terminal.

[1311] Step 10:

[1312] The device displays the message to the user. In the case of voice output, the text is converted to speech and read aloud.

[1313] Example 2: Inquiry about unpaid wages

[1314] Step 1:

[1315] The user inputs a voice message saying, "My salary hasn't been paid, who should I ask about it?"

[1316] Step 2:

[1317] The device receives the audio and converts it to text using a speech recognition engine.

[1318] Step 3:

[1319] The terminal sends text data to the server.

[1320] Step 4:

[1321] The server's natural language processing engine analyzes the user's text. It processes the text in the following order: tokenization, part-of-speech tagging, and semantic analysis, identifying the user's intent as "an inquiry about unpaid wages."

[1322] Step 5:

[1323] The server executes queries against the knowledge base based on the analysis results.

[1324] Step 6:

[1325] The server retrieves information on "contact information for inquiries regarding unpaid wages" from its knowledge base. Here, it retrieves information on the HR department contact person.

[1326] Step 7:

[1327] The server's response generation engine creates a message stating, "Inquiries regarding unpaid wages will be escalated to the HR department."

[1328] Step 8:

[1329] The server's emotion engine analyzes the user's input text to determine the degree of emotion. If the user indicates strong anger or sadness, the action execution engine sends an escalation notification to the HR department.

[1330] Step 9:

[1331] The server generates an escalation notification and sends it to the HR department.

[1332] Step 10:

[1333] The server sends the generated message to the terminal.

[1334] Step 11:

[1335] The device displays the message to the user and confirms that the notification has been sent. In the case of voice output, it converts the text to speech and reads it aloud.

[1336] In this way, we can provide appropriate information in response to user questions, respond to their emotions, and quickly take necessary actions.

[1337] (Example 2)

[1338] 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."

[1339] Traditional question-answering systems only provide information in response to user questions, failing to adequately improve user satisfaction by not considering user emotions or escalating problems. Furthermore, the processing and analysis of voice-input questions were sometimes inefficient. This made it difficult for users to obtain quick and appropriate answers, and particularly led to delays in responses when users were emotionally agitated.

[1340] The identification processing performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving a question entered by the user, natural language processing means for analyzing the received question and searching for relevant information, means for obtaining information from a database based on the analysis results, response generation means for providing the obtained information to the user, and emotion recognition means for analyzing emotions from the user's input. This makes it possible to provide quick and appropriate information in response to the user's question, and to automatically perform responses that take the user's emotions into consideration and necessary escalations.

[1341] A "user" refers to an individual or group that enters a question into the system.

[1342] "Means of receiving questions" refers to a device or software module that has the functionality to receive text or audio input from a user.

[1343] "Natural language processing means" refers to all technologies used to analyze received questions and understand the user's intent.

[1344] "Means of obtaining information from a database" refers to a device or software module that has the function of searching for and extracting relevant information from a database based on an analyzed query.

[1345] "Response generation means" refers to a device or software module that has the function of creating a response to be provided to the user based on acquired information.

[1346] "Emotion recognition means" refers to a device or software module that has the function of identifying emotions from user input and analyzing their degree.

[1347] "Action execution means" refers to a device or software module that has the functionality to perform notifications or escalations based on the detected user's emotions.

[1348] "Speech recognition means" refers to a device or software module that has the function of analyzing data input as speech and converting its content into text format.

[1349] This invention is a system that analyzes user-inputted questions and quickly provides relevant information. Furthermore, it incorporates a function to recognize the user's emotions and adjust its response accordingly. An embodiment of this system is described in detail below.

[1350] User Interface (UI)

[1351] Terminal: Provides an interface for users to query information. This interface includes text boxes for text input and buttons for voice input. In the case of voice input, a speech recognition mechanism is built in to convert the voice data into text.

[1352] Natural Language Processing (NLP) Engine

[1353] Server: It has a natural language processing engine that analyzes text data sent by the user and understands the user's intent. This engine consists of the following technologies:

[1354] Tokenization: Breaking down a text into individual words or phrases.

[1355] Part-of-speech tagging: Identifying the part of speech of each word.

[1356] Semantic analysis: Analyzes the meaning of the entire text to identify the user's intent.

[1357] Knowledge Base

[1358] Server: It has a large database for searching relevant information based on user intent. This database contains information such as internal and external regulations, procedures, steps, guidance, and training.

[1359] Response generation engine

[1360] Server: There is an engine that retrieves information from a knowledge base based on the results analyzed by the NLP engine and generates appropriate responses for the user. This response generation engine has the following functions:

[1361] Template Selection: Select the appropriate template.

[1362] Information Embedding: Embed the retrieved information into the template.

[1363] Emotional Engine

[1364] Server: It has an emotion engine for recognizing emotions from user input. This engine analyzes the degree of emotion from the user's text and voice input and identifies emotions such as joy, anger, and sadness.

[1365] Action execution engine

[1366] Server: Includes an engine that takes notification or escalation actions in response to specific inquiries. For example, if a user's emotions exceed a threshold for anger or sadness, it sends a notification to a specific department.

[1367] Information provision means

[1368] Terminal: Displays the response message sent from the server to the user. This response may be displayed in text format or read aloud.

[1369] Specific example

[1370] Example 1: Inquiry about leave application procedures

[1371] 1. User: Types the following text: "How do I apply for leave?"

[1372] 2. Terminal: Sends user input to the server.

[1373] 3. Server: The NLP engine analyzes the question and identifies information related to "leave application procedures".

[1374] 4. Server: Retrieves relevant information from the knowledge base and generates a response stating, "To submit a leave request, follow these steps: 1. Log in to the portal. 2. Click the 'Leave Request' tab. 3. Enter the required information and click the submit button."

[1375] 5. Server: The emotion engine analyzes the user's emotions. Since it is determined to be a normal question, it determines that no special action is required.

[1376] 6. Terminal: Displays the generated response to the user.

[1377] Example 2: Inquiry about unpaid wages

[1378] 1. User: "My salary hasn't been paid. Who should I contact?" (Voice input)

[1379] 2. Terminal: Converts speech to text and sends it to the server.

[1380] 3. Server: The NLP engine analyzes the question and identifies the "contact point for inquiries regarding unpaid wages."

[1381] 4. Server: Retrieves information on the HR department contact person from the knowledge base and generates a response stating, "Please direct inquiries regarding unpaid wages to the HR department contact person."

[1382] 5. Server: The emotion engine analyzes the user's emotions. If the user is showing strong anger or sadness, the action execution engine sends an escalation notification to the HR department.

[1383] 6. Terminal: Display the generated response to the user and confirm that the notification has been sent.

[1384] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1385] Step 1:

[1386] User: The user enters their question through the interface. For example, they might type "Please tell me about the company's benefits" into the text box. Alternatively, using voice input, they might say into the microphone, "My salary hasn't been paid, who should I contact?" The input is sent to the terminal as text or voice data.

[1387] Step 2:

[1388] Terminal: Receives user input, and if the input is voice data, converts the voice data into text data using speech recognition. This converted text data, or the original text data, is sent to the server. The input is the original user's question, and the output is text data.

[1389] Step 3:

[1390] Server: Receives text data and analyzes it using a natural language processing (NLP) engine. Specifically, it first tokenizes the input sentence, breaking it down into words and phrases. Then, it performs part-of-speech tagging to identify the grammatical role of each word. Next, it performs semantic analysis to analyze the meaning of the entire sentence and identify the user's intent. The input is text data, and the output is analyzed intent data.

[1391] Step 4:

[1392] Server: Based on intent data analyzed by the NLP engine, it searches for relevant information from a knowledge base. This knowledge base includes internal and external regulations, procedures, steps, and guidelines. It executes database queries within the knowledge base to retrieve appropriate information. The input is the analyzed intent data, and the output is the retrieved information.

[1393] Step 5:

[1394] Server: Uses a response generation engine to generate a response to provide to the user based on the information obtained. First, it selects an appropriate template, and then it embeds the obtained information into the template to form the final response. The input is the information obtained, and the output is the generated response text.

[1395] Step 6:

[1396] Server: Uses an emotion engine to analyze the user's emotions from the generated response and the user's input text. The emotion engine analyzes specific words, phrases, and tones in the user's text to identify emotions such as joy, anger, and sadness. The input is the user's input text and the generated response, and the output is emotion data.

[1397] Step 7:

[1398] Server: Based on the emotion data identified by the emotion engine, the action execution engine performs specific actions. For example, if a user's emotion exceeds the threshold for anger or sadness, a notification is sent to a specific department. The input is emotion data, and the output is an escalation notification.

[1399] Step 8:

[1400] Terminal: Receives response messages sent from the server and displays them to the user. This response is displayed in text format or read aloud. Specifically, text is displayed on the terminal's display, or audio is played through the speaker. The input is the response from the server, and the output is the display or audio output to the user.

[1401] (Application Example 2)

[1402] 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."

[1403] In autonomous vehicles, there is a need for a system that can respond quickly and appropriately to passengers' questions, and furthermore, respond flexibly according to the passengers' emotional state. However, conventional systems have had difficulty realizing responses that take passengers' emotions into account or escalation processing based on specific emotional states. As a result, passenger dissatisfaction and problems may not be properly resolved. To solve these problems, this invention provides a new system that combines natural language processing and sentiment analysis.

[1404] 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.

[1405] In this invention, the server includes means for receiving user input, means for natural language processing for analyzing user input, means for obtaining information from a knowledge base based on the analysis results, means for generating a response that provides the obtained information to the user, means for analyzing emotions from user input, and means for executing necessary actions based on the analyzed emotions. This makes it possible to answer user questions quickly and appropriately, and to respond flexibly based on the user's emotional state.

[1406] "Means for receiving user-entered questions" refers to an interface for receiving questions entered by the user in text or voice.

[1407] "Natural language processing means" are technical means for analyzing received text data and understanding its intent.

[1408] "Means of obtaining information from a knowledge base" refers to a function that extracts necessary information from relevant databases based on the analysis results.

[1409] A "response generation means" is an engine that generates answers to be provided to the user based on the acquired information.

[1410] An "emotion analysis tool" is a system that identifies emotions from user input text or voice and analyzes their emotional state.

[1411] An "action execution mechanism" is a function that performs a specific action (such as notification or escalation) based on the analyzed emotional state.

[1412] "Speech recognition means" refers to technical means for converting speech input into text data.

[1413] "Means of providing information to the user" refers to an interface for presenting the generated response to the user in text or audio format.

[1414] The present invention is implemented as a system for responding quickly and appropriately to passenger questions in autonomous vehicles. Specifically, it has the function of analyzing questions entered by the user, obtaining appropriate information from a knowledge base, providing the generated response to the user, and also analyzing the user's emotional state and taking necessary actions based on that analysis.

[1415] Hardware and software configuration

[1416] This system consists of the following hardware and software.

[1417] hardware

[1418] Server: Provides the primary computing resources for backend processing.

[1419] Terminal: Infotainment system within an autonomous vehicle. Includes user interfaces such as touchscreens and voice recognition microphones.

[1420] software

[1421] Natural Language Processing Engine (NLP Engine): Uses the Hugging Face natural language processing model. It has the capability to analyze questions using the transformers library.

[1422] Emotion analysis engine: Also using Hugging Face's pipeline, it identifies the user's emotional state.

[1423] Knowledge base: A large-scale database containing vehicle functions, operational information, etc.

[1424] Response generation engine: Generates responses to the user based on the acquired information.

[1425] Action Execution Engine: Based on sentiment analysis results, it executes actions (notifications or escalations) based on specific emotional states.

[1426] System operation

[1427] 1. Receiving and analyzing user input

[1428] The user enters a question via voice or text through the vehicle's infotainment system. In the case of voice input, a speech recognition system converts it into text. The converted text is sent to a server, where a natural language processing engine analyzes the intent of the question.

[1429] 2. Information acquisition from knowledge bases

[1430] Based on the results analyzed by the natural language processing engine, the server searches for and retrieves relevant information from its knowledge base. This information covers a wide range of topics, including vehicle operation methods and route information.

[1431] 3. Response generation and sentiment analysis

[1432] Based on the acquired information, the response generation engine generates an appropriate answer. Simultaneously, the sentiment analysis engine analyzes the user's emotional state from their questions and other inputs, and determines the analysis result.

[1433] 4. Performing a specific action

[1434] If the emotion analysis results are "negative," for example, if anger or frustration is strong, the action execution engine will send a notification to a specific department or service center. This notification is a measure to ensure that appropriate action is taken quickly.

[1435] Specific example

[1436] Example question: "Please tell me how to use the air conditioner."

[1437] Analysis results and response: The NLP engine analyzes the question, retrieves information about "how to use the air conditioner" from the knowledge base, and generates the response, "To operate the air conditioner, tap the 'Air Conditioner' tab on the central touchscreen."

[1438] Sentiment Analysis and Action: The sentiment analysis engine determines the user's emotion to be "normal." In this case, no special action is required.

[1439] Example question: "I want to change the background music in my car, how do I do that?"

[1440] Analysis results and response: The NLP engine analyzes the question, retrieves information about "how to change background music" from the knowledge base, and generates the response: "Click the menu icon at the top of the central touchscreen -> 'Settings' -> 'Audio Settings' -> Select from 'Playlist'."

[1441] Thus, the present invention provides a system for autonomous vehicles that can respond quickly to passengers' questions and further provide flexible responses according to their emotional state.

[1442] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1443] Step 1:

[1444] The terminal receives user input.

[1445] The user enters their question into the vehicle's infotainment system via voice or text.

[1446] In the case of voice input, a speech recognition system converts the speech into text and generates text data. The entered text is then sent from the terminal to the server.

[1447] Step 2:

[1448] The server analyzes the question.

[1449] The server uses a natural language processing engine to analyze the received text data and identify the intent of the input. This process involves tokenization, part-of-speech tagging, and semantic analysis to understand the structure and content of the user's question. The input is the user's question text, and the output is the intent of the question and important keywords.

[1450] Step 3:

[1451] The server retrieves information from the knowledge base.

[1452] The server searches for relevant information from a knowledge base based on the analysis results of the natural language processing engine. The knowledge base includes vehicle operation instructions and route information, and appropriate data is extracted. The input is the analysis result, and the output is the data of the relevant information.

[1453] Step 4:

[1454] The server generates a response.

[1455] The server uses a response generation engine to generate answers for the user based on the acquired information. The generated answers are created by embedding information into a template. The input is information acquired from the knowledge base, and the output is the answer text provided to the user.

[1456] Step 5:

[1457] The server performs sentiment analysis.

[1458] The server uses an emotion analysis engine to identify emotions from the user's input text. Here, the emotional state the user was in when asking a question (e.g., joy, anger, sadness, etc.) is determined. The input is the user's question text, and the output is the result of the emotional state analysis.

[1459] Step 6:

[1460] The server performs the necessary actions.

[1461] The server performs actions based on the sentiment analysis results for specific emotional states. For example, if a user is highly angry, the action execution engine sends a notification to a specific department or person in charge. The input is the sentiment analysis result, and the output is the result of the action (such as sending a notification).

[1462] Step 7:

[1463] The device provides information to the user.

[1464] The response text generated by the server is sent to the terminal and presented to the user. It may be displayed in text or audio format. The input is the response text from the server, and the output is information provided to the passenger.

[1465] 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.

[1466] 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.

[1467] 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.

[1468] [Fourth Embodiment]

[1469] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1470] 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.

[1471] 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).

[1472] 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.

[1473] 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.

[1474] 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).

[1475] 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.

[1476] 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.

[1477] 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.

[1478] 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.

[1479] 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.

[1480] 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.

[1481] 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".

[1482] This invention is a system that analyzes questions entered by a user and quickly provides the relevant information. Its specific form is described below.

[1483] User Interface (UI)

[1484] Terminal: Provides an interface for users to query information. This includes text boxes for text input and voice input buttons. In the case of voice input, a speech recognition mechanism is built in to convert speech into text.

[1485] Natural Language Processing (NLP) Engine

[1486] Server: It has a natural language processing engine that analyzes text data received from users and understands their intent. This engine consists of the following technologies:

[1487] Tokenization: Breaking down a text into individual words or phrases.

[1488] Part-of-speech tagging: Identifying the part of speech of each word.

[1489] Semantic analysis: Analyzes the meaning of the entire text to identify the user's intent.

[1490] Knowledge Base

[1491] Server: It has a large database for searching relevant information based on user intent. This database includes internal and external regulations, procedures, steps, guidelines, training, etc.

[1492] Response generation engine

[1493] Server: There is an engine that retrieves information from a knowledge base based on the results analyzed by the NLP engine and generates appropriate responses for the user. The response generation engine has the following functions:

[1494] Template Selection: Choose the appropriate template.

[1495] Information Embedding: Embed the retrieved information into the template.

[1496] Action execution engine

[1497] Server: Includes an engine that takes notification or escalation actions in response to specific inquiries. For example, if a user's inquiry is about unpaid wages, it sends a notification to the HR department.

[1498] Information provision means

[1499] Terminal: Displays response messages sent from the server to the user. These may be displayed as text or read aloud.

[1500] Specific example

[1501] Example 1: Inquiry about leave application procedures

[1502] 1. User: Types the following text: "How do I apply for leave?"

[1503] 2. Terminal: Sends user input to the server.

[1504] 3. Server: The server uses a natural language processing engine to analyze the question and identify information related to "leave application procedures."

[1505] 4. Server: Retrieves relevant information from the knowledge base and generates a response stating, "To submit a leave request, follow these steps: 1. Log in to the HR portal. 2. Click the 'Leave Request' tab. 3. Enter the required information and click the submit button."

[1506] 5. Terminal: Displays the generated response to the user.

[1507] Example 2: Inquiry about unpaid wages

[1508] 1. User: "My salary hasn't been paid, who should I ask about it?" (Voice input)

[1509] 2. Terminal: Converts speech to text and sends it to the server.

[1510] 3. Server: Uses a natural language processing engine to analyze the question and identify the "contact information for inquiries regarding unpaid wages."

[1511] 4. Server: Retrieves HR department contact information from the knowledge base and generates a response stating, "Inquiries regarding unpaid wages will be escalated to the HR department contact."

[1512] 5. Server: Sends an escalation notification to the HR department.

[1513] 6. Terminal: Display the generated response to the user and confirm that the notification has been sent.

[1514] This invention provides a system that, by combining these functions, can instantly provide appropriate information in response to a variety of questions entered by the user and quickly process the necessary actions.

[1515] The following describes the processing flow.

[1516] Detailed program processing flow

[1517] Example 1: Inquiry about leave application procedures

[1518] Step 1:

[1519] The user types the text, "How do I apply for leave?"

[1520] Step 2:

[1521] The device receives user input. If the user provides voice input, it is first converted to text by the speech recognition engine.

[1522] Step 3:

[1523] The terminal sends the received text data to the server.

[1524] Step 4:

[1525] The server's natural language processing engine analyzes the user's text. First, it tokenizes the text, then tags it with parts of speech, and finally performs semantic analysis to determine the user's intent.

[1526] Step 5:

[1527] Based on the analysis results, the server queries the knowledge base for information regarding "leave application procedures."

[1528] Step 6:

[1529] The server retrieves the relevant information from the knowledge base.

[1530] Step 7:

[1531] The server's response generation engine creates a reply message based on the information it has obtained. Specifically, it generates a message that says, "To request leave, follow these steps: 1. Log in to the HR portal. 2. Click the 'Leave Request' tab. 3. Enter the required information and click the submit button."

[1532] Step 8:

[1533] The server sends the generated message to the terminal.

[1534] Step 9:

[1535] The device displays the message to the user. In the case of voice output, the text is converted to speech and read aloud.

[1536] Example 2: Inquiry about unpaid wages

[1537] Step 1:

[1538] The user inputs a voice message saying, "My salary hasn't been paid, who should I ask about it?"

[1539] Step 2:

[1540] The device receives the audio and converts it to text using a speech recognition engine.

[1541] Step 3:

[1542] The terminal sends text data to the server.

[1543] Step 4:

[1544] The server's natural language processing engine analyzes the user's text. It processes the text in the following order: tokenization, part-of-speech tagging, and semantic analysis, identifying the user's intent as "an inquiry about unpaid wages."

[1545] Step 5:

[1546] The server executes queries against the knowledge base based on the analysis results.

[1547] Step 6:

[1548] The server retrieves information on "contact information for inquiries regarding unpaid wages" from its knowledge base. Here, it retrieves information on the HR department contact person.

[1549] Step 7:

[1550] The server's response generation engine creates a message stating, "Inquiries regarding unpaid wages will be escalated to the HR department."

[1551] Step 8:

[1552] The server's action execution engine generates an escalation notification and sends it to the HR department.

[1553] Step 9:

[1554] The server sends the generated message to the terminal.

[1555] Step 10:

[1556] The device displays messages to the user and reads them aloud if necessary.

[1557] These steps enable us to provide immediate and accurate answers to user inquiries and take necessary actions.

[1558] (Example 1)

[1559] 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".

[1560] In today's world, systems that allow users to obtain information quickly and accurately are crucial. However, many systems lack the ability to accurately understand user questions and fail to provide appropriate information. Furthermore, while notifications and escalations are necessary in certain situations, few systems automate these processes. Therefore, there is a need to develop systems that can efficiently respond to user questions and take necessary actions.

[1561] 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.

[1562] In this invention, the server includes means for receiving questions entered by the user, natural language processing means for analyzing the questions and searching for relevant information, means for obtaining information from a knowledge base based on the analysis results, response generation means for providing the obtained information to the user, and action execution means for performing notifications and escalations based on the user's questions. This makes it possible to provide information quickly and accurately in response to a wide range of user questions and to automatically perform necessary actions.

[1563] A "user" is an individual or group that uses the system to input questions and obtain information.

[1564] A "question" is a text or voice message that a user enters to ask the system for information.

[1565] A "terminal" refers to a device (such as a PC, smartphone, or tablet) that a user uses to input questions and receive responses from a server.

[1566] A "server" is a central processing unit that analyzes questions received from users and searches for, retrieves, and provides the necessary information.

[1567] "Natural language processing methods" refer to technologies for analyzing text data received from users and understanding its intent. Specifically, this includes tokenization, part-of-speech tagging, and semantic analysis.

[1568] A "tokenization method" is a technique that breaks down a question into words or phrases.

[1569] A "part-of-speech tagging method" is a technique for identifying the part of speech of each word.

[1570] "Semantic analysis means" refers to technologies that analyze the meaning of an entire text and identify the user's intent.

[1571] A "knowledge base" is a database that stores information to provide appropriate answers to user questions.

[1572] A "response generation method" is a technology that generates answers to users based on information obtained from a knowledge base. Specifically, it includes template selection and information embedding.

[1573] "Template selection method" refers to the technique of selecting the appropriate response template.

[1574] "Information embedding means" refers to a technology that generates a final response by embedding the necessary information into a selected template.

[1575] An "action execution method" is a technology that automatically performs notifications or escalations in response to specific inquiries.

[1576] "Voice recognition means" refers to technology that converts voice input from a user into text.

[1577] A "notification" is a message that informs specific recipients of important information.

[1578] "Escalation" refers to the procedure of transferring a specific problem or inquiry to a senior officer or another department.

[1579] This invention relates to a system that analyzes user-inputted questions and quickly provides relevant information. This system provides information quickly and accurately in response to a wide range of user questions and automatically performs necessary actions.

[1580] User Interface (UI)

[1581] The terminal provides an interface for users to query information. The terminal includes text boxes for text input and voice input buttons, providing means for users to input questions. In the case of voice input, speech recognition technology is used to convert speech into text. Specifically, devices such as PCs, smartphones, and tablets are used.

[1582] Natural Language Processing (NLP) Engine

[1583] The server has a natural language processing engine that analyzes text data received from users and understands their intent. This engine uses tools such as Google Cloud Natural Language API and Amazon Comprehend to perform the following processes:

[1584] Tokenization: Breaking down a question into words or phrases.

[1585] Part-of-speech tagging: Identifying the part of speech of each word.

[1586] Semantic analysis: Analyzes the meaning of the entire question to identify the user's intent.

[1587] Knowledge Base

[1588] The server has a database for searching for relevant information based on the user's intent. This database contains information such as internal and external regulations, procedures, steps, guidelines, and training. Specific examples of databases used include MongoDB and MySQL.

[1589] Response generation engine

[1590] The server has an engine that retrieves information from a knowledge base based on the results analyzed by the NLP engine and generates appropriate responses for the user. This response generation engine has the following functions:

[1591] Template Selection: Choose the appropriate template.

[1592] Information Embedding: Embed retrieved information into the template. Specifically, use a template engine such as Mustache or Handlebars.

[1593] Action execution engine

[1594] The server includes an engine that performs notification and escalation actions in response to specific inquiries. For example, if there is an inquiry about unpaid wages, it will send a notification to the HR department. Notifications may use the Slack API or an SMTP server for sending emails.

[1595] Information provision means

[1596] The terminal displays the response message sent from the server to the user. The display method is either text display or text-to-speech.

[1597] Specific example

[1598] Example 1: Inquiry about leave application procedures

[1599] 1. The user types the text, "How do I apply for leave?"

[1600] 2. The terminal sends user input to the server.

[1601] 3. The server uses a natural language processing engine to analyze the question and identify information related to "leave application procedures."

[1602] 4. The server retrieves the relevant information from the knowledge base and generates a response similar to the following: "To submit a leave request, follow these steps: 1. Log in to the HR portal. 2. Click the 'Leave Request' tab. 3. Enter the required information and click the submit button."

[1603] 5. The terminal displays this response to the user.

[1604] Example 2: Inquiry about unpaid wages

[1605] 1. The user voice-inputs, "My salary hasn't been paid, who should I ask about it?"

[1606] 2. The device converts the audio to text using the Google Speech-to-Text API and sends it to the server.

[1607] 3. The server analyzes the question using a natural language processing engine and identifies the "contact information for inquiries regarding unpaid wages."

[1608] 4. The server retrieves information about the HR department contact person from its knowledge base and generates a response stating, "Inquiries regarding unpaid wages will be escalated to the HR department contact person."

[1609] 5. The server uses the Slack API to send an escalation notification to the HR department.

[1610] 6. The device displays the generated response to the user and confirms that the notification has been sent.

[1611] Example of a prompt:

[1612] "Could you please explain the procedure for requesting leave?"

[1613] "Please tell me how to deal with unpaid wages."

[1614] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1615] Step 1:

[1616] The user enters a question into the terminal's interface.

[1617] Input methods include text input or voice input.

[1618] In terms of specific actions, the user either types "How do I go about applying for leave?" into the text box or presses the voice input button to speak the question.

[1619] The entered text or audio will be sent to the next step.

[1620] Step 2:

[1621] The terminal sends the user's text input or voice to the server.

[1622] In the case of voice input, the device uses speech recognition to convert the speech into text.

[1623] Input: Text or audio data entered by the user.

[1624] Data processing: Speech-to-text conversion using speech recognition technology.

[1625] Output: User questions in text format

[1626] Specifically, the process involves converting speech to text using APIs such as the Google Speech-to-Text API, and then sending the converted text to a server over the network.

[1627] Step 3:

[1628] The server analyzes the received text data using a natural language processing (NLP) engine.

[1629] The NLP engine performs tokenization, part-of-speech tagging, and semantic analysis.

[1630] Input: A text-formatted question sent from the device.

[1631] Data processing:

[1632] Tokenization: Breaking down a question into words or phrases.

[1633] Part-of-speech tagging: Identify the part of speech of each token.

[1634] Semantic analysis: Interpret the meaning of the entire question and identify the user's intent.

[1635] Output: Structured data indicating user intent

[1636] Specifically, the system uses Google Cloud Natural Language API and Amazon Comprehend to analyze the questions.

[1637] Step 4:

[1638] The server searches for relevant information from its knowledge base based on the results analyzed by the NLP engine.

[1639] Input: Structured data indicating user intent

[1640] Data processing: Executing queries against knowledge bases (such as MongoDB or MySQL)

[1641] Output: Appropriate information in response to the user's question

[1642] Specifically, the process involves executing a database query to retrieve information regarding "leave application procedures."

[1643] Step 5:

[1644] The server's response generation engine generates an appropriate answer based on the information it has received.

[1645] Input: Information obtained from a knowledge base

[1646] Data processing:

[1647] Template Selection: Select the appropriate response template.

[1648] Information Embedding: Embed the retrieved information into the template.

[1649] Output: Final response message to the user

[1650] Specifically, the system generates response messages using template engines such as Mustache or Handlebars.

[1651] Step 6:

[1652] The server performs the necessary actions (notifications or escalation) in response to specific queries.

[1653] Input: User's questions and analysis results

[1654] Data processing: Creating and sending notification messages

[1655] Output: Escalation notifications and reports

[1656] Specifically, the system uses the Slack API to send escalation notifications to the HR department.

[1657] Step 7:

[1658] The terminal displays the response sent from the server to the user.

[1659] The display method is either text display or text-to-speech.

[1660] Input: Response message from the server

[1661] Data processing: Display in a user-friendly format.

[1662] Output: Response message displayed to the user

[1663] Specifically, the user interface will display the following message: "To request leave, follow these steps: 1. Log in to the HR portal. 2. Click the 'Leave Request' tab. 3. Enter the required information and click the submit button."

[1664] (Application Example 1)

[1665] 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".

[1666] Modern users demand interactive and personalized information. However, traditional systems have struggled to effectively analyze user input and quickly deliver the most relevant content to individual users. Furthermore, the rapid increase in video content has led to problems with users spending a lot of time finding the right videos. This could potentially lead to decreased user satisfaction and engagement.

[1667] 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.

[1668] In this invention, the server includes means for receiving a question entered by the user, means for analyzing the received question and searching for relevant information, means for obtaining information from a knowledge base based on the analysis results, means for generating a response to provide the obtained information to the user, and means for providing the user with personalized video recommendations. This makes it possible to quickly and efficiently recommend relevant video content based on the question entered by the user, thereby improving the user experience.

[1669] "Means for receiving user-entered questions" refers to an interface that allows users to input questions into the system via text boxes or voice input and send those questions to the server in digital format.

[1670] "Natural language processing means for searching for relevant information" refers to technology that analyzes text data received from users to understand their intent and search for appropriate information or content.

[1671] A "knowledge base" is a database that stores a large amount of related information and data, and is used to quickly provide appropriate information in response to user questions.

[1672] A "response generation means" is a means for generating an appropriate answer to provide to the user based on the analysis results and information obtained from a knowledge base.

[1673] "Means of providing personalized video recommendations to users" refers to methods for selecting and presenting the most suitable video content for each individual user based on their questions and past behavioral data.

[1674] "Action execution means for sending notifications and escalating" refers to means for taking appropriate actions, such as sending notifications to relevant departments or personnel, based on the content of a question from a specific user.

[1675] "Voice recognition means" refers to a technology for converting questions entered by a user via voice into text format, and is a means for analyzing voice signals and converting their content into character data.

[1676] Modes for carrying out the invention

[1677] The present invention is a system that personalizedly recommends relevant information and video content based on questions entered by the user. Specific embodiments for carrying out the present invention are described below.

[1678] Hardware and software to use

[1679] hardware

[1680] Smartphone: A device used by users to input questions and receive responses.

[1681] Server: A device that performs data processing and manages the knowledge base.

[1682] software

[1683] Natural Language Processing Engine (NLP): For example, the Google Cloud Natural Language API. It analyzes user text and voice input.

[1684] Video recommendation algorithms: For example, APIs from video streaming services (YouTube API) or proprietary recommendation systems. These recommend appropriate videos to users.

[1685] Database: For example, MySQL or MongoDB. Used as a knowledge base.

[1686] Data processing and data calculation

[1687] smartphone

[1688] The smartphone provides a text box or voice input button for the user to enter a question. This input is sent to the server via an API. In the case of voice input, a speech recognition mechanism is used to convert the speech to text.

[1689] server

[1690] The server will perform the following actions:

[1691] Natural language processing engines analyze text data received from users to understand their intent. This includes tokenization, part-of-speech tagging, and semantic analysis.

[1692] The knowledge base searches for appropriate information and video content based on the results analyzed by the natural language processing engine.

[1693] The response generation engine generates appropriate answers for the user based on information retrieved from a knowledge base. This includes video links and related information.

[1694] The video recommendation algorithm provides personalized video recommendations based on the user's past viewing history and feedback.

[1695] Smartphone (response display)

[1696] The smartphone displays the response received from the server to the user. A text-to-speech function is also provided as needed.

[1697] Specific examples and prompt statements

[1698] Example 1: Desire for knowledge

[1699] The user types "I want to learn more about artificial intelligence." The system responds as follows:

[1700] User: "I want to learn more about artificial intelligence."

[1701] system:

[1702] I am searching for information about artificial intelligence...

[1703] I recommend this video:

[1704] 1. Fundamentals of Artificial Intelligence

[1705] 2. Latest AI technology

[1706] 3. The Future of AI

[1707] Specific example 2: The desire for entertainment

[1708] The user types "Tell me about some popular movies lately." The system responds as follows:

[1709] User: "Tell me about some popular movies lately."

[1710] system:

[1711] I'm searching for recently popular movies...

[1712] These movies are popular:

[1713] 1. Trailer for the movie 'XYZ'

[1714] 2. Review of the movie 'ABC'

[1715] 3. Interview about the film '123'

[1716] By combining these components, this invention provides users with instantly appropriate information and video content in response to a variety of questions they enter, thereby improving the user experience.

[1717] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1718] Step 1:

[1719] The user enters a question. The user enters the question using the text box or voice input button on their smartphone. The smartphone receives the entered text data (or voice data) and sends it to the server. The user's question is given as input and sent to the server as output.

[1720] Step 2:

[1721] The server receives input data from the user. The server receives the input text data or audio data, and if it is audio data, it converts it to text data using speech recognition. Text data (or converted text data) is given as input, and text data for analysis is obtained as output.

[1722] Step 3:

[1723] The server starts up a natural language processing engine and analyzes the text data. The natural language processing engine uses techniques such as tokenization, part-of-speech tagging, and semantic analysis to understand the intent of the user's question. Text data is given as input, and the result of intent analysis is obtained as output.

[1724] Step 4:

[1725] The server retrieves relevant information from the knowledge base. Based on the analysis results, the server searches for and retrieves relevant information and content from the knowledge base (database). The result of intent analysis is given as input, and relevant information is obtained as output.

[1726] Step 5:

[1727] The server uses a video recommendation algorithm to select personalized videos. The video recommendation algorithm considers the user's past viewing history and feedback to calculate data for recommending the most suitable videos. The analysis results and the user's history data are given as input, and a list of recommended videos is obtained as output.

[1728] Step 6:

[1729] The server uses a response generation engine to generate a response to provide to the user. The response generation engine generates a response to the user based on the acquired information and the recommended video list. Relevant information and the recommended video list are given as input, and the final response message is obtained as output.

[1730] Step 7:

[1731] The server sends a response message to the smartphone. The final response message is sent from the server to the smartphone. The final response message is given as input, and the output is displayed on the smartphone.

[1732] Step 8:

[1733] The smartphone displays the response message to the user. The smartphone displays the response message received from the server to the user. It also provides a voice reading function if necessary. The final response message is given as input, and it is displayed to the user as output.

[1734] 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.

[1735] This invention incorporates a function that recognizes the user's emotions and adjusts the response accordingly, in addition to analyzing user-inputted questions and providing relevant information quickly. Its specific form is described below.

[1736] User Interface (UI)

[1737] Terminal: Provides an interface for users to query information. This includes text boxes for text input and voice input buttons. In the case of voice input, a speech recognition mechanism is built in to convert speech into text.

[1738] Natural Language Processing (NLP) Engine

[1739] Server: It has a natural language processing engine that analyzes text data received from users and understands their intent. This engine consists of the following technologies:

[1740] Tokenization: Breaking down a text into individual words or phrases.

[1741] Part-of-speech tagging: Identifying the part of speech of each word.

[1742] Semantic analysis: Analyzes the meaning of the entire text to identify the user's intent.

[1743] Knowledge Base

[1744] Server: It has a large database for searching relevant information based on user intent. This database includes internal and external regulations, procedures, steps, guidelines, training, etc.

[1745] Response generation engine

[1746] Server: There is an engine that retrieves information from a knowledge base based on the results analyzed by the NLP engine and generates appropriate responses for the user. The response generation engine has the following functions:

[1747] Template Selection: Choose the appropriate template.

[1748] Information Embedding: Embed the retrieved information into the template.

[1749] Emotional Engine

[1750] Server: It has an emotion engine for recognizing emotions from user input. This engine analyzes the degree of emotion from the user's text and voice input and identifies emotions such as joy, anger, and sadness.

[1751] Action execution engine

[1752] Server: Includes an engine that takes notification or escalation actions in response to specific inquiries. For example, if a user's emotions exceed a threshold for anger or sadness, it sends a notification to a specific department.

[1753] Information provision means

[1754] Terminal: Displays response messages sent from the server to the user. These may be displayed as text or read aloud.

[1755] Specific example

[1756] Example 1: Inquiry about leave application procedures

[1757] 1. User: Types the following text: "How do I apply for leave?"

[1758] 2. Terminal: Sends user input to the server.

[1759] 3. Server: The server uses a natural language processing engine to analyze the question and identify information related to "leave application procedures."

[1760] 4. Server: Retrieves relevant information from the knowledge base and generates a response stating, "To submit a leave request, follow these steps: 1. Log in to the HR portal. 2. Click the 'Leave Request' tab. 3. Enter the required information and click the submit button."

[1761] 5. Server: The emotion engine analyzes the user's emotions. Since it is determined to be a normal question, it determines that no special action is required.

[1762] 6. Terminal: Displays the generated response to the user.

[1763] Example 2: Inquiry about unpaid wages

[1764] 1. User: "My salary hasn't been paid, who should I ask about it?" (Voice input)

[1765] 2. Terminal: Converts speech to text and sends it to the server.

[1766] 3. Server: Uses a natural language processing engine to analyze the question and identify the "contact information for inquiries regarding unpaid wages."

[1767] 4. Server: Retrieves HR department contact information from the knowledge base and generates a response stating, "Inquiries regarding unpaid wages will be escalated to the HR department contact."

[1768] 5. Server: The emotion engine analyzes the user's emotions. If the user is showing strong anger or sadness, the action execution engine sends an escalation notification to the HR department.

[1769] 6. Terminal: Display the generated response to the user and confirm that the notification has been sent.

[1770] By combining these functions, the present invention provides a system that can immediately provide appropriate information in response to user inquiries, respond in a way that takes user emotions into consideration, and quickly execute necessary actions.

[1771] The following describes the processing flow.

[1772] Detailed program processing flow

[1773] Example 1: Inquiry about leave application procedures

[1774] Step 1:

[1775] The user types the text, "How do I apply for leave?"

[1776] Step 2:

[1777] The device receives user input. If the user provides voice input, the speech recognition engine first converts the speech into text.

[1778] Step 3:

[1779] The terminal sends the received text data to the server.

[1780] Step 4:

[1781] The server's natural language processing engine analyzes the user's text. First, it tokenizes the text, then tags it with parts of speech, and finally performs semantic analysis to determine the user's intent.

[1782] Step 5:

[1783] Based on the analysis results, the server queries the knowledge base for information regarding "leave application procedures."

[1784] Step 6:

[1785] The server retrieves the relevant information from the knowledge base.

[1786] Step 7:

[1787] The server's response generation engine creates a reply message based on the information it has obtained. Specifically, it generates a message that says, "To request leave, follow these steps: 1. Log in to the HR portal. 2. Click the 'Leave Request' tab. 3. Enter the required information and click the submit button."

[1788] Step 8:

[1789] The server's emotion engine analyzes the user's input text and determines the degree of emotion. In this case, it recognizes it as a normal question and determines that there is no special emotional response.

[1790] Step 9:

[1791] The server sends the generated message to the terminal.

[1792] Step 10:

[1793] The device displays the message to the user. In the case of voice output, the text is converted to speech and read aloud.

[1794] Example 2: Inquiry about unpaid wages

[1795] Step 1:

[1796] The user inputs a voice message saying, "My salary hasn't been paid, who should I ask about it?"

[1797] Step 2:

[1798] The device receives the audio and converts it to text using a speech recognition engine.

[1799] Step 3:

[1800] The terminal sends text data to the server.

[1801] Step 4:

[1802] The server's natural language processing engine analyzes the user's text. It processes the text in the following order: tokenization, part-of-speech tagging, and semantic analysis, identifying the user's intent as "an inquiry about unpaid wages."

[1803] Step 5:

[1804] The server executes queries against the knowledge base based on the analysis results.

[1805] Step 6:

[1806] The server retrieves information on "contact information for inquiries regarding unpaid wages" from its knowledge base. Here, it retrieves information on the HR department contact person.

[1807] Step 7:

[1808] The server's response generation engine creates a message stating, "Inquiries regarding unpaid wages will be escalated to the HR department."

[1809] Step 8:

[1810] The server's emotion engine analyzes the user's input text to determine the degree of emotion. If the user indicates strong anger or sadness, the action execution engine sends an escalation notification to the HR department.

[1811] Step 9:

[1812] The server generates an escalation notification and sends it to the HR department.

[1813] Step 10:

[1814] The server sends the generated message to the terminal.

[1815] Step 11:

[1816] The device displays the message to the user and confirms that the notification has been sent. In the case of voice output, it converts the text to speech and reads it aloud.

[1817] In this way, we can provide appropriate information in response to user questions, respond to their emotions, and quickly take necessary actions.

[1818] (Example 2)

[1819] 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".

[1820] Traditional question-answering systems only provide information in response to user questions, failing to adequately improve user satisfaction by not considering user emotions or escalating problems. Furthermore, the processing and analysis of voice-input questions were sometimes inefficient. This made it difficult for users to obtain quick and appropriate answers, and particularly led to delays in responses when users were emotionally agitated.

[1821] The identification processing performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving a question entered by the user, natural language processing means for analyzing the received question and searching for relevant information, means for obtaining information from a database based on the analysis results, response generation means for providing the obtained information to the user, and emotion recognition means for analyzing emotions from the user's input. This makes it possible to provide quick and appropriate information in response to the user's question, and to automatically perform responses that take the user's emotions into consideration and necessary escalations.

[1822] A "user" refers to an individual or group that enters a question into the system.

[1823] "Means of receiving questions" refers to a device or software module that has the functionality to receive text or audio input from a user.

[1824] "Natural language processing means" refers to all technologies used to analyze received questions and understand the user's intent.

[1825] "Means of obtaining information from a database" refers to a device or software module that has the function of searching for and extracting relevant information from a database based on an analyzed query.

[1826] "Response generation means" refers to a device or software module that has the function of creating a response to be provided to the user based on acquired information.

[1827] "Emotion recognition means" refers to a device or software module that has the function of identifying emotions from user input and analyzing their degree.

[1828] "Action execution means" refers to a device or software module that has the functionality to perform notifications or escalations based on the detected user's emotions.

[1829] "Speech recognition means" refers to a device or software module that has the function of analyzing data input as speech and converting its content into text format.

[1830] This invention is a system that analyzes user-inputted questions and quickly provides relevant information. Furthermore, it incorporates a function to recognize the user's emotions and adjust its response accordingly. An embodiment of this system is described in detail below.

[1831] User Interface (UI)

[1832] Terminal: Provides an interface for users to query information. This interface includes text boxes for text input and buttons for voice input. In the case of voice input, a speech recognition mechanism is built in to convert the voice data into text.

[1833] Natural Language Processing (NLP) Engine

[1834] Server: It has a natural language processing engine that analyzes text data sent by the user and understands the user's intent. This engine consists of the following technologies:

[1835] Tokenization: Breaking down a text into individual words or phrases.

[1836] Part-of-speech tagging: Identifying the part of speech of each word.

[1837] Semantic analysis: Analyzes the meaning of the entire text to identify the user's intent.

[1838] Knowledge Base

[1839] Server: It has a large database for searching relevant information based on user intent. This database contains information such as internal and external regulations, procedures, steps, guidance, and training.

[1840] Response generation engine

[1841] Server: There is an engine that retrieves information from a knowledge base based on the results analyzed by the NLP engine and generates appropriate responses for the user. This response generation engine has the following functions:

[1842] Template Selection: Select the appropriate template.

[1843] Information Embedding: Embed the retrieved information into the template.

[1844] Emotional Engine

[1845] Server: It has an emotion engine for recognizing emotions from user input. This engine analyzes the degree of emotion from the user's text and voice input and identifies emotions such as joy, anger, and sadness.

[1846] Action execution engine

[1847] Server: Includes an engine that takes notification or escalation actions in response to specific inquiries. For example, if a user's emotions exceed a threshold for anger or sadness, it sends a notification to a specific department.

[1848] Information provision means

[1849] Terminal: Displays the response message sent from the server to the user. This response may be displayed in text format or read aloud.

[1850] Specific example

[1851] Example 1: Inquiry about leave application procedures

[1852] 1. User: Types the following text: "How do I apply for leave?"

[1853] 2. Terminal: Sends user input to the server.

[1854] 3. Server: The NLP engine analyzes the question and identifies information related to "leave application procedures".

[1855] 4. Server: Retrieves relevant information from the knowledge base and generates a response stating, "To submit a leave request, follow these steps: 1. Log in to the portal. 2. Click the 'Leave Request' tab. 3. Enter the required information and click the submit button."

[1856] 5. Server: The emotion engine analyzes the user's emotions. Since it is determined to be a normal question, it determines that no special action is required.

[1857] 6. Terminal: Displays the generated response to the user.

[1858] Example 2: Inquiry about unpaid wages

[1859] 1. User: "My salary hasn't been paid. Who should I contact?" (Voice input)

[1860] 2. Terminal: Converts speech to text and sends it to the server.

[1861] 3. Server: The NLP engine analyzes the question and identifies the "contact point for inquiries regarding unpaid wages."

[1862] 4. Server: Retrieves information on the HR department contact person from the knowledge base and generates a response stating, "Please direct inquiries regarding unpaid wages to the HR department contact person."

[1863] 5. Server: The emotion engine analyzes the user's emotions. If the user is showing strong anger or sadness, the action execution engine sends an escalation notification to the HR department.

[1864] 6. Terminal: Display the generated response to the user and confirm that the notification has been sent.

[1865] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1866] Step 1:

[1867] User: The user enters their question through the interface. For example, they might type "Please tell me about the company's benefits" into the text box. Alternatively, using voice input, they might say into the microphone, "My salary hasn't been paid, who should I contact?" The input is sent to the terminal as text or voice data.

[1868] Step 2:

[1869] Terminal: Receives user input, and if the input is voice data, converts the voice data into text data using speech recognition. This converted text data, or the original text data, is sent to the server. The input is the original user's question, and the output is text data.

[1870] Step 3:

[1871] Server: Receives text data and analyzes it using a natural language processing (NLP) engine. Specifically, it first tokenizes the input sentence, breaking it down into words and phrases. Then, it performs part-of-speech tagging to identify the grammatical role of each word. Next, it performs semantic analysis to analyze the meaning of the entire sentence and identify the user's intent. The input is text data, and the output is analyzed intent data.

[1872] Step 4:

[1873] Server: Based on intent data analyzed by the NLP engine, it searches for relevant information from a knowledge base. This knowledge base includes internal and external regulations, procedures, steps, and guidelines. It executes database queries within the knowledge base to retrieve appropriate information. The input is the analyzed intent data, and the output is the retrieved information.

[1874] Step 5:

[1875] Server: Uses a response generation engine to generate a response to provide to the user based on the information obtained. First, it selects an appropriate template, and then it embeds the obtained information into the template to form the final response. The input is the information obtained, and the output is the generated response text.

[1876] Step 6:

[1877] Server: Uses an emotion engine to analyze the user's emotions from the generated response and the user's input text. The emotion engine analyzes specific words, phrases, and tones in the user's text to identify emotions such as joy, anger, and sadness. The input is the user's input text and the generated response, and the output is emotion data.

[1878] Step 7:

[1879] Server: Based on the emotion data identified by the emotion engine, the action execution engine performs specific actions. For example, if a user's emotion exceeds the threshold for anger or sadness, a notification is sent to a specific department. The input is emotion data, and the output is an escalation notification.

[1880] Step 8:

[1881] Terminal: Receives response messages sent from the server and displays them to the user. This response is displayed in text format or read aloud. Specifically, text is displayed on the terminal's display, or audio is played through the speaker. The input is the response from the server, and the output is the display or audio output to the user.

[1882] (Application Example 2)

[1883] 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".

[1884] In autonomous vehicles, there is a need for a system that can respond quickly and appropriately to passengers' questions, and furthermore, respond flexibly according to the passengers' emotional state. However, conventional systems have had difficulty realizing responses that take passengers' emotions into account or escalation processing based on specific emotional states. As a result, passenger dissatisfaction and problems may not be properly resolved. To solve these problems, this invention provides a new system that combines natural language processing and sentiment analysis.

[1885] 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.

[1886] In this invention, the server includes means for receiving user input, means for natural language processing for analyzing user input, means for obtaining information from a knowledge base based on the analysis results, means for generating a response that provides the obtained information to the user, means for analyzing emotions from user input, and means for executing necessary actions based on the analyzed emotions. This makes it possible to answer user questions quickly and appropriately, and to respond flexibly based on the user's emotional state.

[1887] "Means for receiving user-entered questions" refers to an interface for receiving questions entered by the user in text or voice.

[1888] "Natural language processing means" are technical means for analyzing received text data and understanding its intent.

[1889] "Means of obtaining information from a knowledge base" refers to a function that extracts necessary information from relevant databases based on the analysis results.

[1890] A "response generation means" is an engine that generates answers to be provided to the user based on the acquired information.

[1891] An "emotion analysis tool" is a system that identifies emotions from user input text or voice and analyzes their emotional state.

[1892] An "action execution mechanism" is a function that performs a specific action (such as notification or escalation) based on the analyzed emotional state.

[1893] "Speech recognition means" refers to technical means for converting speech input into text data.

[1894] "Means of providing information to the user" refers to an interface for presenting the generated response to the user in text or audio format.

[1895] The present invention is implemented as a system for responding quickly and appropriately to passenger questions in autonomous vehicles. Specifically, it has the function of analyzing questions entered by the user, obtaining appropriate information from a knowledge base, providing the generated response to the user, and also analyzing the user's emotional state and taking necessary actions based on that analysis.

[1896] Hardware and software configuration

[1897] This system consists of the following hardware and software.

[1898] hardware

[1899] Server: Provides the primary computing resources for backend processing.

[1900] Terminal: Infotainment system within an autonomous vehicle. Includes user interfaces such as touchscreens and voice recognition microphones.

[1901] software

[1902] Natural Language Processing Engine (NLP Engine): Uses the Hugging Face natural language processing model. It has the capability to analyze questions using the transformers library.

[1903] Emotion analysis engine: Also using Hugging Face's pipeline, it identifies the user's emotional state.

[1904] Knowledge base: A large-scale database containing vehicle functions, operational information, etc.

[1905] Response generation engine: Generates responses to the user based on the acquired information.

[1906] Action Execution Engine: Based on sentiment analysis results, it executes actions (notifications or escalations) based on specific emotional states.

[1907] System operation

[1908] 1. Receiving and analyzing user input

[1909] The user enters a question via voice or text through the vehicle's infotainment system. In the case of voice input, a speech recognition system converts it into text. The converted text is sent to a server, where a natural language processing engine analyzes the intent of the question.

[1910] 2. Information acquisition from knowledge bases

[1911] Based on the results analyzed by the natural language processing engine, the server searches for and retrieves relevant information from its knowledge base. This information covers a wide range of topics, including vehicle operation methods and route information.

[1912] 3. Response generation and sentiment analysis

[1913] Based on the acquired information, the response generation engine generates an appropriate answer. Simultaneously, the sentiment analysis engine analyzes the user's emotional state from their questions and other inputs, and determines the analysis result.

[1914] 4. Performing a specific action

[1915] If the emotion analysis results are "negative," for example, if anger or frustration is strong, the action execution engine will send a notification to a specific department or service center. This notification is a measure to ensure that appropriate action is taken quickly.

[1916] Specific example

[1917] Example question: "Please tell me how to use the air conditioner."

[1918] Analysis results and response: The NLP engine analyzes the question, retrieves information about "how to use the air conditioner" from the knowledge base, and generates the response, "To operate the air conditioner, tap the 'Air Conditioner' tab on the central touchscreen."

[1919] Sentiment Analysis and Action: The sentiment analysis engine determines the user's emotion to be "normal." In this case, no special action is required.

[1920] Example question: "I want to change the background music in my car, how do I do that?"

[1921] Analysis results and response: The NLP engine analyzes the question, retrieves information about "how to change background music" from the knowledge base, and generates the response: "Click the menu icon at the top of the central touchscreen -> 'Settings' -> 'Audio Settings' -> Select from 'Playlist'."

[1922] Thus, the present invention provides a system for autonomous vehicles that can respond quickly to passengers' questions and further provide flexible responses according to their emotional state.

[1923] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1924] Step 1:

[1925] The terminal receives user input.

[1926] The user enters their question into the vehicle's infotainment system via voice or text.

[1927] In the case of voice input, a speech recognition system converts the speech into text and generates text data. The entered text is then sent from the terminal to the server.

[1928] Step 2:

[1929] The server analyzes the question.

[1930] The server uses a natural language processing engine to analyze the received text data and identify the intent of the input. This process involves tokenization, part-of-speech tagging, and semantic analysis to understand the structure and content of the user's question. The input is the user's question text, and the output is the intent of the question and important keywords.

[1931] Step 3:

[1932] The server retrieves information from the knowledge base.

[1933] The server searches for relevant information from a knowledge base based on the analysis results of the natural language processing engine. The knowledge base includes vehicle operation instructions and route information, and appropriate data is extracted. The input is the analysis result, and the output is the data of the relevant information.

[1934] Step 4:

[1935] The server generates a response.

[1936] The server uses a response generation engine to generate answers for the user based on the acquired information. The generated answers are created by embedding information into a template. The input is information acquired from the knowledge base, and the output is the answer text provided to the user.

[1937] Step 5:

[1938] The server performs sentiment analysis.

[1939] The server uses an emotion analysis engine to identify emotions from the user's input text. Here, the emotional state the user was in when asking a question (e.g., joy, anger, sadness, etc.) is determined. The input is the user's question text, and the output is the result of the emotional state analysis.

[1940] Step 6:

[1941] The server performs the necessary actions.

[1942] The server performs actions based on the sentiment analysis results for specific emotional states. For example, if a user is highly angry, the action execution engine sends a notification to a specific department or person in charge. The input is the sentiment analysis result, and the output is the result of the action (such as sending a notification).

[1943] Step 7:

[1944] The device provides information to the user.

[1945] The response text generated by the server is sent to the terminal and presented to the user. It may be displayed in text or audio format. The input is the response text from the server, and the output is information provided to the passenger.

[1946] 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.

[1947] 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.

[1948] 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.

[1949] 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.

[1950] 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.

[1951] 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.

[1952] 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.

[1953] 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.

[1954] 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."

[1955] 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.

[1956] 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.

[1957] 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.

[1958] 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.

[1959] 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.

[1960] 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.

[1961] 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.

[1962] 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.

[1963] 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.

[1964] 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.

[1965] 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.

[1966] 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.

[1967] The following is further disclosed regarding the embodiments described above.

[1968] (Claim 1)

[1969] A means of receiving questions entered by the user,

[1970] A natural language processing means for analyzing received questions and searching for relevant information,

[1971] A means of obtaining information from a knowledge base based on the analysis results,

[1972] A means for generating a response to provide the acquired information to the user,

[1973] Means of providing information to users,

[1974] A system that includes this.

[1975] (Claim 2)

[1976] The system according to claim 1, further comprising means for performing actions such as notifications or escalations based on user inquiries.

[1977] (Claim 3)

[1978] The system according to claim 1, further comprising speech recognition means for converting speech input into text.

[1979] "Example 1"

[1980] (Claim 1)

[1981] A means of receiving questions entered by the user,

[1982] A natural language processing means for analyzing received questions and searching for relevant information,

[1983] A means of obtaining information from a knowledge base based on the analysis results,

[1984] A means for generating a response to provide the acquired information to the user,

[1985] Means of providing information to users,

[1986] A system that includes this.

[1987] (Claim 2)

[1988] The system according to claim 1, further comprising means for performing actions such as notifications or escalations based on user inquiries.

[1989] (Claim 3)

[1990] The system according to claim 1, further comprising speech recognition means for converting speech input into text.

[1991] (Claim 4)

[1992] The system according to claim 1, further comprising a tokenization means for tokenizing a question entered by a user and breaking it down into words or phrases.

[1993] (Claim 5)

[1994] The system according to claim 4, further comprising a part-of-speech tagging means for identifying the part of speech of each word.

[1995] (Claim 6)

[1996] The system according to claim 5, further comprising semantic analysis means for analyzing the meaning of the entire user question and identifying the user's intent.

[1997] (Claim 7)

[1998] The system according to claim 1, further comprising an information embedding means for selecting a template and embedding acquired information into the template.

[1999] "Application Example 1"

[2000] (Claim 1)

[2001] A means of receiving questions entered by the user,

[2002] A natural language processing means for analyzing received questions and searching for relevant information,

[2003] A means of obtaining information from a knowledge base based on the analysis results,

[2004] A means for generating a response to provide the acquired information to the user,

[2005] A means of providing users with personalized video recommendations,

[2006] A system that includes this.

[2007] (Claim 2)

[2008] The system according to claim 1, further comprising means for performing actions such as notifications or escalations based on user inquiries.

[2009] (Claim 3)

[2010] The system according to claim 1, further comprising speech recognition means for converting speech input into text.

[2011] "Example 2 of combining an emotion engine"

[2012] (Claim 1)

[2013] A means of receiving questions entered by the user,

[2014] A natural language processing means for analyzing received questions and searching for relevant information,

[2015] A means of obtaining information from a database based on the analysis results,

[2016] A response generation means for providing the acquired information to the user,

[2017] An emotion recognition method that analyzes emotions from user input,

[2018] Means of providing information to users,

[2019] A system that includes this.

[2020] (Claim 2)

[2021] The system according to claim 1, further comprising means for performing actions such as notifications or escalations based on user inquiries.

[2022] (Claim 3)

[2023] The system according to claim 1, further comprising speech recognition means for converting speech input into text.

[2024] "Application example 2 when combining with an emotional engine"

[2025] (Claim 1)

[2026] A means of receiving questions entered by the user,

[2027] A natural language processing means for analyzing received questions and searching for relevant information,

[2028] A means of obtaining information from a knowledge base based on the analysis results,

[2029] A means for generating a response to provide the acquired information to the user,

[2030] A means of analyzing emotions from user input,

[2031] An action execution mechanism that performs necessary actions based on analyzed emotions,

[2032] Means of providing information to users,

[2033] A system that includes this.

[2034] (Claim 2)

[2035] The system according to claim 1, further comprising means for performing actions such as notifications or escalations based on a specific emotional state.

[2036] (Claim 3)

[2037] The system according to claim 1, further comprising speech recognition means for converting speech input into text. [Explanation of Symbols]

[2038] 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 questions entered by the user, A natural language processing means for analyzing received questions and searching for relevant information, A means of obtaining information from a knowledge base based on the analysis results, A means for generating a response to provide the acquired information to the user, Means of providing information to users, A system that includes this.

2. The system according to claim 1, further comprising means for executing actions such as notifications and escalations based on user inquiries.

3. The system according to claim 1, further comprising a speech recognition means for converting speech input into text.

Images