system

Abstract

Provide a system. A receiving means; a text analysis means for analyzing the received text data to extract key elements; A reply generation means for generating reply candidates based on the user's past message history and communication style; a transmitting means for transmitting the generated reply candidates to a user terminal; A system including:

Description

[Technical Field]

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

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

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

[0004] In modern communication, there is a demand for quick and appropriate replies to text data sent and received as messages. However, it is difficult for busy users to come up with an appropriate reply immediately when receiving a message, resulting in problems such as delayed replies and messages being read but ignored. There is a need to provide a system that supports efficient and smooth communication by solving this problem. [Means for solving the problem]

[0005] In order to solve the above problems, the present invention provides the following means. The system is equipped with a text analysis means that analyzes messages received by a user terminal and extracts their main elements, thereby accurately understanding the intent of the message. The system also provides a reply generation means that generates appropriate reply candidates based on the user's past message history and communication style, and a transmission means that transmits the generated reply candidates to the user terminal. This provides a system that allows the user to simply select an appropriate reply from the candidates, enabling quick and appropriate communication.

[0006] The "receiving means" is a device or software that has the function of receiving text data sent to the user terminal.

[0007] "Text analysis means" is software that has the function of analyzing the content of received text data and extracting key elements and intent from it.

[0008] The "reply generation means" is software that has the function of generating appropriate reply message candidates based on the user's past message history and communication style.

[0009] The "transmission means" is a device or software that has the function of transmitting the generated reply candidates to the user terminal.

[0010] "User terminal" means a device (e.g., smartphone, tablet, computer, etc.) used by a user to send or receive messages.

[0011] "Natural language processing technology" is a technology that analyzes text data to understand its content and intent, and is primarily achieved using machine learning models and algorithms.

[0012] "Reply candidates" are multiple candidate messages that the system suggests as a reply to a message received by the user.

[0013] "Message History Database" means a database for recording and storing messages sent and received by a user in the past. [Brief explanation of the drawings]

[0014] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14]FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0015] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0016] First, the terms used in the following description will be explained.

[0017] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0018] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0019] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0020] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0021] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0022] [First embodiment]

[0023] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0024] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0025] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0026] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0027] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0028] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0029] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0030] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0031] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0032] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0033] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0034] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0035] The present invention relates to a system and method for making users' message replies more efficient and supporting appropriate communication, and detailed embodiments thereof are described below.

[0036] The present invention is a system consisting of a user terminal, a server, a text analysis engine, and an AI model, which supports a user in replying promptly and appropriately to a message received.

[0037] Overview of program processing flow

[0038] Receiving and parsing messages

[0039] 1. A user receives a message using a chat application, for example, from a friend, "Want to go see a movie with me this weekend?"

[0040] 2. The device sends the received message to the server, where it is sent as text data via a communication protocol.

[0041] 3. The server's text analysis engine receives the message and begins analyzing it. The text analysis engine uses natural language processing technology to understand the subject and intent of the message. For example, it extracts keywords such as "weekend," "movie," and "go together" and classifies the intent of the message as a "request to confirm plans."

[0042] Generate reply suggestions

[0043] 4. The server accesses the user's past message history database and evaluates the user's communication style, thereby understanding how the user has responded in the past and how they prefer to express themselves.

[0044] 5. The server's AI model generates appropriate reply suggestions based on the text analysis results and the user's communication style. For example, suggestions like the following might be generated:

[0045] "Thanks, I'd love to go! What time would be good?"

[0046] "Sorry, I have plans this weekend so I can't make it. See you next time."

[0047] I haven't decided on a schedule yet, so I'll think about it for a bit and get back to you.

[0048] Sending and selecting suggested replies

[0049] 6. The server sends the generated reply candidates to the device, where they are displayed on the user interface (UI).

[0050] 7. The user selects the appropriate reply from the suggested replies, for example, "Thanks, I'd love to! What time would be good?"

[0051] 8. The terminal sends the selected reply message to the server.

[0052] 9. The server delivers the selected reply message to the recipient friend, thereby completing the communication promptly and appropriately.

[0053] Specific examples

[0054] For example, when a user receives a message from a friend saying, "Would you like to go see a movie with me this weekend?", the system works as follows:

[0055] When a user receives a message, the device sends it to the server. The server uses a text analysis engine to analyze the message and extract key elements. In this case, the extracted result is "Request to confirm schedule." The server then generates appropriate reply candidates for this element based on the user's past message history. The generated reply candidates are then sent to the device and displayed to the user. For example, a candidate might be "Thanks, I'd love to go! What time would be good?" The user selects an appropriate reply and presses the send button on the device, which then sends the reply to their friend via the server.

[0056] In this way, the present invention supports users in responding to messages quickly and appropriately, thereby improving the efficiency of user communication.

[0057] The processing flow will be explained below.

[0058] Step 1:

[0059] A user has a chat app open and receives a message from a friend saying, "Want to go see a movie with me this weekend?"

[0060] Step 2:

[0061] The terminal displays the received message and simultaneously sends the message as text data to the server.

[0062] Step 3:

[0063] The server sends the received message as a request to the text analysis engine, which uses natural language processing technology to analyze the content of the message.

[0064] Step 4:

[0065] The text analysis engine analyzes the message and extracts key elements (e.g., "weekend," "movie," and "going together"). As a result of the analysis, the intent of the message is classified as a "request to confirm plans."

[0066] Step 5:

[0067] The server accesses the user's past message history database to acquire the user's communication style, thereby learning what expressions the user has used in the past.

[0068] Step 6:

[0069] The server's AI model generates multiple reply suggestions based on the text analysis results and the user's communication style, for example:

[0070] "Thanks, I'd love to go! What time would be good?"

[0071] "Sorry, I have plans this weekend so I can't make it. See you next time."

[0072] I haven't decided on a schedule yet, so I'll think about it for a bit and get back to you.

[0073] Step 7:

[0074] The server generates reply suggestions and sends them to the device, where they are displayed to the user.

[0075] Step 8:

[0076] The user selects the appropriate reply from the suggested replies, for example, "Thanks, I'd love to! What time would be good?"

[0077] Step 9:

[0078] The terminal sends the selected reply message to the server.

[0079] Step 10:

[0080] The server delivers the final reply message to the recipient friend, who then displays the message on their device.

[0081] This series of steps allows users to quickly and appropriately reply after receiving a message.

[0082] Example 1

[0083] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0084] In modern communication, users are expected to respond quickly and appropriately. However, constantly rethinking the content of their replies to messages is a time-consuming task for users. Therefore, there is a need for a way for users to efficiently generate replies to received messages and communicate quickly.

[0085] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0086] In this invention, the server includes a receiving means, a text analyzing means for analyzing the received text data and extracting key elements, a reply generating means for generating reply candidates based on the user's past message history and communication style, a transmitting means for transmitting the generated reply candidates to the user terminal, and a distributing means for receiving a reply message selected from the user terminal and transmitting it to the receiving terminal, thereby enabling the user to reply quickly and appropriately to the message received.

[0087] The "receiving means" is a means for processing a message received by a user through a chat application and sending it to a server.

[0088] The "text analysis means" is a means for analyzing received text data and extracting key elements and keywords.

[0089] The "reply generation means" is a means for generating appropriate reply candidates based on the elements and keywords extracted by the text analysis means, as well as the user's past message history and communication style.

[0090] The "transmitting means" is a means for transmitting the generated reply candidates to the user terminal and displaying them on the user interface of the user terminal.

[0091] The "distribution means" is a means for sending a selected reply message from a user terminal back to the server, and finally sending the reply message to the receiving terminal.

[0092] "Natural language processing technology" is a field of computer science that is a technology for understanding, analyzing, and generating human language.

[0093] An "AI model" is an algorithm or data structure that uses artificial intelligence to perform various tasks, and in natural language processing in particular, it is a model for generating and analyzing text.

[0094] The present invention relates to a system and method for improving the efficiency of user message replies and supporting appropriate communication. The system comprises a user terminal, a server, a text analysis engine, and an AI model.

[0095] Receiving and parsing messages

[0096] First, a user receives a message using a chat application. For example, the received message might say, "Would you like to go see a movie with me this weekend?" The device receives this message and sends it to the server. The HTTP POST request is typically used as the communication protocol.

[0097] The server passes the received message to a text analysis engine. This text analysis engine uses natural language processing technology, such as Google® Cloud Natural Language API or IBM Watson® Natural Language Understanding. The text analysis engine analyzes the message and extracts key elements and keywords. Specifically, the keywords "weekend," "movie," and "go together" are extracted, and the intent of the message is classified as a "request to confirm plans."

[0098] Generate reply suggestions

[0099] The server then accesses a database of the user's past message history, which can be a relational or NoSQL database like MySQL or MongoDB, and runs a database query to retrieve the user's past reply patterns and communication style.

[0100] Based on the acquired data, the server uses an AI model (e.g., OpenAI's GPT-3 or BERT) to generate appropriate reply suggestions. This AI model receives text analysis results and the user's communication style as input data. The reply suggestions generated include:

[0101] "Thanks, I'd love to go! What time would be good?"

[0102] "Sorry, I have plans this weekend so I can't make it. See you next time."

[0103] I haven't decided on a schedule yet, so I'll think about it for a bit and get back to you.

[0104] Sending and selecting suggested replies

[0105] The generated reply candidates are sent by the server to the device. The transmission is done through an HTTP response, and the reply candidates are encoded in JSON format. The device decodes them and displays them in the application's user interface. The user can select the appropriate reply from multiple reply candidates. For example, the reply "Thanks, I'd love to go! What time would be good?" is selected.

[0106] The selected reply message is then sent back to the server, which then delivers it to the recipient device using the appropriate protocol (e.g., XMPP or MQTT) for the chat application server, ensuring fast and accurate communication.

[0107] Specific examples

[0108] For example, if a user receives a message from a friend saying, "Want to go see a movie with me this weekend?", the system operates as follows:

[0109] The user receives the message.

[0110] 2. The device sends the message to the server.

[0111] 3. The server passes the message to a text analysis engine, which extracts the key elements: "weekend," "movie," and "going together."

[0112] 4. The server accesses the user's history database to check past response patterns.

[0113] 5. The server's AI model generates appropriate reply suggestions.

[0114] 6. The server sends the generated reply candidates to the terminal, which displays them.

[0115] 7. The user selects the reply "Thanks, I'd love to! What time would be good?"

[0116] 8. The device sends the selected reply to the server, which distributes it to the friend.

[0117] In this way, the present invention supports users in quickly and appropriately replying to messages they receive, thereby improving the efficiency of user communication.

[0118] Example prompts to input to the generative AI model

[0119] Generate a suitable reply to a message from a friend saying "Want to go see a movie with me this weekend?". The output should be in the following format:

[0120] 1. "Reply candidate #1"

[0121] 2. "Reply candidate #2"

[0122] 3. "Reply candidate #3"

[0123] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0124] Step 1:

[0125] A user receives a message using a chat application. For example, a friend might send a message saying, "Want to go see a movie with me this weekend?", and the message is received by the user's device. The input is the message from the friend, and the output is the received message on the device.

[0126] Step 2:

[0127] The terminal sends the message received to the server. Specifically, it uses an HTTP POST request to send the message content to the server. The input is the received message, and the output is the text data sent to the server.

[0128] Step 3:

[0129] The server passes the received message to the text analysis engine and begins analysis. The text analysis engine analyzes the content of the message and extracts key elements and keywords. For example, the keywords "weekend," "movie," and "go together" are extracted using the Google Cloud Natural Language API. The input is the text data sent to the server, and the output is the analysis results (the extraction of key elements).

[0130] Step 4:

[0131] The server accesses a database of the user's past message history. It executes a database query to obtain past reply patterns and the user's communication style. The database can be, for example, MySQL or MongoDB. The input is the analysis result and the user's ID, and the output is the past message history data.

[0132] Step 5:

[0133] The server uses an AI model to generate appropriate reply suggestions. Based on the analysis results and past message history data, the AI ​​model (e.g., OpenAI's GPT-3) generates reply suggestions. The generated reply suggestions include:

[0134] "Thanks, I'd love to go! What time would be good?"

[0135] "Sorry, I have plans this weekend so I can't make it. See you next time."

[0136] I haven't decided on a schedule yet, so I'll think about it for a bit and get back to you.

[0137] The input is the analysis results and the user's message history data, and the output is a list of reply candidates.

[0138] Step 6:

[0139] The server generates reply candidates and sends them to the terminal. The HTTP response is used to send reply candidates encoded in JSON format to the terminal. The input is the reply candidate list, and the output is the data sent to the terminal.

[0140] Step 7:

[0141] The device decodes the reply candidates received and displays them on the user interface. The user selects the appropriate reply from the displayed multiple reply candidates. The input is the reply candidate data sent from the server, and the output is the candidates displayed on the UI. For example, the reply "Thanks, I'd love to go! What time would be good?" is displayed.

[0142] Step 8:

[0143] The user confirms the selected reply on the terminal and sends the reply message to the server. Again, an HTTP POST request is used to send the selected reply message to the server. The input is the reply message selected by the user, and the output is the data sent to the server.

[0144] Step 9:

[0145] The server receives the user's reply and delivers it to the recipient friend. It sends the reply message to the friend's device using an appropriate communication protocol (e.g., XMPP or MQTT). The input is the reply message selected by the user, and the output is the message delivered to the recipient's device. This completes fast and appropriate communication.

[0146] (Application example 1)

[0147] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0148] In recent years, security systems have been required to process warning messages quickly and accurately and to take appropriate action. However, the volume and variety of warning messages received by users has made it difficult to respond quickly. Therefore, it is necessary to develop a system that can instantly understand the content of warning messages and generate appropriate responses.

[0149] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0150] In this invention, the server includes a receiving means, a text analyzing means, a reply generating means, a sending means, and a security analyzing means, which enable the server to quickly and accurately analyze the received warning message, generate appropriate reply candidates, and present them to the user.

[0151] "Receiving means" refers to devices or software for receiving data or messages from the outside.

[0152] "Text analysis means" refers to a device or software for analyzing received text data and extracting key elements.

[0153] A "reply generator" is a device or software that generates appropriate reply candidates based on a user's past message history and communication style.

[0154] "Transmission means" refers to a device or software for transmitting the generated reply candidates to the user terminal.

[0155] "Security analysis means" refers to devices or software that analyze received warning messages and provide appropriate response candidates.

[0156] The present invention relates to a system that improves the efficiency of users' message replies and supports appropriate communication. The system includes a receiving unit, a text analyzing unit, a reply generating unit, a sending unit, and a security analyzing unit.

[0157] The user uses the security assistant application to receive warning messages sent from the security system or camera. This reception can be performed using an application installed on a device such as a smartphone. The received messages are then sent to the server.

[0158] The server analyzes the received text data using natural language processing technology based on Hugging Face's Transformers library. The text analysis method extracts key elements of the received warning message and uses this to understand the message's content and intent.

[0159] Based on the analysis results, the reply generator evaluates the user's past message history and communication style to generate appropriate reply candidates, using the user's response history database as well.

[0160] The generated reply candidates are sent to the user terminal by a sending means, and the user selects an appropriate reply from the presented reply candidates through the application interface.

[0161] For example, if a warning message is received stating, "The window sensor has detected an abnormality," the server analyzes the message and generates appropriate response candidates, such as, "We will call the police," or "Please check the security camera footage." These response candidates are displayed on the user's device, and the user selects an appropriate response, which is then sent via the server to the security system.

[0162] An example prompt is:

[0163] Alert message: "Window sensor has detected an abnormality"

[0164] Potential responses generated: "Call the police." "Check your security camera footage." "I need your help right away."

[0165] In this way, the present invention can quickly and accurately analyze received warning messages, generate appropriate response candidates, and present them to the user, thereby streamlining the user's security response and providing a safer environment.

[0166] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0167] Step 1:

[0168] The user uses the security assistant application to receive an alert message. This input is an alert message sent from a security system or camera in a home or office. The device receives this and sends it to the next step.

[0169] Step 2:

[0170] The terminal sends the received message to the server. In this process, the warning message is transferred to the server as text data. The input data is the warning message sent via the terminal, and the output data is the text data received by the server.

[0171] Step 3:

[0172] The server uses the Hugging Face Transformers library to analyze the received text data. The input data is the received text data, and the analyzed results include the main elements of the message (e.g., "sensor abnormality" and "window").

[0173] Step 4:

[0174] The server uses security analysis means to understand the contents of the warning message based on the analysis results and determines an appropriate response. This process determines what response is appropriate based on the analysis results (e.g., "sensor abnormality"). The input data is the analysis results, and the output data is information about the intended response (e.g., "warning response").

[0175] Step 5:

[0176] The reply generator evaluates the user's past message history and communication style to generate appropriate reply candidates. The input data are the analysis results and user history data, and the output data are the generated reply candidates (e.g., "I will call the police." "Please check the security camera footage.").

[0177] Step 6:

[0178] The server transmits the generated reply candidates to the user terminal, where the input data are the generated reply candidates and the output data are the multiple reply candidates displayed on the user terminal.

[0179] Step 7:

[0180] The user selects an appropriate reply from the presented reply candidates. The input data are the reply candidates presented to the user, and the output data is the reply selected by the user.

[0181] Step 8:

[0182] The terminal sends the selected reply message to the server, which delivers the reply to the security system. The input data is the reply selected by the user, and the output data is the final reply sent to the security system.

[0183] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0184] The present invention relates to a system and method for improving the efficiency of users' message replies and supporting appropriate communication, and is characterized in that reply candidates are generated taking into account the user's emotions by combining an emotion engine. Detailed embodiments are described below.

[0185] This system consists of a user terminal, a server, a text analysis engine, an AI model, and an emotion engine. The operation flow of the system will be explained in detail below.

[0186] Overview of program processing flow

[0187] Receiving and parsing messages

[0188] 1. A user receives a message using a chat application, for example, from a friend, "Want to go see a movie with me this weekend?"

[0189] 2. The terminal displays the received message and sends the message as text data to the server.

[0190] 3. The server sends the received message as a request to the text analysis engine and emotion engine. The analysis engine uses natural language processing technology to analyze the content of the message and extract key elements.

[0191] Emotion Analysis

[0192] 4. The emotion engine analyzes the message and recognizes the user's emotions. For example, emotions such as "happiness," "sadness," and "surprise" are identified by the emotion engine. In this case, the message "Would you like to go see a movie with me?" is analyzed as "curiosity" and "anticipation."

[0193] Generate reply suggestions

[0194] 5. The server accesses the user's past message history database to obtain the user's communication style, taking into account the analyzed emotional information.

[0195] 6. The server's AI model generates appropriate reply suggestions based on the text analysis results, the user's communication style, and emotional information. For example, the following suggestions might be generated:

[0196] "Thanks, I'd love to go! Sounds interesting."

[0197] "Sorry, I have plans this weekend so I can't make it. See you next time."

[0198] I haven't decided on a schedule yet, so I'll think about it for a bit and get back to you.

[0199] Sending and selecting suggested replies

[0200] 7. The server sends the generated reply candidates to the device, where they are displayed on the user interface (UI).

[0201] 8. The user selects the appropriate reply from the suggested replies displayed. For example, the user selects the reply "Thanks, I'd love to go! Sounds interesting."

[0202] 9. The terminal sends the selected reply message to the server.

[0203] 10. The server delivers the final reply message to the recipient friend, who then displays the message on their device.

[0204] Specific examples

[0205] For example, if a user receives a message from a friend saying, "Want to go see a movie with me this weekend?", the system works as follows:

[0206] While a user has a chat app open, they receive a message from a friend. The device sends this message to a server, which then asks the text analysis engine and emotion engine to analyze the message. The text analysis engine analyzes the message and classifies it as a "request to confirm plans," and the emotion engine extracts the emotions of "curiosity" and "anticipation." The server generates reply candidates based on the user's past message history and the results of these analyses. For example, a candidate might be "Thanks, I'd love to go! Sounds interesting." The server sends this to the device so the user can select one. After the user selects and sends an appropriate reply, the server distributes the reply to the friend.

[0207] In this way, the present invention provides a system that takes into account the user's emotions to create more personalized replies, making users' communications more efficient.

[0208] The processing flow will be explained below.

[0209] Step 1:

[0210] A user is using a chat app and receives a message from a friend saying, "Want to go see a movie with me this weekend?"

[0211] Step 2:

[0212] The terminal sends this message as text data to the server.

[0213] Step 3:

[0214] The server sends the received message as a request to the text analysis engine, which uses natural language processing technology to analyze the content of the message and extract key elements.

[0215] Step 4:

[0216] A text analysis engine analyzes the message, extracts key elements (e.g., "weekend," "movie," "going together"), and classifies the intent of the message as "a request to confirm plans."

[0217] Step 5:

[0218] The server simultaneously sends the message to the emotion engine, which uses natural language processing technology to analyze the emotions in the message.

[0219] Step 6:

[0220] The emotion engine analyzes messages and identifies emotions. For example, for the message "Would you like to go see a movie with me?", it extracts the emotions "curiosity" and "anticipation."

[0221] Step 7:

[0222] The server accesses a database of the user's past messages to obtain the user's communication style, for example, whether the user previously preferred polite replies.

[0223] Step 8:

[0224] The server's AI model generates multiple reply suggestions based on the text analysis results, the user's communication style, and emotional information, such as:

[0225] "Thanks, I'd love to go! Sounds interesting."

[0226] "Sorry, I have plans this weekend so I can't make it. Let's go next time."

[0227] I haven't decided on a schedule yet, so I'll think about it for a bit and get back to you.

[0228] Step 9:

[0229] The server sends these reply suggestions to the device, which then displays them on the user interface (UI).

[0230] Step 10:

[0231] The user selects the appropriate reply from the suggested replies, for example, "Thanks, I'd love to go! Sounds interesting."

[0232] Step 11:

[0233] The terminal sends the selected reply message to the server.

[0234] Step 12:

[0235] The server delivers the final reply message to the friend's device, which displays "Thanks, I'd love to go! Sounds interesting."

[0236] This series of steps allows users to quickly respond appropriately after receiving a message. By taking into account the user's emotions, the system provides more personalized replies and achieves effective communication.

[0237] Example 2

[0238] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0239] Conventional message reply systems have difficulty generating personalized and appropriate replies because they do not take into account the user's emotions or past communication styles. As a result, improvements to communication efficiency and user experience have been limited.

[0240] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0241] In this invention, the server includes a receiving means, a text analysis means for analyzing the received text data and extracting key elements, a sentiment analysis means for recognizing the user's sentiment based on the analyzed data, a reply generation means for generating appropriate reply candidates based on the user's past message history and communication style, and a transmission means for transmitting the generated reply candidates to the user terminal. This makes it possible to generate and provide personalized and appropriate reply candidates that take into account the user's sentiment and past communication style.

[0242] "Receiving means" means a piece of hardware or software designed to receive messages or data from a user.

[0243] "Text analysis means" means a piece of hardware or software that has the function of analyzing received text data and extracting key elements or information.

[0244] "Emotion analysis means" means a piece of hardware or software that employs technology to recognize and identify a user's emotions based on analyzed data.

[0245] A "reply generator" is a piece of hardware or software that takes into account a user's past message history and communication style to generate appropriate reply candidates.

[0246] The "transmitting means" is a piece of hardware or software for transferring the generated reply candidates to the user terminal.

[0247] "Natural language processing technology" is a technology that enables computers to understand, interpret, and generate human language.

[0248] A "user terminal" is a device that is directly operated by a user, and includes computers, smartphones, tablets, etc.

[0249] "Communication style" refers to characteristics such as tone, vocabulary, and expression used in messages a user has used in the past.

[0250] "Suggested replies" are system-generated suggested messages for a user to send.

[0251] A "database" is a structured collection of data for efficient storage and retrieval of specific information.

[0252] A "generative AI model" is a model that uses artificial intelligence technology to generate appropriate output based on user input.

[0253] The present invention relates to a system and method for improving the efficiency of users' message replies and supporting appropriate communication. The system includes a user terminal, a server, a text analysis engine, an AI model, and a sentiment analysis engine. The operational flow of this system is specifically shown below.

[0254] Hardware and Software

[0255] The system uses the following components:

[0256] User devices: computers, smartphones, tablets, etc.

[0257] Server: Cloud service or on-premise server.

[0258] Text analysis engines: Natural language processing libraries (e.g., SpaCy, NLTK).

[0259] Sentiment Analysis Engine: A sentiment analysis library (e.g., TextBlob, VADER).

[0260] Generative AI model: A model that uses artificial intelligence techniques (e.g., TENSORFLOW (registered trademark), PyTorch).

[0261] Receiving and parsing messages

[0262] A user uses a chat application and receives a message, for example, "Would you like to go see a movie with me this weekend?" The device displays this message and sends it as text data to the server. The server then sends the received message as a request to the text analysis engine and sentiment analysis engine. The text analysis engine analyzes the received message and extracts key elements.

[0263] Emotion Analysis

[0264] The emotion analysis engine analyzes messages to recognize the user's emotions. For example, in the message "Would you like to go see a movie with me this weekend?", it identifies the emotions "curiosity" and "anticipation."

[0265] Generate reply suggestions

[0266] The server accesses the user's past message history database to obtain the user's consistent communication style. Based on the analyzed text data and emotional information, the generative AI model generates reply candidates. For example, the following candidates are generated:

[0267] "Thanks, I'd love to go! Sounds interesting."

[0268] "Sorry, I have plans this weekend so I can't make it. See you next time."

[0269] I haven't decided on a schedule yet, so I'll think about it for a bit and get back to you.

[0270] Sending and selecting suggested replies

[0271] The server generates reply candidates and sends them to the device, which displays them on the user interface (UI). The user selects the most appropriate reply from the displayed candidates. For example, if the user selects "Thanks, I'd love to go! Sounds interesting," the device sends the selected reply message to the server. The server then distributes the final reply message to the friend's device, where the message is displayed.

[0272] Specific examples

[0273] For example, if a user receives a message from a friend saying, "Want to go see a movie with me this weekend?", the system works as follows:

[0274] While a user has a chat app open, the user receives a message from a friend. The device sends this message to a server, which then asks the text analysis engine and sentiment analysis engine to analyze the message. The text analysis engine analyzes the message and classifies it as a "request to confirm plans," and the sentiment analysis engine extracts the emotions of "curiosity" and "anticipation." The server generates reply candidates based on the user's past message history and the results of these analyses. For example, a candidate might be "Thanks, I'd love to go! Sounds interesting." The server sends this to the device so the user can select one. After the user selects and sends an appropriate reply, the server distributes the reply to the friend.

[0275] In this way, the present invention provides a system that takes into account the user's emotions to create more personalized replies, making users' communications more efficient.

[0276] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0277] Processing Steps

[0278] Step 1: Receiving a message

[0279] Input: A user receives a message using a chat application (e.g., "Want to go see a movie with me this weekend?").

[0280] Operation:

[0281] The user opens a chat app.

[0282] A message is sent from a friend and appears on your device.

[0283] Output: The received message displayed.

[0284] Step 2: Sending the message to the server

[0285] Input: The received message displayed on the terminal.

[0286] Operation:

[0287] The terminal transmits the received message to the server as text data.

[0288] Messages are sent to the server via HTTP POST requests.

[0289] Output: The text data received by the server.

[0290] Step 3: Send a text analysis request

[0291] Input: Text data received by the server.

[0292] Operation:

[0293] The server sends a request to the text analysis engine to analyze the message.

[0294] A text analysis engine (e.g., SpaCy, NLTK) analyzes the message content and extracts key elements.

[0295] Output: Data containing the parsed key elements.

[0296] Step 4: Send a sentiment analysis request

[0297] Input: Text data received by the server.

[0298] Operation:

[0299] The server sends a request to the sentiment analysis engine to analyze the message.

[0300] A sentiment analysis engine (e.g., TextBlob, VADER) analyzes the message and identifies the user's sentiment.

[0301] Output: Parsed emotion information.

[0302] Step 5: Obtaining user's past message data

[0303] Input: The server accesses the database based on the user ID.

[0304] Operation:

[0305] The server executes a database query to retrieve the user's past messaging data and communication style.

[0306] A database (e.g. SQL, NoSQL) returns the user history.

[0307] Output: User's past messaging data and communication style.

[0308] Step 6: Generate reply suggestions

[0309] Input: analyzed key elements, sentiment information, user's previous messages and communication style.

[0310] Operation:

[0311] The server uses this data to input prompt sentences into the generative AI model.

[0312] A generative AI model (e.g., TensorFlow, PyTorch) generates potential replies (example prompt: "Generate a reply when a friend invites you to the movies.").

[0313] Output: Multiple reply suggestions.

[0314] Step 7: Send reply suggestions to your device

[0315] Input: The generated multiple reply suggestions.

[0316] Operation:

[0317] The server transmits the generated reply candidates to the terminal.

[0318] Messages are sent to the device via HTTP POST requests.

[0319] Output: Suggested replies displayed on the device's user interface (UI).

[0320] Step 8: Select a reply suggestion

[0321] Input: Multiple reply suggestions displayed in the device's user interface.

[0322] Operation:

[0323] The user selects the appropriate reply suggestion.

[0324] The terminal sends the selected reply message to the server.

[0325] Output: The selected reply message.

[0326] Step 9: Sending a final reply

[0327] Input: The selected reply message received by the server.

[0328] Operation:

[0329] The server delivers the received message to the friend's terminal.

[0330] Messages are transferred using a real-time communication framework (e.g., WebSocket, Socket.IO).

[0331] Output: The final reply message displayed on your friend's device.

[0332] The above is a concrete explanation of the processing of the system program, broken down into steps, and shows the flow including input and output at each processing step.

[0333] (Application example 2)

[0334] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0335] Modern self-driving vehicles lack dialogue systems that take user emotions into account, resulting in a lack of a satisfactory user experience. Furthermore, navigation and environmental adjustments do not provide optimal support based on the user's mood and state, resulting in low user satisfaction. To address these issues, a system incorporating user emotion analysis is needed.

[0336] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a receiving means, a text analysis means that analyzes received text data and extracts key elements, a reply generation means that generates reply candidates based on the user's past message history and communication style, a transmission means that transmits the generated reply candidates to the user terminal, and in an in-car dialogue system, a sentiment analysis means that analyzes the sentiment of the user's message, a model generation means that uses a generative AI model to generate optimal reply candidates based on the analysis results, and an execution means that performs navigation and environmental adjustment using prompt sentences. This makes it possible to analyze the user's sentiment and adjust the navigation and environment optimally.

[0337] "Receiving means" refers to a device or system that receives messages or data from a user.

[0338] "Text analysis means" refers to technology or devices that analyze received text data and extract key elements and intent.

[0339] A "reply generator" is a system or algorithm that generates appropriate reply suggestions based on a user's past messaging history and communication style.

[0340] "Transmission means" refers to a technique or device for transmitting the generated reply candidates to a user terminal.

[0341] An "interactive system" is a system or device that enables two-way communication between a user and the system.

[0342] "Emotion analysis means" refers to technology or devices that analyze and recognize emotions from user messages.

[0343] A "generative AI model" is a model or algorithm that uses artificial intelligence to automatically generate appropriate reply suggestions or actions.

[0344] A "prompt" is a sentence or phrase that can be input into a generative AI model to elicit a specific outcome or response.

[0345] "Navigation" refers to systems and functions that present and guide vehicles and users on the optimal route to their destination.

[0346] "Environmental control" refers to technologies and functions that adjust in-car conditions (temperature, music, lighting, etc.) to improve user comfort.

[0347] The present invention relates to a system and method for analyzing user sentiment in dialogue systems within autonomous vehicles, and providing appropriate responses and navigation support. The system aims to provide a more personalized experience by combining user sentiment analysis with generative AI models.

[0348] System configuration

[0349] 1. User Device:

[0350] A device that allows users to input information, such as a smartphone or an in-car display.

[0351] 2. Server:

[0352] It is a high-performance server that uses an instance of a cloud service provider (e.g., a virtual machine of a cloud service). The server is configured to include a text analysis engine, a sentiment analysis engine, and a generative AI model.

[0353] 3. Text Analysis Engine:

[0354] The received message is analyzed and its main elements and intent are extracted. Specifically, an engine that uses natural language processing technology (for example, the API of a natural language processing cloud service provider) is used.

[0355] 4. Sentiment Analysis Engine:

[0356] An engine for analyzing and recognizing emotions from user messages. For example, we use a cloud-based emotion analysis engine.

[0357] 5. Generative AI Models:

[0358] An AI model that generates appropriate reply candidates and actions based on the results of user sentiment analysis and past data. For example, it is possible to use the latest model of generative AI.

[0359] 6. Dialogue Systems:

[0360] It is an interface for realizing two-way communication within the vehicle, and includes execution means for executing generated responses and actions.

[0361] Specific operation of the system

[0362] In this system, the user inputs a message into the in-car dialogue system using voice or text. The user's device then sends this message to the server. The server first analyzes the message content using a text analysis engine to extract key elements. It then uses an emotion analysis engine to recognize the user's emotions. Once the emotion analysis is complete, the server's generative AI model generates optimal reply candidates based on the analysis results and past data.

[0363] The generated reply candidates are sent to the user's device and displayed on the in-car display. The user can select an appropriate reply from the displayed candidates. The selected reply or action is then executed by the dialogue system, adjusting the navigation or the in-car environment.

[0364] Examples and prompts

[0365] Specific examples

[0366] If a user is driving and says to the in-car system, "I'm tired today and want to get home early," the system will:

[0367] 1. The user device sends a message to the server.

[0368] 2. The text analysis engine analyzes "I want to go home early" and extracts the main intent.

[0369] 3. The emotion analysis engine detects "fatigue."

[0370] 4. The generative AI model generates the following reply suggestions:

[0371] "Thank you for your hard work. I'm calculating the quickest route home."

[0372] "Play some relaxing music?"

[0373] 5. Candidates are displayed on the in-car display and the user selects one.

[0374] 6. The selected reply is executed and navigation changes to the shortest route.

[0375] Prompt Sentence Examples

[0376] If user fatigue is detected:

[0377] "User fatigue detected. Suggest navigation for the quickest route home."

[0378] "The user is tired, so please suggest three relaxing music pieces."

[0379] Using the above configuration, the present invention can provide a more personalized driving experience that takes into account the user's emotions, and realize comfortable communication and navigation.

[0380] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0381] Step 1:

[0382] The user inputs a message to the in-car dialogue system.

[0383] Input: User's voice or text message

[0384] Output: Message data

[0385] Action: The user tells the dialogue system, "I'm tired today and want to go home early."

[0386] Step 2:

[0387] The terminal sends the received message to the server.

[0388] Input: Message data

[0389] Output: Message data sent to the server

[0390] How it works: The device digitizes the user's message and sends it to the server.

[0391] Step 3:

[0392] The server sends the received message to a text analysis engine to extract key elements.

[0393] Input: Message data

[0394] Output: Key element data (e.g., "I want to go home early")

[0395] How it works: A text analysis engine in the server analyzes the message and extracts the main idea.

[0396] Step 4:

[0397] The server sends the results of the text analysis engine to the sentiment analysis engine to identify the user's sentiment.

[0398] Input: Main element data

[0399] Output: Emotion data (e.g., "fatigue")

[0400] How it works: The sentiment analysis engine analyzes and identifies the user's sentiment based on the text analysis results.

[0401] Step 5:

[0402] The server generates reply suggestions using a generative AI model based on the sentiment analysis results and past user data.

[0403] Input: Emotion data, past user data

[0404] Output: Reply candidate data

[0405] How it works: A generative AI model combines sentiment data with past user data to generate appropriate reply suggestions.

[0406] Step 6:

[0407] The server sends the generated reply candidates to the terminal and presents them to the user.

[0408] Input: Reply candidate data

[0409] Output: Suggested replies displayed on the user's device

[0410] How it works: The server sends reply candidates to the device, which are then displayed on the in-car display.

[0411] Step 7:

[0412] The user selects from the suggested replies.

[0413] Type: Show suggested replies

[0414] Output: Selected reply suggestion

[0415] What happens: The user selects the reply suggestion "Great work! Calculating the quickest route home."

[0416] Step 8:

[0417] The terminal sends the selected reply to the server.

[0418] Enter: Selected reply suggestion

[0419] Output: Selected replies sent to the server

[0420] How it works: The device sends the user's selection to the server.

[0421] Step 9:

[0422] The server performs navigation and environmental adjustments based on the selected reply.

[0423] Input: Selected reply data

[0424] Output: Navigation route, environment settings

[0425] How it works: The server sends commands to the navigation system to calculate and set the shortest route, and also makes appropriate environmental adjustments (e.g., playing music).

[0426] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0427] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0428] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0429] [Second embodiment]

[0430] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0431] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0432] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0433] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0434] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0435] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0436] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0437] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0438] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0439] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0440] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0441] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0442] The present invention relates to a system and method for making users' message replies more efficient and supporting appropriate communication, and detailed embodiments thereof are described below.

[0443] The present invention is a system consisting of a user terminal, a server, a text analysis engine, and an AI model, which supports a user in replying promptly and appropriately to a message received.

[0444] Overview of program processing flow

[0445] Receiving and parsing messages

[0446] 1. A user receives a message using a chat application, for example, from a friend, "Want to go see a movie with me this weekend?"

[0447] 2. The device sends the received message to the server, where it is sent as text data via a communication protocol.

[0448] 3. The server's text analysis engine receives the message and begins analyzing it. The text analysis engine uses natural language processing technology to understand the subject and intent of the message. For example, it extracts keywords such as "weekend," "movie," and "go together" and classifies the intent of the message as a "request to confirm plans."

[0449] Generate reply suggestions

[0450] 4. The server accesses the user's past message history database and evaluates the user's communication style, thereby understanding how the user has responded in the past and how they prefer to express themselves.

[0451] 5. The server's AI model generates appropriate reply suggestions based on the text analysis results and the user's communication style. For example, suggestions like the following might be generated:

[0452] "Thanks, I'd love to go! What time would be good?"

[0453] "Sorry, I have plans this weekend so I can't make it. See you next time."

[0454] I haven't decided on a schedule yet, so I'll think about it for a bit and get back to you.

[0455] Sending and selecting suggested replies

[0456] 6. The server sends the generated reply candidates to the device, where they are displayed on the user interface (UI).

[0457] 7. The user selects the appropriate reply from the suggested replies, for example, "Thanks, I'd love to! What time would be good?"

[0458] 8. The terminal sends the selected reply message to the server.

[0459] 9. The server delivers the selected reply message to the recipient friend, thereby completing the communication promptly and appropriately.

[0460] Specific examples

[0461] For example, when a user receives a message from a friend saying, "Would you like to go see a movie with me this weekend?", the system works as follows:

[0462] When a user receives a message, the device sends it to the server. The server uses a text analysis engine to analyze the message and extract key elements. In this case, the extracted result is "Request to confirm schedule." The server then generates appropriate reply candidates for this element based on the user's past message history. The generated reply candidates are then sent to the device and displayed to the user. For example, a candidate might be "Thanks, I'd love to go! What time would be good?" The user selects an appropriate reply and presses the send button on the device, which then sends the reply to their friend via the server.

[0463] In this way, the present invention supports users in responding to messages quickly and appropriately, thereby improving the efficiency of user communication.

[0464] The processing flow will be explained below.

[0465] Step 1:

[0466] A user has a chat app open and receives a message from a friend saying, "Want to go see a movie with me this weekend?"

[0467] Step 2:

[0468] The terminal displays the received message and simultaneously sends the message as text data to the server.

[0469] Step 3:

[0470] The server sends the received message as a request to the text analysis engine, which uses natural language processing technology to analyze the content of the message.

[0471] Step 4:

[0472] The text analysis engine analyzes the message and extracts key elements (e.g., "weekend," "movie," and "going together"). As a result of the analysis, the intent of the message is classified as a "request to confirm plans."

[0473] Step 5:

[0474] The server accesses the user's past message history database to acquire the user's communication style, thereby learning what expressions the user has used in the past.

[0475] Step 6:

[0476] The server's AI model generates multiple reply suggestions based on the text analysis results and the user's communication style, for example:

[0477] "Thanks, I'd love to go! What time would be good?"

[0478] "Sorry, I have plans this weekend so I can't make it. See you next time."

[0479] I haven't decided on a schedule yet, so I'll think about it for a bit and get back to you.

[0480] Step 7:

[0481] The server generates reply suggestions and sends them to the device, where they are displayed to the user.

[0482] Step 8:

[0483] The user selects the appropriate reply from the suggested replies, for example, "Thanks, I'd love to! What time would be good?"

[0484] Step 9:

[0485] The terminal sends the selected reply message to the server.

[0486] Step 10:

[0487] The server delivers the final reply message to the recipient friend, who then displays the message on their device.

[0488] This series of steps allows users to quickly and appropriately reply after receiving a message.

[0489] Example 1

[0490] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0491] In modern communication, users are expected to respond quickly and appropriately. However, constantly rethinking the content of their replies to messages is a time-consuming task for users. Therefore, there is a need for a way for users to efficiently generate replies to received messages and communicate quickly.

[0492] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0493] In this invention, the server includes a receiving means, a text analyzing means for analyzing the received text data and extracting key elements, a reply generating means for generating reply candidates based on the user's past message history and communication style, a transmitting means for transmitting the generated reply candidates to the user terminal, and a distributing means for receiving a reply message selected from the user terminal and transmitting it to the receiving terminal, thereby enabling the user to reply quickly and appropriately to the message received.

[0494] The "receiving means" is a means for processing a message received by a user through a chat application and sending it to a server.

[0495] The "text analysis means" is a means for analyzing received text data and extracting key elements and keywords.

[0496] The "reply generation means" is a means for generating appropriate reply candidates based on the elements and keywords extracted by the text analysis means, as well as the user's past message history and communication style.

[0497] The "transmitting means" is a means for transmitting the generated reply candidates to the user terminal and displaying them on the user interface of the user terminal.

[0498] The "distribution means" is a means for sending a selected reply message from a user terminal back to the server, and finally sending the reply message to the receiving terminal.

[0499] "Natural language processing technology" is a field of computer science that is a technology for understanding, analyzing, and generating human language.

[0500] An "AI model" is an algorithm or data structure that uses artificial intelligence to perform various tasks, and in natural language processing in particular, it is a model for generating and analyzing text.

[0501] The present invention relates to a system and method for improving the efficiency of user message replies and supporting appropriate communication. The system comprises a user terminal, a server, a text analysis engine, and an AI model.

[0502] Receiving and parsing messages

[0503] First, a user receives a message using a chat application. For example, the received message might say, "Would you like to go see a movie with me this weekend?" The device receives this message and sends it to the server. The HTTP POST request is typically used as the communication protocol.

[0504] The server passes the received message to a text analysis engine. This text analysis engine uses natural language processing technologies, such as Google Cloud Natural Language API or IBM Watson Natural Language Understanding. The text analysis engine analyzes the message and extracts key elements and keywords. Specifically, the keywords "weekend," "movie," and "go together" are extracted, and the intent of the message is classified as a "request to confirm plans."

[0505] Generate reply suggestions

[0506] The server then accesses a database of the user's past messages, which can be a relational or NoSQL database like MySQL or MongoDB, and runs a database query to retrieve the user's past reply patterns and communication style.

[0507] Based on the acquired data, the server uses an AI model (e.g., OpenAI's GPT-3 or BERT) to generate appropriate reply suggestions. This AI model receives text analysis results and the user's communication style as input data. The generated reply suggestions include:

[0508] "Thanks, I'd love to go! What time would be good?"

[0509] "Sorry, I have plans this weekend so I can't make it. See you next time."

[0510] I haven't decided on a schedule yet, so I'll think about it for a bit and get back to you.

[0511] Sending and selecting suggested replies

[0512] The generated reply candidates are sent by the server to the device. The transmission is done through an HTTP response, and the reply candidates are encoded in JSON format. The device decodes them and displays them in the application's user interface. The user can select the appropriate reply from multiple reply candidates. For example, the reply "Thanks, I'd love to go! What time would be good?" is selected.

[0513] The selected reply message is then sent back to the server, which then delivers it to the recipient device using the appropriate protocol (e.g., XMPP or MQTT) for the chat application server, ensuring fast and accurate communication.

[0514] Specific examples

[0515] For example, if a user receives a message from a friend saying, "Want to go see a movie with me this weekend?", the system operates as follows:

[0516] The user receives the message.

[0517] 2. The device sends the message to the server.

[0518] 3. The server passes the message to a text analysis engine, which extracts the key elements: "weekend," "movie," and "going together."

[0519] 4. The server accesses the user's history database to check past response patterns.

[0520] 5. The server's AI model generates appropriate reply suggestions.

[0521] 6. The server sends the generated reply candidates to the terminal, which displays them.

[0522] 7. The user selects the reply "Thanks, I'd love to! What time would be good?"

[0523] 8. The device sends the selected reply to the server, which distributes it to the friend.

[0524] In this way, the present invention supports users in quickly and appropriately replying to messages they receive, thereby improving the efficiency of user communication.

[0525] Example prompts to input to the generative AI model

[0526] Generate a suitable reply to a message from a friend saying "Want to go see a movie with me this weekend?". The output should be in the following format:

[0527] 1. "Reply candidate #1"

[0528] 2. "Reply candidate #2"

[0529] 3. "Reply candidate #3"

[0530] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0531] Step 1:

[0532] A user receives a message using a chat application. For example, a friend might send a message saying, "Want to go see a movie with me this weekend?", and the message is received by the user's device. The input is the message from the friend, and the output is the received message on the device.

[0533] Step 2:

[0534] The terminal sends the message received to the server. Specifically, it uses an HTTP POST request to send the message content to the server. The input is the received message, and the output is the text data sent to the server.

[0535] Step 3:

[0536] The server passes the received message to the text analysis engine and begins analysis. The text analysis engine analyzes the content of the message and extracts key elements and keywords. For example, the keywords "weekend," "movie," and "go together" are extracted using the Google Cloud Natural Language API. The input is the text data sent to the server, and the output is the analysis results (the extraction of key elements).

[0537] Step 4:

[0538] The server accesses a database of the user's past message history. It executes a database query to obtain past reply patterns and the user's communication style. The database can be, for example, MySQL or MongoDB. The input is the analysis result and the user's ID, and the output is the past message history data.

[0539] Step 5:

[0540] The server uses an AI model to generate appropriate reply suggestions. Based on the analysis results and past message history data, the AI ​​model (e.g., OpenAI's GPT-3) generates reply suggestions. The generated reply suggestions include:

[0541] "Thanks, I'd love to go! What time would be good?"

[0542] "Sorry, I have plans this weekend so I can't make it. See you next time."

[0543] I haven't decided on a schedule yet, so I'll think about it for a bit and get back to you.

[0544] The input is the analysis results and the user's message history data, and the output is a list of reply candidates.

[0545] Step 6:

[0546] The server generates reply candidates and sends them to the terminal. The HTTP response is used to send reply candidates encoded in JSON format to the terminal. The input is the reply candidate list, and the output is the data sent to the terminal.

[0547] Step 7:

[0548] The device decodes the reply candidates received and displays them on the user interface. The user selects the appropriate reply from the displayed multiple reply candidates. The input is the reply candidate data sent from the server, and the output is the candidates displayed on the UI. For example, the reply "Thanks, I'd love to go! What time would be good?" is displayed.

[0549] Step 8:

[0550] The user confirms the selected reply on the terminal and sends the reply message to the server. Again, an HTTP POST request is used to send the selected reply message to the server. The input is the reply message selected by the user, and the output is the data sent to the server.

[0551] Step 9:

[0552] The server receives the user's reply and delivers it to the recipient friend. It sends the reply message to the friend's device using an appropriate communication protocol (e.g., XMPP or MQTT). The input is the reply message selected by the user, and the output is the message delivered to the recipient's device. This completes fast and appropriate communication.

[0553] (Application example 1)

[0554] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0555] In recent years, security systems have been required to process warning messages quickly and accurately and to take appropriate action. However, the volume and variety of warning messages received by users has made it difficult to respond quickly. Therefore, it is necessary to develop a system that can instantly understand the content of warning messages and generate appropriate responses.

[0556] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0557] In this invention, the server includes a receiving means, a text analyzing means, a reply generating means, a sending means, and a security analyzing means, which enable the server to quickly and accurately analyze the received warning message, generate appropriate reply candidates, and present them to the user.

[0558] "Receiving means" refers to devices or software for receiving data or messages from the outside.

[0559] "Text analysis means" refers to a device or software for analyzing received text data and extracting key elements.

[0560] A "reply generator" is a device or software that generates appropriate reply candidates based on a user's past message history and communication style.

[0561] "Transmission means" refers to a device or software for transmitting the generated reply candidates to the user terminal.

[0562] "Security analysis means" refers to devices or software that analyze received warning messages and provide appropriate response candidates.

[0563] The present invention relates to a system that improves the efficiency of users' message replies and supports appropriate communication. The system includes a receiving unit, a text analyzing unit, a reply generating unit, a sending unit, and a security analyzing unit.

[0564] The user uses the security assistant application to receive warning messages sent from the security system or camera. This reception can be performed using an application installed on a device such as a smartphone. The received messages are then sent to the server.

[0565] The server analyzes the received text data using natural language processing technology based on Hugging Face's Transformers library. The text analysis method extracts key elements of the received warning message and uses this to understand the message's content and intent.

[0566] Based on the analysis results, the reply generator evaluates the user's past message history and communication style to generate appropriate reply candidates, using the user's response history database as well.

[0567] The generated reply candidates are sent to the user terminal by a sending means, and the user selects an appropriate reply from the presented reply candidates through the application interface.

[0568] For example, if a warning message is received stating, "The window sensor has detected an abnormality," the server analyzes the message and generates appropriate response candidates, such as, "We will call the police," or "Please check the security camera footage." These response candidates are displayed on the user's device, and the user selects an appropriate response, which is then sent via the server to the security system.

[0569] An example prompt is:

[0570] Alert message: "Window sensor has detected an abnormality"

[0571] Potential responses generated: "Call the police." "Check your security camera footage." "I need your help right away."

[0572] In this way, the present invention can quickly and accurately analyze received warning messages, generate appropriate response candidates, and present them to the user, thereby streamlining the user's security response and providing a safer environment.

[0573] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0574] Step 1:

[0575] The user uses the security assistant application to receive an alert message. This input is an alert message sent from a security system or camera in a home or office. The device receives this and sends it to the next step.

[0576] Step 2:

[0577] The terminal sends the received message to the server. In this process, the warning message is transferred to the server as text data. The input data is the warning message sent via the terminal, and the output data is the text data received by the server.

[0578] Step 3:

[0579] The server uses the Hugging Face Transformers library to analyze the received text data. The input data is the received text data, and the analyzed results include the main elements of the message (e.g., "sensor abnormality" and "window").

[0580] Step 4:

[0581] The server uses security analysis means to understand the contents of the warning message based on the analysis results and determines an appropriate response. This process determines what response is appropriate based on the analysis results (e.g., "sensor abnormality"). The input data is the analysis results, and the output data is information about the intended response (e.g., "warning response").

[0582] Step 5:

[0583] The reply generator evaluates the user's past message history and communication style to generate appropriate reply candidates. The input data are the analysis results and user history data, and the output data are the generated reply candidates (e.g., "I will call the police." "Please check the security camera footage.").

[0584] Step 6:

[0585] The server transmits the generated reply candidates to the user terminal, where the input data are the generated reply candidates and the output data are the multiple reply candidates displayed on the user terminal.

[0586] Step 7:

[0587] The user selects an appropriate reply from the presented reply candidates. The input data are the reply candidates presented to the user, and the output data is the reply selected by the user.

[0588] Step 8:

[0589] The terminal sends the selected reply message to the server, which delivers the reply to the security system. The input data is the reply selected by the user, and the output data is the final reply sent to the security system.

[0590] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0591] The present invention relates to a system and method for improving the efficiency of users' message replies and supporting appropriate communication, and is characterized in that reply candidates are generated taking into account the user's emotions by combining an emotion engine. Detailed embodiments are described below.

[0592] This system consists of a user terminal, a server, a text analysis engine, an AI model, and an emotion engine. The operation flow of the system will be explained in detail below.

[0593] Overview of program processing flow

[0594] Receiving and parsing messages

[0595] 1. A user receives a message using a chat application, for example, from a friend, "Want to go see a movie with me this weekend?"

[0596] 2. The terminal displays the received message and sends the message as text data to the server.

[0597] 3. The server sends the received message as a request to the text analysis engine and emotion engine. The analysis engine uses natural language processing technology to analyze the content of the message and extract key elements.

[0598] Emotion Analysis

[0599] 4. The emotion engine analyzes the message and recognizes the user's emotions. For example, emotions such as "happiness," "sadness," and "surprise" are identified by the emotion engine. In this case, the message "Would you like to go see a movie with me?" is analyzed as "curiosity" and "anticipation."

[0600] Generate reply suggestions

[0601] 5. The server accesses the user's past message history database to obtain the user's communication style, taking into account the analyzed emotional information.

[0602] 6. The server's AI model generates appropriate reply suggestions based on the text analysis results, the user's communication style, and emotional information. For example, the following suggestions might be generated:

[0603] "Thanks, I'd love to go! Sounds interesting."

[0604] "Sorry, I have plans this weekend so I can't make it. See you next time."

[0605] I haven't decided on a schedule yet, so I'll think about it for a bit and get back to you.

[0606] Sending and selecting suggested replies

[0607] 7. The server sends the generated reply candidates to the device, where they are displayed on the user interface (UI).

[0608] 8. The user selects the appropriate reply from the suggested replies displayed. For example, the user selects the reply "Thanks, I'd love to go! Sounds interesting."

[0609] 9. The terminal sends the selected reply message to the server.

[0610] 10. The server delivers the final reply message to the recipient friend, who then displays the message on their device.

[0611] Specific examples

[0612] For example, if a user receives a message from a friend saying, "Want to go see a movie with me this weekend?", the system works as follows:

[0613] While a user has a chat app open, they receive a message from a friend. The device sends this message to a server, which then asks the text analysis engine and emotion engine to analyze the message. The text analysis engine analyzes the message and classifies it as a "request to confirm plans," and the emotion engine extracts the emotions of "curiosity" and "anticipation." The server generates reply candidates based on the user's past message history and the results of these analyses. For example, a candidate might be "Thanks, I'd love to go! Sounds interesting." The server sends this to the device so the user can select one. After the user selects and sends an appropriate reply, the server distributes the reply to the friend.

[0614] In this way, the present invention provides a system that takes into account the user's emotions to create more personalized replies, making users' communications more efficient.

[0615] The processing flow will be explained below.

[0616] Step 1:

[0617] A user is using a chat app and receives a message from a friend saying, "Want to go see a movie with me this weekend?"

[0618] Step 2:

[0619] The terminal sends this message as text data to the server.

[0620] Step 3:

[0621] The server sends the received message as a request to the text analysis engine, which uses natural language processing technology to analyze the content of the message and extract key elements.

[0622] Step 4:

[0623] A text analysis engine analyzes the message, extracts key elements (e.g., "weekend," "movie," "going together"), and classifies the intent of the message as "a request to confirm plans."

[0624] Step 5:

[0625] The server simultaneously sends the message to the emotion engine, which uses natural language processing technology to analyze the emotions in the message.

[0626] Step 6:

[0627] The emotion engine analyzes messages and identifies emotions. For example, for the message "Would you like to go see a movie with me?", it extracts the emotions "curiosity" and "anticipation."

[0628] Step 7:

[0629] The server accesses a database of the user's past messages to obtain the user's communication style, for example, whether the user previously preferred polite replies.

[0630] Step 8:

[0631] The server's AI model generates multiple reply suggestions based on the text analysis results, the user's communication style, and emotional information, such as:

[0632] "Thanks, I'd love to go! Sounds interesting."

[0633] "Sorry, I have plans this weekend so I can't make it. Let's go next time."

[0634] I haven't decided on a schedule yet, so I'll think about it for a bit and get back to you.

[0635] Step 9:

[0636] The server sends these reply suggestions to the device, which then displays them on the user interface (UI).

[0637] Step 10:

[0638] The user selects the appropriate reply from the suggested replies, for example, "Thanks, I'd love to go! Sounds interesting."

[0639] Step 11:

[0640] The terminal sends the selected reply message to the server.

[0641] Step 12:

[0642] The server delivers the final reply message to the friend's device, which displays "Thanks, I'd love to go! Sounds interesting."

[0643] This series of steps allows users to quickly respond appropriately after receiving a message. By taking into account the user's emotions, the system provides more personalized replies and achieves effective communication.

[0644] Example 2

[0645] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0646] Conventional message reply systems have difficulty generating personalized and appropriate replies because they do not take into account the user's emotions or past communication styles. As a result, improvements to communication efficiency and user experience have been limited.

[0647] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0648] In this invention, the server includes a receiving means, a text analysis means for analyzing the received text data and extracting key elements, a sentiment analysis means for recognizing the user's sentiment based on the analyzed data, a reply generation means for generating appropriate reply candidates based on the user's past message history and communication style, and a transmission means for transmitting the generated reply candidates to the user terminal. This makes it possible to generate and provide personalized and appropriate reply candidates that take into account the user's sentiment and past communication style.

[0649] "Receiving means" means a piece of hardware or software designed to receive messages or data from a user.

[0650] "Text analysis means" means a piece of hardware or software that has the function of analyzing received text data and extracting key elements or information.

[0651] "Emotion analysis means" means a piece of hardware or software that employs technology to recognize and identify a user's emotions based on analyzed data.

[0652] A "reply generator" is a piece of hardware or software that takes into account a user's past message history and communication style to generate appropriate reply candidates.

[0653] The "transmitting means" is a piece of hardware or software for transferring the generated reply candidates to the user terminal.

[0654] "Natural language processing technology" is a technology that enables computers to understand, interpret, and generate human language.

[0655] A "user terminal" is a device that is directly operated by a user, and includes computers, smartphones, tablets, etc.

[0656] "Communication style" refers to characteristics such as tone, vocabulary, and expression used in messages a user has used in the past.

[0657] "Suggested replies" are system-generated suggested messages for a user to send.

[0658] A "database" is a structured collection of data for efficient storage and retrieval of specific information.

[0659] A "generative AI model" is a model that uses artificial intelligence technology to generate appropriate output based on user input.

[0660] The present invention relates to a system and method for improving the efficiency of users' message replies and supporting appropriate communication. The system includes a user terminal, a server, a text analysis engine, an AI model, and a sentiment analysis engine. The operational flow of this system is specifically shown below.

[0661] Hardware and Software

[0662] The system uses the following components:

[0663] User devices: computers, smartphones, tablets, etc.

[0664] Server: Cloud service or on-premise server.

[0665] Text analysis engines: Natural language processing libraries (e.g., SpaCy, NLTK).

[0666] Sentiment Analysis Engine: A sentiment analysis library (e.g., TextBlob, VADER).

[0667] Generative AI models: Models that use artificial intelligence techniques (e.g., TensorFlow, PyTorch).

[0668] Receiving and parsing messages

[0669] A user uses a chat application and receives a message, for example, "Would you like to go see a movie with me this weekend?" The device displays this message and sends it as text data to the server. The server then sends the received message as a request to the text analysis engine and sentiment analysis engine. The text analysis engine analyzes the received message and extracts key elements.

[0670] Emotion Analysis

[0671] The emotion analysis engine analyzes messages to recognize the user's emotions. For example, in the message "Would you like to go see a movie with me this weekend?", it identifies the emotions "curiosity" and "anticipation."

[0672] Generate reply suggestions

[0673] The server accesses the user's past message history database to obtain the user's consistent communication style. Based on the analyzed text data and emotional information, the generative AI model generates reply candidates. For example, the following candidates are generated:

[0674] "Thanks, I'd love to go! Sounds interesting."

[0675] "Sorry, I have plans this weekend so I can't make it. See you next time."

[0676] I haven't decided on a schedule yet, so I'll think about it for a bit and get back to you.

[0677] Sending and selecting suggested replies

[0678] The server generates reply candidates and sends them to the device, which displays them on the user interface (UI). The user selects the most appropriate reply from the displayed candidates. For example, if the user selects "Thanks, I'd love to go! Sounds interesting," the device sends the selected reply message to the server. The server then distributes the final reply message to the friend's device, where the message is displayed.

[0679] Specific examples

[0680] For example, if a user receives a message from a friend saying, "Want to go see a movie with me this weekend?", the system works as follows:

[0681] While a user has a chat app open, the user receives a message from a friend. The device sends this message to a server, which then asks the text analysis engine and sentiment analysis engine to analyze the message. The text analysis engine analyzes the message and classifies it as a "request to confirm plans," and the sentiment analysis engine extracts the emotions of "curiosity" and "anticipation." The server generates reply candidates based on the user's past message history and the results of these analyses. For example, a candidate might be "Thanks, I'd love to go! Sounds interesting." The server sends this to the device so the user can select one. After the user selects and sends an appropriate reply, the server distributes the reply to the friend.

[0682] In this way, the present invention provides a system that takes into account the user's emotions to create more personalized replies, making users' communications more efficient.

[0683] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0684] Processing Steps

[0685] Step 1: Receiving a message

[0686] Input: A user receives a message using a chat application (e.g., "Want to go see a movie with me this weekend?").

[0687] Operation:

[0688] The user opens a chat app.

[0689] A message is sent from a friend and appears on your device.

[0690] Output: The received message displayed.

[0691] Step 2: Sending the message to the server

[0692] Input: The received message displayed on the terminal.

[0693] Operation:

[0694] The terminal transmits the received message to the server as text data.

[0695] Messages are sent to the server via HTTP POST requests.

[0696] Output: The text data received by the server.

[0697] Step 3: Send a text analysis request

[0698] Input: Text data received by the server.

[0699] Operation:

[0700] The server sends a request to the text analysis engine to analyze the message.

[0701] A text analysis engine (e.g., SpaCy, NLTK) analyzes the message content and extracts key elements.

[0702] Output: Data containing the parsed key elements.

[0703] Step 4: Send a sentiment analysis request

[0704] Input: Text data received by the server.

[0705] Operation:

[0706] The server sends a request to the sentiment analysis engine to analyze the message.

[0707] A sentiment analysis engine (e.g., TextBlob, VADER) analyzes the message and identifies the user's sentiment.

[0708] Output: Parsed emotion information.

[0709] Step 5: Obtaining user's past message data

[0710] Input: The server accesses the database based on the user ID.

[0711] Operation:

[0712] The server executes a database query to retrieve the user's past messaging data and communication style.

[0713] A database (e.g. SQL, NoSQL) returns the user history.

[0714] Output: User's past messaging data and communication style.

[0715] Step 6: Generate reply suggestions

[0716] Input: analyzed key elements, sentiment information, user's previous messages and communication style.

[0717] Operation:

[0718] The server uses this data to input prompt sentences into the generative AI model.

[0719] A generative AI model (e.g., TensorFlow, PyTorch) generates potential replies (example prompt: "Generate a reply when a friend invites you to the movies.").

[0720] Output: Multiple reply suggestions.

[0721] Step 7: Send reply suggestions to your device

[0722] Input: The generated multiple reply suggestions.

[0723] Operation:

[0724] The server transmits the generated reply candidates to the terminal.

[0725] Messages are sent to the device via HTTP POST requests.

[0726] Output: Suggested replies displayed on the device's user interface (UI).

[0727] Step 8: Select a reply suggestion

[0728] Input: Multiple reply suggestions displayed in the device's user interface.

[0729] Operation:

[0730] The user selects the appropriate reply suggestion.

[0731] The terminal sends the selected reply message to the server.

[0732] Output: The selected reply message.

[0733] Step 9: Sending a final reply

[0734] Input: The selected reply message received by the server.

[0735] Operation:

[0736] The server delivers the received message to the friend's terminal.

[0737] Messages are transferred using a real-time communication framework (e.g., WebSocket, Socket.IO).

[0738] Output: The final reply message displayed on your friend's device.

[0739] The above is a concrete explanation of the processing of the system program, broken down into steps, and shows the flow including input and output at each processing step.

[0740] (Application example 2)

[0741] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0742] Modern self-driving vehicles lack dialogue systems that take user emotions into account, resulting in a lack of a satisfactory user experience. Furthermore, navigation and environmental adjustments do not provide optimal support based on the user's mood and state, resulting in low user satisfaction. To address these issues, a system incorporating user emotion analysis is needed.

[0743] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a receiving means, a text analysis means that analyzes received text data and extracts key elements, a reply generation means that generates reply candidates based on the user's past message history and communication style, a transmission means that transmits the generated reply candidates to the user terminal, and in an in-car dialogue system, a sentiment analysis means that analyzes the sentiment of the user's message, a model generation means that uses a generative AI model to generate optimal reply candidates based on the analysis results, and an execution means that performs navigation and environmental adjustment using prompt sentences. This makes it possible to analyze the user's sentiment and adjust the navigation and environment optimally.

[0744] "Receiving means" refers to a device or system that receives messages or data from a user.

[0745] "Text analysis means" refers to technology or devices that analyze received text data and extract key elements and intent.

[0746] A "reply generator" is a system or algorithm that generates appropriate reply suggestions based on a user's past messaging history and communication style.

[0747] "Transmission means" refers to a technique or device for transmitting the generated reply candidates to a user terminal.

[0748] An "interactive system" is a system or device that enables two-way communication between a user and the system.

[0749] "Emotion analysis means" refers to technology or devices that analyze and recognize emotions from user messages.

[0750] A "generative AI model" is a model or algorithm that uses artificial intelligence to automatically generate appropriate reply suggestions or actions.

[0751] A "prompt" is a sentence or phrase that can be input into a generative AI model to elicit a specific outcome or response.

[0752] "Navigation" refers to systems and functions that present and guide vehicles and users on the optimal route to their destination.

[0753] "Environmental control" refers to technologies and functions that adjust in-car conditions (temperature, music, lighting, etc.) to improve user comfort.

[0754] The present invention relates to a system and method for analyzing user sentiment in dialogue systems within autonomous vehicles, and providing appropriate responses and navigation support. The system aims to provide a more personalized experience by combining user sentiment analysis with generative AI models.

[0755] System configuration

[0756] 1. User Device:

[0757] A device that allows users to input information, such as a smartphone or an in-car display.

[0758] 2. Server:

[0759] It is a high-performance server that uses an instance of a cloud service provider (e.g., a virtual machine of a cloud service). The server is configured to include a text analysis engine, a sentiment analysis engine, and a generative AI model.

[0760] 3. Text Analysis Engine:

[0761] The received message is analyzed and its main elements and intent are extracted. Specifically, an engine that uses natural language processing technology (for example, the API of a natural language processing cloud service provider) is used.

[0762] 4. Sentiment Analysis Engine:

[0763] An engine for analyzing and recognizing emotions from user messages. For example, we use a cloud-based emotion analysis engine.

[0764] 5. Generative AI Models:

[0765] An AI model that generates appropriate reply candidates and actions based on the results of user sentiment analysis and past data. For example, it is possible to use the latest model of generative AI.

[0766] 6. Dialogue Systems:

[0767] It is an interface for realizing two-way communication within the vehicle, and includes execution means for executing generated responses and actions.

[0768] Specific operation of the system

[0769] In this system, the user inputs a message into the in-car dialogue system using voice or text. The user's device then sends this message to the server. The server first analyzes the message content using a text analysis engine to extract key elements. It then uses an emotion analysis engine to recognize the user's emotions. Once the emotion analysis is complete, the server's generative AI model generates optimal reply candidates based on the analysis results and past data.

[0770] The generated reply candidates are sent to the user's device and displayed on the in-car display. The user can select an appropriate reply from the displayed candidates. The selected reply or action is then executed by the dialogue system, adjusting the navigation or the in-car environment.

[0771] Examples and prompts

[0772] Specific examples

[0773] If a user is driving and says to the in-car system, "I'm tired today and want to get home early," the system will:

[0774] 1. The user device sends a message to the server.

[0775] 2. The text analysis engine analyzes "I want to go home early" and extracts the main intent.

[0776] 3. The emotion analysis engine detects "fatigue."

[0777] 4. The generative AI model generates the following reply suggestions:

[0778] "Thank you for your hard work. I'm calculating the quickest route home."

[0779] "Play some relaxing music?"

[0780] 5. Candidates are displayed on the in-car display and the user selects one.

[0781] 6. The selected reply is executed and navigation changes to the shortest route.

[0782] Prompt Sentence Examples

[0783] If user fatigue is detected:

[0784] "User fatigue detected. Suggest navigation for the quickest route home."

[0785] "The user is tired, so please suggest three relaxing music pieces."

[0786] Using the above configuration, the present invention can provide a more personalized driving experience that takes into account the user's emotions, and realize comfortable communication and navigation.

[0787] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0788] Step 1:

[0789] The user inputs a message to the in-car dialogue system.

[0790] Input: User's voice or text message

[0791] Output: Message data

[0792] Action: The user tells the dialogue system, "I'm tired today and want to go home early."

[0793] Step 2:

[0794] The terminal sends the received message to the server.

[0795] Input: Message data

[0796] Output: Message data sent to the server

[0797] How it works: The device digitizes the user's message and sends it to the server.

[0798] Step 3:

[0799] The server sends the received message to a text analysis engine to extract key elements.

[0800] Input: Message data

[0801] Output: Key element data (e.g., "I want to go home early")

[0802] How it works: A text analysis engine in the server analyzes the message and extracts the main idea.

[0803] Step 4:

[0804] The server sends the results of the text analysis engine to the sentiment analysis engine to identify the user's sentiment.

[0805] Input: Main element data

[0806] Output: Emotion data (e.g., "fatigue")

[0807] How it works: The sentiment analysis engine analyzes and identifies the user's sentiment based on the text analysis results.

[0808] Step 5:

[0809] The server generates reply suggestions using a generative AI model based on the sentiment analysis results and past user data.

[0810] Input: Emotion data, past user data

[0811] Output: Reply candidate data

[0812] How it works: A generative AI model combines sentiment data with past user data to generate appropriate reply suggestions.

[0813] Step 6:

[0814] The server sends the generated reply candidates to the terminal and presents them to the user.

[0815] Input: Reply candidate data

[0816] Output: Suggested replies displayed on the user's device

[0817] How it works: The server sends reply candidates to the device, which are then displayed on the in-car display.

[0818] Step 7:

[0819] The user selects from the suggested replies.

[0820] Type: Show suggested replies

[0821] Output: Selected reply suggestion

[0822] What happens: The user selects the reply suggestion "Great work! Calculating the quickest route home."

[0823] Step 8:

[0824] The terminal sends the selected reply to the server.

[0825] Enter: Selected reply suggestion

[0826] Output: Selected replies sent to the server

[0827] How it works: The device sends the user's selection to the server.

[0828] Step 9:

[0829] The server performs navigation and environmental adjustments based on the selected reply.

[0830] Input: Selected reply data

[0831] Output: Navigation route, environment settings

[0832] How it works: The server sends commands to the navigation system to calculate and set the shortest route, and also makes appropriate environmental adjustments (e.g., playing music).

[0833] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0834] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0835] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0836] [Third embodiment]

[0837] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0838] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0839] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0840] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0841] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0842] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0843] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0844] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0845] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0846] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0847] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0848] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0849] The present invention relates to a system and method for making users' message replies more efficient and supporting appropriate communication, and detailed embodiments thereof are described below.

[0850] The present invention is a system consisting of a user terminal, a server, a text analysis engine, and an AI model, which supports a user in replying promptly and appropriately to a message received.

[0851] Overview of program processing flow

[0852] Receiving and parsing messages

[0853] 1. A user receives a message using a chat application, for example, from a friend, "Want to go see a movie with me this weekend?"

[0854] 2. The device sends the received message to the server, where it is sent as text data via a communication protocol.

[0855] 3. The server's text analysis engine receives the message and begins analyzing it. The text analysis engine uses natural language processing technology to understand the subject and intent of the message. For example, it extracts keywords such as "weekend," "movie," and "go together" and classifies the intent of the message as a "request to confirm plans."

[0856] Generate reply suggestions

[0857] 4. The server accesses the user's past message history database and evaluates the user's communication style, thereby understanding how the user has responded in the past and how they prefer to express themselves.

[0858] 5. The server's AI model generates appropriate reply suggestions based on the text analysis results and the user's communication style. For example, suggestions like the following might be generated:

[0859] "Thanks, I'd love to go! What time would be good?"

[0860] "Sorry, I have plans this weekend so I can't make it. See you next time."

[0861] I haven't decided on a schedule yet, so I'll think about it for a bit and get back to you.

[0862] Sending and selecting suggested replies

[0863] 6. The server sends the generated reply candidates to the device, where they are displayed on the user interface (UI).

[0864] 7. The user selects the appropriate reply from the suggested replies, for example, "Thanks, I'd love to! What time would be good?"

[0865] 8. The terminal sends the selected reply message to the server.

[0866] 9. The server delivers the selected reply message to the recipient friend, thereby completing the communication promptly and appropriately.

[0867] Specific examples

[0868] For example, when a user receives a message from a friend saying, "Would you like to go see a movie with me this weekend?", the system works as follows:

[0869] When a user receives a message, the device sends it to the server. The server uses a text analysis engine to analyze the message and extract key elements. In this case, the extracted result is "Request to confirm schedule." The server then generates appropriate reply candidates for this element based on the user's past message history. The generated reply candidates are then sent to the device and displayed to the user. For example, a candidate might be "Thanks, I'd love to go! What time would be good?" The user selects an appropriate reply and presses the send button on the device, which then sends the reply to their friend via the server.

[0870] In this way, the present invention supports users in responding to messages quickly and appropriately, thereby improving the efficiency of user communication.

[0871] The processing flow will be explained below.

[0872] Step 1:

[0873] A user has a chat app open and receives a message from a friend saying, "Want to go see a movie with me this weekend?"

[0874] Step 2:

[0875] The terminal displays the received message and simultaneously sends the message as text data to the server.

[0876] Step 3:

[0877] The server sends the received message as a request to the text analysis engine, which uses natural language processing technology to analyze the content of the message.

[0878] Step 4:

[0879] The text analysis engine analyzes the message and extracts key elements (e.g., "weekend," "movie," and "going together"). As a result of the analysis, the intent of the message is classified as a "request to confirm plans."

[0880] Step 5:

[0881] The server accesses the user's past message history database to acquire the user's communication style, thereby learning what expressions the user has used in the past.

[0882] Step 6:

[0883] The server's AI model generates multiple reply suggestions based on the text analysis results and the user's communication style, for example:

[0884] "Thanks, I'd love to go! What time would be good?"

[0885] "Sorry, I have plans this weekend so I can't make it. See you next time."

[0886] I haven't decided on a schedule yet, so I'll think about it for a bit and get back to you.

[0887] Step 7:

[0888] The server generates reply suggestions and sends them to the device, where they are displayed to the user.

[0889] Step 8:

[0890] The user selects the appropriate reply from the suggested replies, for example, "Thanks, I'd love to! What time would be good?"

[0891] Step 9:

[0892] The terminal sends the selected reply message to the server.

[0893] Step 10:

[0894] The server delivers the final reply message to the recipient friend, who then displays the message on their device.

[0895] This series of steps allows users to quickly and appropriately reply after receiving a message.

[0896] Example 1

[0897] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0898] In modern communication, users are expected to respond quickly and appropriately. However, constantly rethinking the content of their replies to messages is a time-consuming task for users. Therefore, there is a need for a way for users to efficiently generate replies to received messages and communicate quickly.

[0899] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0900] In this invention, the server includes a receiving means, a text analyzing means for analyzing the received text data and extracting key elements, a reply generating means for generating reply candidates based on the user's past message history and communication style, a transmitting means for transmitting the generated reply candidates to the user terminal, and a distributing means for receiving a reply message selected from the user terminal and transmitting it to the receiving terminal, thereby enabling the user to reply quickly and appropriately to the message received.

[0901] The "receiving means" is a means for processing a message received by a user through a chat application and sending it to a server.

[0902] The "text analysis means" is a means for analyzing received text data and extracting key elements and keywords.

[0903] The "reply generation means" is a means for generating appropriate reply candidates based on the elements and keywords extracted by the text analysis means, as well as the user's past message history and communication style.

[0904] The "transmitting means" is a means for transmitting the generated reply candidates to the user terminal and displaying them on the user interface of the user terminal.

[0905] The "distribution means" is a means for sending a selected reply message from a user terminal back to the server, and finally sending the reply message to the receiving terminal.

[0906] "Natural language processing technology" is a field of computer science that is a technology for understanding, analyzing, and generating human language.

[0907] An "AI model" is an algorithm or data structure that uses artificial intelligence to perform various tasks, and in natural language processing in particular, it is a model for generating and analyzing text.

[0908] The present invention relates to a system and method for improving the efficiency of user message replies and supporting appropriate communication. The system comprises a user terminal, a server, a text analysis engine, and an AI model.

[0909] Receiving and parsing messages

[0910] First, a user receives a message using a chat application. For example, the received message might say, "Would you like to go see a movie with me this weekend?" The device receives this message and sends it to the server. The HTTP POST request is typically used as the communication protocol.

[0911] The server passes the received message to a text analysis engine. This text analysis engine uses natural language processing technologies, such as Google Cloud Natural Language API or IBM Watson Natural Language Understanding. The text analysis engine analyzes the message and extracts key elements and keywords. Specifically, the keywords "weekend," "movie," and "go together" are extracted, and the intent of the message is classified as a "request to confirm plans."

[0912] Generate reply suggestions

[0913] The server then accesses a database of the user's past messages, which can be a relational or NoSQL database like MySQL or MongoDB, and runs a database query to retrieve the user's past reply patterns and communication style.

[0914] Based on the acquired data, the server uses an AI model (e.g., OpenAI's GPT-3 or BERT) to generate appropriate reply suggestions. This AI model receives text analysis results and the user's communication style as input data. The generated reply suggestions include:

[0915] "Thanks, I'd love to go! What time would be good?"

[0916] "Sorry, I have plans this weekend so I can't make it. See you next time."

[0917] I haven't decided on a schedule yet, so I'll think about it for a bit and get back to you.

[0918] Sending and selecting suggested replies

[0919] The generated reply candidates are sent by the server to the device. The transmission is done through an HTTP response, and the reply candidates are encoded in JSON format. The device decodes them and displays them in the application's user interface. The user can select the appropriate reply from multiple reply candidates. For example, the reply "Thanks, I'd love to go! What time would be good?" is selected.

[0920] The selected reply message is then sent back to the server, which then delivers it to the recipient device using the appropriate protocol (e.g., XMPP or MQTT) for the chat application server, ensuring fast and accurate communication.

[0921] Specific examples

[0922] For example, if a user receives a message from a friend saying, "Want to go see a movie with me this weekend?", the system operates as follows:

[0923] The user receives the message.

[0924] 2. The device sends the message to the server.

[0925] 3. The server passes the message to a text analysis engine, which extracts the key elements: "weekend," "movie," and "going together."

[0926] 4. The server accesses the user's history database to check past response patterns.

[0927] 5. The server's AI model generates appropriate reply suggestions.

[0928] 6. The server sends the generated reply candidates to the terminal, which displays them.

[0929] 7. The user selects the reply "Thanks, I'd love to! What time would be good?"

[0930] 8. The device sends the selected reply to the server, which distributes it to the friend.

[0931] In this way, the present invention supports users in quickly and appropriately replying to messages they receive, thereby improving the efficiency of user communication.

[0932] Example prompts to input to the generative AI model

[0933] Generate a suitable reply to a message from a friend saying "Want to go see a movie with me this weekend?". The output should be in the following format:

[0934] 1. "Reply candidate #1"

[0935] 2. "Reply candidate #2"

[0936] 3. "Reply candidate #3"

[0937] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0938] Step 1:

[0939] A user receives a message using a chat application. For example, a friend might send a message saying, "Want to go see a movie with me this weekend?", and the message is received by the user's device. The input is the message from the friend, and the output is the received message on the device.

[0940] Step 2:

[0941] The terminal sends the message received to the server. Specifically, it uses an HTTP POST request to send the message content to the server. The input is the received message, and the output is the text data sent to the server.

[0942] Step 3:

[0943] The server passes the received message to the text analysis engine and begins analysis. The text analysis engine analyzes the content of the message and extracts key elements and keywords. For example, the keywords "weekend," "movie," and "go together" are extracted using the Google Cloud Natural Language API. The input is the text data sent to the server, and the output is the analysis results (the extraction of key elements).

[0944] Step 4:

[0945] The server accesses a database of the user's past message history. It executes a database query to obtain past reply patterns and the user's communication style. The database can be, for example, MySQL or MongoDB. The input is the analysis result and the user's ID, and the output is the past message history data.

[0946] Step 5:

[0947] The server uses an AI model to generate appropriate reply suggestions. Based on the analysis results and past message history data, the AI ​​model (e.g., OpenAI's GPT-3) generates reply suggestions. The generated reply suggestions include:

[0948] "Thanks, I'd love to go! What time would be good?"

[0949] "Sorry, I have plans this weekend so I can't make it. See you next time."

[0950] I haven't decided on a schedule yet, so I'll think about it for a bit and get back to you.

[0951] The input is the analysis results and the user's message history data, and the output is a list of reply candidates.

[0952] Step 6:

[0953] The server generates reply candidates and sends them to the terminal. The HTTP response is used to send reply candidates encoded in JSON format to the terminal. The input is the reply candidate list, and the output is the data sent to the terminal.

[0954] Step 7:

[0955] The device decodes the reply candidates received and displays them on the user interface. The user selects the appropriate reply from the displayed multiple reply candidates. The input is the reply candidate data sent from the server, and the output is the candidates displayed on the UI. For example, the reply "Thanks, I'd love to go! What time would be good?" is displayed.

[0956] Step 8:

[0957] The user confirms the selected reply on the terminal and sends the reply message to the server. Again, an HTTP POST request is used to send the selected reply message to the server. The input is the reply message selected by the user, and the output is the data sent to the server.

[0958] Step 9:

[0959] The server receives the user's reply and delivers it to the recipient friend. It sends the reply message to the friend's device using an appropriate communication protocol (e.g., XMPP or MQTT). The input is the reply message selected by the user, and the output is the message delivered to the recipient's device. This completes fast and appropriate communication.

[0960] (Application example 1)

[0961] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0962] In recent years, security systems have been required to process warning messages quickly and accurately and to take appropriate action. However, the volume and variety of warning messages received by users has made it difficult to respond quickly. Therefore, it is necessary to develop a system that can instantly understand the content of warning messages and generate appropriate responses.

[0963] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0964] In this invention, the server includes a receiving means, a text analyzing means, a reply generating means, a sending means, and a security analyzing means, which enable the server to quickly and accurately analyze the received warning message, generate appropriate reply candidates, and present them to the user.

[0965] "Receiving means" refers to devices or software for receiving data or messages from the outside.

[0966] "Text analysis means" refers to a device or software for analyzing received text data and extracting key elements.

[0967] A "reply generator" is a device or software that generates appropriate reply candidates based on a user's past message history and communication style.

[0968] "Transmission means" refers to a device or software for transmitting the generated reply candidates to the user terminal.

[0969] "Security analysis means" refers to devices or software that analyze received warning messages and provide appropriate response candidates.

[0970] The present invention relates to a system that improves the efficiency of users' message replies and supports appropriate communication. The system includes a receiving unit, a text analyzing unit, a reply generating unit, a sending unit, and a security analyzing unit.

[0971] The user uses the security assistant application to receive warning messages sent from the security system or camera. This reception can be performed using an application installed on a device such as a smartphone. The received messages are then sent to the server.

[0972] The server analyzes the received text data using natural language processing technology based on Hugging Face's Transformers library. The text analysis method extracts key elements of the received warning message and uses this to understand the message's content and intent.

[0973] Based on the analysis results, the reply generator evaluates the user's past message history and communication style to generate appropriate reply candidates, using the user's response history database as well.

[0974] The generated reply candidates are sent to the user terminal by a sending means, and the user selects an appropriate reply from the presented reply candidates through the application interface.

[0975] For example, if a warning message is received stating, "The window sensor has detected an abnormality," the server analyzes the message and generates appropriate response candidates, such as, "We will call the police," or "Please check the security camera footage." These response candidates are displayed on the user's device, and the user selects an appropriate response, which is then sent via the server to the security system.

[0976] An example prompt is:

[0977] Alert message: "Window sensor has detected an abnormality"

[0978] Potential responses generated: "Call the police." "Check your security camera footage." "I need your help right away."

[0979] In this way, the present invention can quickly and accurately analyze received warning messages, generate appropriate response candidates, and present them to the user, thereby streamlining the user's security response and providing a safer environment.

[0980] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0981] Step 1:

[0982] The user uses the security assistant application to receive an alert message. This input is an alert message sent from a security system or camera in a home or office. The device receives this and sends it to the next step.

[0983] Step 2:

[0984] The terminal sends the received message to the server. In this process, the warning message is transferred to the server as text data. The input data is the warning message sent via the terminal, and the output data is the text data received by the server.

[0985] Step 3:

[0986] The server uses the Hugging Face Transformers library to analyze the received text data. The input data is the received text data, and the analyzed results include the main elements of the message (e.g., "sensor abnormality" and "window").

[0987] Step 4:

[0988] The server uses security analysis means to understand the contents of the warning message based on the analysis results and determines an appropriate response. This process determines what response is appropriate based on the analysis results (e.g., "sensor abnormality"). The input data is the analysis results, and the output data is information about the intended response (e.g., "warning response").

[0989] Step 5:

[0990] The reply generator evaluates the user's past message history and communication style to generate appropriate reply candidates. The input data are the analysis results and user history data, and the output data are the generated reply candidates (e.g., "I will call the police." "Please check the security camera footage.").

[0991] Step 6:

[0992] The server transmits the generated reply candidates to the user terminal, where the input data are the generated reply candidates and the output data are the multiple reply candidates displayed on the user terminal.

[0993] Step 7:

[0994] The user selects an appropriate reply from the presented reply candidates. The input data are the reply candidates presented to the user, and the output data is the reply selected by the user.

[0995] Step 8:

[0996] The terminal sends the selected reply message to the server, which delivers the reply to the security system. The input data is the reply selected by the user, and the output data is the final reply sent to the security system.

[0997] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0998] The present invention relates to a system and method for improving the efficiency of users' message replies and supporting appropriate communication, and is characterized in that reply candidates are generated taking into account the user's emotions by combining an emotion engine. Detailed embodiments are described below.

[0999] This system consists of a user terminal, a server, a text analysis engine, an AI model, and an emotion engine. The operation flow of the system will be explained in detail below.

[1000] Overview of program processing flow

[1001] Receiving and parsing messages

[1002] 1. A user receives a message using a chat application, for example, from a friend, "Want to go see a movie with me this weekend?"

[1003] 2. The terminal displays the received message and sends the message as text data to the server.

[1004] 3. The server sends the received message as a request to the text analysis engine and emotion engine. The analysis engine uses natural language processing technology to analyze the content of the message and extract key elements.

[1005] Emotion Analysis

[1006] 4. The emotion engine analyzes the message and recognizes the user's emotions. For example, emotions such as "happiness," "sadness," and "surprise" are identified by the emotion engine. In this case, the message "Would you like to go see a movie with me?" is analyzed as "curiosity" and "anticipation."

[1007] Generate reply suggestions

[1008] 5. The server accesses the user's past message history database to obtain the user's communication style, taking into account the analyzed emotional information.

[1009] 6. The server's AI model generates appropriate reply suggestions based on the text analysis results, the user's communication style, and emotional information. For example, the following suggestions might be generated:

[1010] "Thanks, I'd love to go! Sounds interesting."

[1011] "Sorry, I have plans this weekend so I can't make it. See you next time."

[1012] I haven't decided on a schedule yet, so I'll think about it for a bit and get back to you.

[1013] Sending and selecting suggested replies

[1014] 7. The server sends the generated reply candidates to the device, where they are displayed on the user interface (UI).

[1015] 8. The user selects the appropriate reply from the suggested replies displayed. For example, the user selects the reply "Thanks, I'd love to go! Sounds interesting."

[1016] 9. The terminal sends the selected reply message to the server.

[1017] 10. The server delivers the final reply message to the recipient friend, who then displays the message on their device.

[1018] Specific examples

[1019] For example, if a user receives a message from a friend saying, "Want to go see a movie with me this weekend?", the system works as follows:

[1020] While a user has a chat app open, they receive a message from a friend. The device sends this message to a server, which then asks the text analysis engine and emotion engine to analyze the message. The text analysis engine analyzes the message and classifies it as a "request to confirm plans," and the emotion engine extracts the emotions of "curiosity" and "anticipation." The server generates reply candidates based on the user's past message history and the results of these analyses. For example, a candidate might be "Thanks, I'd love to go! Sounds interesting." The server sends this to the device so the user can select one. After the user selects and sends an appropriate reply, the server distributes the reply to the friend.

[1021] In this way, the present invention provides a system that takes into account the user's emotions to create more personalized replies, making users' communications more efficient.

[1022] The processing flow will be explained below.

[1023] Step 1:

[1024] A user is using a chat app and receives a message from a friend saying, "Want to go see a movie with me this weekend?"

[1025] Step 2:

[1026] The terminal sends this message as text data to the server.

[1027] Step 3:

[1028] The server sends the received message as a request to the text analysis engine, which uses natural language processing technology to analyze the content of the message and extract key elements.

[1029] Step 4:

[1030] A text analysis engine analyzes the message, extracts key elements (e.g., "weekend," "movie," "going together"), and classifies the intent of the message as "a request to confirm plans."

[1031] Step 5:

[1032] The server simultaneously sends the message to the emotion engine, which uses natural language processing technology to analyze the emotions in the message.

[1033] Step 6:

[1034] The emotion engine analyzes messages and identifies emotions. For example, for the message "Would you like to go see a movie with me?", it extracts the emotions "curiosity" and "anticipation."

[1035] Step 7:

[1036] The server accesses a database of the user's past messages to obtain the user's communication style, for example, whether the user previously preferred polite replies.

[1037] Step 8:

[1038] The server's AI model generates multiple reply suggestions based on the text analysis results, the user's communication style, and emotional information, such as:

[1039] "Thanks, I'd love to go! Sounds interesting."

[1040] "Sorry, I have plans this weekend so I can't make it. Let's go next time."

[1041] I haven't decided on a schedule yet, so I'll think about it for a bit and get back to you.

[1042] Step 9:

[1043] The server sends these reply suggestions to the device, which then displays them on the user interface (UI).

[1044] Step 10:

[1045] The user selects the appropriate reply from the suggested replies, for example, "Thanks, I'd love to go! Sounds interesting."

[1046] Step 11:

[1047] The terminal sends the selected reply message to the server.

[1048] Step 12:

[1049] The server delivers the final reply message to the friend's device, which displays "Thanks, I'd love to go! Sounds interesting."

[1050] This series of steps allows users to quickly respond appropriately after receiving a message. By taking into account the user's emotions, the system provides more personalized replies and achieves effective communication.

[1051] Example 2

[1052] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1053] Conventional message reply systems have difficulty generating personalized and appropriate replies because they do not take into account the user's emotions or past communication styles. As a result, improvements to communication efficiency and user experience have been limited.

[1054] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1055] In this invention, the server includes a receiving means, a text analysis means for analyzing the received text data and extracting key elements, a sentiment analysis means for recognizing the user's sentiment based on the analyzed data, a reply generation means for generating appropriate reply candidates based on the user's past message history and communication style, and a transmission means for transmitting the generated reply candidates to the user terminal. This makes it possible to generate and provide personalized and appropriate reply candidates that take into account the user's sentiment and past communication style.

[1056] "Receiving means" means a piece of hardware or software designed to receive messages or data from a user.

[1057] "Text analysis means" means a piece of hardware or software that has the function of analyzing received text data and extracting key elements or information.

[1058] "Emotion analysis means" means a piece of hardware or software that employs technology to recognize and identify a user's emotions based on analyzed data.

[1059] A "reply generator" is a piece of hardware or software that takes into account a user's past message history and communication style to generate appropriate reply candidates.

[1060] The "transmitting means" is a piece of hardware or software for transferring the generated reply candidates to the user terminal.

[1061] "Natural language processing technology" is a technology that enables computers to understand, interpret, and generate human language.

[1062] A "user terminal" is a device that is directly operated by a user, and includes computers, smartphones, tablets, etc.

[1063] "Communication style" refers to characteristics such as tone, vocabulary, and expression used in messages a user has used in the past.

[1064] "Suggested replies" are system-generated suggested messages for a user to send.

[1065] A "database" is a structured collection of data for efficient storage and retrieval of specific information.

[1066] A "generative AI model" is a model that uses artificial intelligence technology to generate appropriate output based on user input.

[1067] The present invention relates to a system and method for improving the efficiency of users' message replies and supporting appropriate communication. The system includes a user terminal, a server, a text analysis engine, an AI model, and a sentiment analysis engine. The operational flow of this system is specifically shown below.

[1068] Hardware and Software

[1069] The system uses the following components:

[1070] User devices: computers, smartphones, tablets, etc.

[1071] Server: Cloud service or on-premise server.

[1072] Text analysis engines: Natural language processing libraries (e.g., SpaCy, NLTK).

[1073] Sentiment Analysis Engine: A sentiment analysis library (e.g., TextBlob, VADER).

[1074] Generative AI models: Models that use artificial intelligence techniques (e.g., TensorFlow, PyTorch).

[1075] Receiving and parsing messages

[1076] A user uses a chat application and receives a message, for example, "Would you like to go see a movie with me this weekend?" The device displays this message and sends it as text data to the server. The server then sends the received message as a request to the text analysis engine and sentiment analysis engine. The text analysis engine analyzes the received message and extracts key elements.

[1077] Emotion Analysis

[1078] The emotion analysis engine analyzes messages to recognize the user's emotions. For example, in the message "Would you like to go see a movie with me this weekend?", it identifies the emotions "curiosity" and "anticipation."

[1079] Generate reply suggestions

[1080] The server accesses the user's past message history database to obtain the user's consistent communication style. Based on the analyzed text data and emotional information, the generative AI model generates reply candidates. For example, the following candidates are generated:

[1081] "Thanks, I'd love to go! Sounds interesting."

[1082] "Sorry, I have plans this weekend so I can't make it. See you next time."

[1083] I haven't decided on a schedule yet, so I'll think about it for a bit and get back to you.

[1084] Sending and selecting suggested replies

[1085] The server generates reply candidates and sends them to the device, which displays them on the user interface (UI). The user selects the most appropriate reply from the displayed candidates. For example, if the user selects "Thanks, I'd love to go! Sounds interesting," the device sends the selected reply message to the server. The server then distributes the final reply message to the friend's device, where the message is displayed.

[1086] Specific examples

[1087] For example, if a user receives a message from a friend saying, "Want to go see a movie with me this weekend?", the system works as follows:

[1088] While a user has a chat app open, the user receives a message from a friend. The device sends this message to a server, which then asks the text analysis engine and sentiment analysis engine to analyze the message. The text analysis engine analyzes the message and classifies it as a "request to confirm plans," and the sentiment analysis engine extracts the emotions of "curiosity" and "anticipation." The server generates reply candidates based on the user's past message history and the results of these analyses. For example, a candidate might be "Thanks, I'd love to go! Sounds interesting." The server sends this to the device so the user can select one. After the user selects and sends an appropriate reply, the server distributes the reply to the friend.

[1089] In this way, the present invention provides a system that takes into account the user's emotions to create more personalized replies, making users' communications more efficient.

[1090] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1091] Processing Steps

[1092] Step 1: Receiving a message

[1093] Input: A user receives a message using a chat application (e.g., "Want to go see a movie with me this weekend?").

[1094] Operation:

[1095] The user opens a chat app.

[1096] A message is sent from a friend and appears on your device.

[1097] Output: The received message displayed.

[1098] Step 2: Sending the message to the server

[1099] Input: The received message displayed on the terminal.

[1100] Operation:

[1101] The terminal transmits the received message to the server as text data.

[1102] Messages are sent to the server via HTTP POST requests.

[1103] Output: The text data received by the server.

[1104] Step 3: Send a text analysis request

[1105] Input: Text data received by the server.

[1106] Operation:

[1107] The server sends a request to the text analysis engine to analyze the message.

[1108] A text analysis engine (e.g., SpaCy, NLTK) analyzes the message content and extracts key elements.

[1109] Output: Data containing the parsed key elements.

[1110] Step 4: Send a sentiment analysis request

[1111] Input: Text data received by the server.

[1112] Operation:

[1113] The server sends a request to the sentiment analysis engine to analyze the message.

[1114] A sentiment analysis engine (e.g., TextBlob, VADER) analyzes the message and identifies the user's sentiment.

[1115] Output: Parsed emotion information.

[1116] Step 5: Obtaining user's past message data

[1117] Input: The server accesses the database based on the user ID.

[1118] Operation:

[1119] The server executes a database query to retrieve the user's past messaging data and communication style.

[1120] A database (e.g. SQL, NoSQL) returns the user history.

[1121] Output: User's past messaging data and communication style.

[1122] Step 6: Generate reply suggestions

[1123] Input: analyzed key elements, sentiment information, user's previous messages and communication style.

[1124] Operation:

[1125] The server uses this data to input prompt sentences into the generative AI model.

[1126] A generative AI model (e.g., TensorFlow, PyTorch) generates potential replies (example prompt: "Generate a reply when a friend invites you to the movies.").

[1127] Output: Multiple reply suggestions.

[1128] Step 7: Send reply suggestions to your device

[1129] Input: The generated multiple reply suggestions.

[1130] Operation:

[1131] The server transmits the generated reply candidates to the terminal.

[1132] Messages are sent to the device via HTTP POST requests.

[1133] Output: Suggested replies displayed on the device's user interface (UI).

[1134] Step 8: Select a reply suggestion

[1135] Input: Multiple reply suggestions displayed in the device's user interface.

[1136] Operation:

[1137] The user selects the appropriate reply suggestion.

[1138] The terminal sends the selected reply message to the server.

[1139] Output: The selected reply message.

[1140] Step 9: Sending a final reply

[1141] Input: The selected reply message received by the server.

[1142] Operation:

[1143] The server delivers the received message to the friend's terminal.

[1144] Messages are transferred using a real-time communication framework (e.g., WebSocket, Socket.IO).

[1145] Output: The final reply message displayed on your friend's device.

[1146] The above is a concrete explanation of the processing of the system program, broken down into steps, and shows the flow including input and output at each processing step.

[1147] (Application example 2)

[1148] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1149] Modern self-driving vehicles lack dialogue systems that take user emotions into account, resulting in a lack of a satisfactory user experience. Furthermore, navigation and environmental adjustments do not provide optimal support based on the user's mood and state, resulting in low user satisfaction. To address these issues, a system incorporating user emotion analysis is needed.

[1150] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a receiving means, a text analysis means that analyzes received text data and extracts key elements, a reply generation means that generates reply candidates based on the user's past message history and communication style, a transmission means that transmits the generated reply candidates to the user terminal, and in an in-car dialogue system, a sentiment analysis means that analyzes the sentiment of the user's message, a model generation means that uses a generative AI model to generate optimal reply candidates based on the analysis results, and an execution means that performs navigation and environmental adjustment using prompt sentences. This makes it possible to analyze the user's sentiment and adjust the navigation and environment optimally.

[1151] "Receiving means" refers to a device or system that receives messages or data from a user.

[1152] "Text analysis means" refers to technology or devices that analyze received text data and extract key elements and intent.

[1153] A "reply generator" is a system or algorithm that generates appropriate reply suggestions based on a user's past messaging history and communication style.

[1154] "Transmission means" refers to a technique or device for transmitting the generated reply candidates to a user terminal.

[1155] An "interactive system" is a system or device that enables two-way communication between a user and the system.

[1156] "Emotion analysis means" refers to technology or devices that analyze and recognize emotions from user messages.

[1157] A "generative AI model" is a model or algorithm that uses artificial intelligence to automatically generate appropriate reply suggestions or actions.

[1158] A "prompt" is a sentence or phrase that can be input into a generative AI model to elicit a specific outcome or response.

[1159] "Navigation" refers to systems and functions that present and guide vehicles and users on the optimal route to their destination.

[1160] "Environmental control" refers to technologies and functions that adjust in-car conditions (temperature, music, lighting, etc.) to improve user comfort.

[1161] The present invention relates to a system and method for analyzing user sentiment in dialogue systems within autonomous vehicles, and providing appropriate responses and navigation support. The system aims to provide a more personalized experience by combining user sentiment analysis with generative AI models.

[1162] System configuration

[1163] 1. User Device:

[1164] A device that allows users to input information, such as a smartphone or an in-car display.

[1165] 2. Server:

[1166] It is a high-performance server that uses an instance of a cloud service provider (e.g., a virtual machine of a cloud service). The server is configured to include a text analysis engine, a sentiment analysis engine, and a generative AI model.

[1167] 3. Text Analysis Engine:

[1168] The received message is analyzed and its main elements and intent are extracted. Specifically, an engine that uses natural language processing technology (for example, the API of a natural language processing cloud service provider) is used.

[1169] 4. Sentiment Analysis Engine:

[1170] An engine for analyzing and recognizing emotions from user messages. For example, we use a cloud-based emotion analysis engine.

[1171] 5. Generative AI Models:

[1172] An AI model that generates appropriate reply candidates and actions based on the results of user sentiment analysis and past data. For example, it is possible to use the latest model of generative AI.

[1173] 6. Dialogue Systems:

[1174] It is an interface for realizing two-way communication within the vehicle, and includes execution means for executing generated responses and actions.

[1175] Specific operation of the system

[1176] In this system, the user inputs a message into the in-car dialogue system using voice or text. The user's device then sends this message to the server. The server first analyzes the message content using a text analysis engine to extract key elements. It then uses an emotion analysis engine to recognize the user's emotions. Once the emotion analysis is complete, the server's generative AI model generates optimal reply candidates based on the analysis results and past data.

[1177] The generated reply candidates are sent to the user's device and displayed on the in-car display. The user can select an appropriate reply from the displayed candidates. The selected reply or action is then executed by the dialogue system, adjusting the navigation or the in-car environment.

[1178] Examples and prompts

[1179] Specific examples

[1180] If a user is driving and says to the in-car system, "I'm tired today and want to get home early," the system will:

[1181] 1. The user device sends a message to the server.

[1182] 2. The text analysis engine analyzes "I want to go home early" and extracts the main intent.

[1183] 3. The emotion analysis engine detects "fatigue."

[1184] 4. The generative AI model generates the following reply suggestions:

[1185] "Thank you for your hard work. I'm calculating the quickest route home."

[1186] "Play some relaxing music?"

[1187] 5. Candidates are displayed on the in-car display and the user selects one.

[1188] 6. The selected reply is executed and navigation changes to the shortest route.

[1189] Prompt Sentence Examples

[1190] If user fatigue is detected:

[1191] "User fatigue detected. Suggest navigation for the quickest route home."

[1192] "The user is tired, so please suggest three relaxing music pieces."

[1193] Using the above configuration, the present invention can provide a more personalized driving experience that takes into account the user's emotions, and realize comfortable communication and navigation.

[1194] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1195] Step 1:

[1196] The user inputs a message to the in-car dialogue system.

[1197] Input: User's voice or text message

[1198] Output: Message data

[1199] Action: The user tells the dialogue system, "I'm tired today and want to go home early."

[1200] Step 2:

[1201] The terminal sends the received message to the server.

[1202] Input: Message data

[1203] Output: Message data sent to the server

[1204] How it works: The device digitizes the user's message and sends it to the server.

[1205] Step 3:

[1206] The server sends the received message to a text analysis engine to extract key elements.

[1207] Input: Message data

[1208] Output: Key element data (e.g., "I want to go home early")

[1209] How it works: A text analysis engine in the server analyzes the message and extracts the main idea.

[1210] Step 4:

[1211] The server sends the results of the text analysis engine to the sentiment analysis engine to identify the user's sentiment.

[1212] Input: Main element data

[1213] Output: Emotion data (e.g., "fatigue")

[1214] How it works: The sentiment analysis engine analyzes and identifies the user's sentiment based on the text analysis results.

[1215] Step 5:

[1216] The server generates reply suggestions using a generative AI model based on the sentiment analysis results and past user data.

[1217] Input: Emotion data, past user data

[1218] Output: Reply candidate data

[1219] How it works: A generative AI model combines sentiment data with past user data to generate appropriate reply suggestions.

[1220] Step 6:

[1221] The server sends the generated reply candidates to the terminal and presents them to the user.

[1222] Input: Reply candidate data

[1223] Output: Suggested replies displayed on the user's device

[1224] How it works: The server sends reply candidates to the device, which are then displayed on the in-car display.

[1225] Step 7:

[1226] The user selects from the suggested replies.

[1227] Type: Show suggested replies

[1228] Output: Selected reply suggestion

[1229] What happens: The user selects the reply suggestion "Great work! Calculating the quickest route home."

[1230] Step 8:

[1231] The terminal sends the selected reply to the server.

[1232] Enter: Selected reply suggestion

[1233] Output: Selected replies sent to the server

[1234] How it works: The device sends the user's selection to the server.

[1235] Step 9:

[1236] The server performs navigation and environmental adjustments based on the selected reply.

[1237] Input: Selected reply data

[1238] Output: Navigation route, environment settings

[1239] How it works: The server sends commands to the navigation system to calculate and set the shortest route, and also makes appropriate environmental adjustments (e.g., playing music).

[1240] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1241] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1242] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1243] [Fourth embodiment]

[1244] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1245] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1246] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1247] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1248] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1249] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1250] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1251] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1252] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1253] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1254] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1255] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1256] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1257] The present invention relates to a system and method for making users' message replies more efficient and supporting appropriate communication, and detailed embodiments thereof are described below.

[1258] The present invention is a system consisting of a user terminal, a server, a text analysis engine, and an AI model, which supports a user in replying promptly and appropriately to a message received.

[1259] Overview of program processing flow

[1260] Receiving and parsing messages

[1261] 1. A user receives a message using a chat application, for example, from a friend, "Want to go see a movie with me this weekend?"

[1262] 2. The device sends the received message to the server, where it is sent as text data via a communication protocol.

[1263] 3. The server's text analysis engine receives the message and begins analyzing it. The text analysis engine uses natural language processing technology to understand the subject and intent of the message. For example, it extracts keywords such as "weekend," "movie," and "go together" and classifies the intent of the message as a "request to confirm plans."

[1264] Generate reply suggestions

[1265] 4. The server accesses the user's past message history database and evaluates the user's communication style, thereby understanding how the user has responded in the past and how they prefer to express themselves.

[1266] 5. The server's AI model generates appropriate reply suggestions based on the text analysis results and the user's communication style. For example, suggestions like the following might be generated:

[1267] "Thanks, I'd love to go! What time would be good?"

[1268] "Sorry, I have plans this weekend so I can't make it. See you next time."

[1269] I haven't decided on a schedule yet, so I'll think about it for a bit and get back to you.

[1270] Sending and selecting suggested replies

[1271] 6. The server sends the generated reply candidates to the device, where they are displayed on the user interface (UI).

[1272] 7. The user selects the appropriate reply from the suggested replies, for example, "Thanks, I'd love to! What time would be good?"

[1273] 8. The terminal sends the selected reply message to the server.

[1274] 9. The server delivers the selected reply message to the recipient friend, thereby completing the communication promptly and appropriately.

[1275] Specific examples

[1276] For example, when a user receives a message from a friend saying, "Would you like to go see a movie with me this weekend?", the system works as follows:

[1277] When a user receives a message, the device sends it to the server. The server uses a text analysis engine to analyze the message and extract key elements. In this case, the extracted result is "Request to confirm schedule." The server then generates appropriate reply candidates for this element based on the user's past message history. The generated reply candidates are then sent to the device and displayed to the user. For example, a candidate might be "Thanks, I'd love to go! What time would be good?" The user selects an appropriate reply and presses the send button on the device, which then sends the reply to their friend via the server.

[1278] In this way, the present invention supports users in responding to messages quickly and appropriately, thereby improving the efficiency of user communication.

[1279] The processing flow will be explained below.

[1280] Step 1:

[1281] A user has a chat app open and receives a message from a friend saying, "Want to go see a movie with me this weekend?"

[1282] Step 2:

[1283] The terminal displays the received message and simultaneously sends the message as text data to the server.

[1284] Step 3:

[1285] The server sends the received message as a request to the text analysis engine, which uses natural language processing technology to analyze the content of the message.

[1286] Step 4:

[1287] The text analysis engine analyzes the message and extracts key elements (e.g., "weekend," "movie," and "going together"). As a result of the analysis, the intent of the message is classified as a "request to confirm plans."

[1288] Step 5:

[1289] The server accesses the user's past message history database to acquire the user's communication style, thereby learning what expressions the user has used in the past.

[1290] Step 6:

[1291] The server's AI model generates multiple reply suggestions based on the text analysis results and the user's communication style, for example:

[1292] "Thanks, I'd love to go! What time would be good?"

[1293] "Sorry, I have plans this weekend so I can't make it. See you next time."

[1294] I haven't decided on a schedule yet, so I'll think about it for a bit and get back to you.

[1295] Step 7:

[1296] The server generates reply suggestions and sends them to the device, where they are displayed to the user.

[1297] Step 8:

[1298] The user selects the appropriate reply from the suggested replies, for example, "Thanks, I'd love to! What time would be good?"

[1299] Step 9:

[1300] The terminal sends the selected reply message to the server.

[1301] Step 10:

[1302] The server delivers the final reply message to the recipient friend, who then displays the message on their device.

[1303] This series of steps allows users to quickly and appropriately reply after receiving a message.

[1304] Example 1

[1305] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1306] In modern communication, users are expected to respond quickly and appropriately. However, constantly rethinking the content of their replies to messages is a time-consuming task for users. Therefore, there is a need for a way for users to efficiently generate replies to received messages and communicate quickly.

[1307] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1308] In this invention, the server includes a receiving means, a text analyzing means for analyzing the received text data and extracting key elements, a reply generating means for generating reply candidates based on the user's past message history and communication style, a transmitting means for transmitting the generated reply candidates to the user terminal, and a distributing means for receiving a reply message selected from the user terminal and transmitting it to the receiving terminal, thereby enabling the user to reply quickly and appropriately to the message received.

[1309] The "receiving means" is a means for processing a message received by a user through a chat application and sending it to a server.

[1310] The "text analysis means" is a means for analyzing received text data and extracting key elements and keywords.

[1311] The "reply generation means" is a means for generating appropriate reply candidates based on the elements and keywords extracted by the text analysis means, as well as the user's past message history and communication style.

[1312] The "transmitting means" is a means for transmitting the generated reply candidates to the user terminal and displaying them on the user interface of the user terminal.

[1313] The "distribution means" is a means for sending a selected reply message from a user terminal back to the server, and finally sending the reply message to the receiving terminal.

[1314] "Natural language processing technology" is a field of computer science that is a technology for understanding, analyzing, and generating human language.

[1315] An "AI model" is an algorithm or data structure that uses artificial intelligence to perform various tasks, and in natural language processing in particular, it is a model for generating and analyzing text.

[1316] The present invention relates to a system and method for improving the efficiency of user message replies and supporting appropriate communication. The system comprises a user terminal, a server, a text analysis engine, and an AI model.

[1317] Receiving and parsing messages

[1318] First, a user receives a message using a chat application. For example, the received message might say, "Would you like to go see a movie with me this weekend?" The device receives this message and sends it to the server. The HTTP POST request is typically used as the communication protocol.

[1319] The server passes the received message to a text analysis engine. This text analysis engine uses natural language processing technologies, such as Google Cloud Natural Language API or IBM Watson Natural Language Understanding. The text analysis engine analyzes the message and extracts key elements and keywords. Specifically, the keywords "weekend," "movie," and "go together" are extracted, and the intent of the message is classified as a "request to confirm plans."

[1320] Generate reply suggestions

[1321] The server then accesses a database of the user's past messages, which can be a relational or NoSQL database like MySQL or MongoDB, and runs a database query to retrieve the user's past reply patterns and communication style.

[1322] Based on the acquired data, the server uses an AI model (e.g., OpenAI's GPT-3 or BERT) to generate appropriate reply suggestions. This AI model receives text analysis results and the user's communication style as input data. The generated reply suggestions include:

[1323] "Thanks, I'd love to go! What time would be good?"

[1324] "Sorry, I have plans this weekend so I can't make it. See you next time."

[1325] I haven't decided on a schedule yet, so I'll think about it for a bit and get back to you.

[1326] Sending and selecting suggested replies

[1327] The generated reply candidates are sent by the server to the device. The transmission is done through an HTTP response, and the reply candidates are encoded in JSON format. The device decodes them and displays them in the application's user interface. The user can select the appropriate reply from multiple reply candidates. For example, the reply "Thanks, I'd love to go! What time would be good?" is selected.

[1328] The selected reply message is then sent back to the server, which then delivers it to the recipient device using the appropriate protocol (e.g., XMPP or MQTT) for the chat application server, ensuring fast and accurate communication.

[1329] Specific examples

[1330] For example, if a user receives a message from a friend saying, "Want to go see a movie with me this weekend?", the system operates as follows:

[1331] The user receives the message.

[1332] 2. The device sends the message to the server.

[1333] 3. The server passes the message to a text analysis engine, which extracts the key elements: "weekend," "movie," and "going together."

[1334] 4. The server accesses the user's history database to check past response patterns.

[1335] 5. The server's AI model generates appropriate reply suggestions.

[1336] 6. The server sends the generated reply candidates to the terminal, which displays them.

[1337] 7. The user selects the reply "Thanks, I'd love to! What time would be good?"

[1338] 8. The device sends the selected reply to the server, which distributes it to the friend.

[1339] In this way, the present invention supports users in quickly and appropriately replying to messages they receive, thereby improving the efficiency of user communication.

[1340] Example prompts to input to the generative AI model

[1341] Generate a suitable reply to a message from a friend saying "Want to go see a movie with me this weekend?". The output should be in the following format:

[1342] 1. "Reply candidate #1"

[1343] 2. "Reply candidate #2"

[1344] 3. "Reply candidate #3"

[1345] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1346] Step 1:

[1347] A user receives a message using a chat application. For example, a friend might send a message saying, "Want to go see a movie with me this weekend?", and the message is received by the user's device. The input is the message from the friend, and the output is the received message on the device.

[1348] Step 2:

[1349] The terminal sends the message received to the server. Specifically, it uses an HTTP POST request to send the message content to the server. The input is the received message, and the output is the text data sent to the server.

[1350] Step 3:

[1351] The server passes the received message to the text analysis engine and begins analysis. The text analysis engine analyzes the content of the message and extracts key elements and keywords. For example, the keywords "weekend," "movie," and "go together" are extracted using the Google Cloud Natural Language API. The input is the text data sent to the server, and the output is the analysis results (the extraction of key elements).

[1352] Step 4:

[1353] The server accesses a database of the user's past message history. It executes a database query to obtain past reply patterns and the user's communication style. The database can be, for example, MySQL or MongoDB. The input is the analysis result and the user's ID, and the output is the past message history data.

[1354] Step 5:

[1355] The server uses an AI model to generate appropriate reply suggestions. Based on the analysis results and past message history data, the AI ​​model (e.g., OpenAI's GPT-3) generates reply suggestions. The generated reply suggestions include:

[1356] "Thanks, I'd love to go! What time would be good?"

[1357] "Sorry, I have plans this weekend so I can't make it. See you next time."

[1358] I haven't decided on a schedule yet, so I'll think about it for a bit and get back to you.

[1359] The input is the analysis results and the user's message history data, and the output is a list of reply candidates.

[1360] Step 6:

[1361] The server generates reply candidates and sends them to the terminal. The HTTP response is used to send reply candidates encoded in JSON format to the terminal. The input is the reply candidate list, and the output is the data sent to the terminal.

[1362] Step 7:

[1363] The device decodes the reply candidates received and displays them on the user interface. The user selects the appropriate reply from the displayed multiple reply candidates. The input is the reply candidate data sent from the server, and the output is the candidates displayed on the UI. For example, the reply "Thanks, I'd love to go! What time would be good?" is displayed.

[1364] Step 8:

[1365] The user confirms the selected reply on the terminal and sends the reply message to the server. Again, an HTTP POST request is used to send the selected reply message to the server. The input is the reply message selected by the user, and the output is the data sent to the server.

[1366] Step 9:

[1367] The server receives the user's reply and delivers it to the recipient friend. It sends the reply message to the friend's device using an appropriate communication protocol (e.g., XMPP or MQTT). The input is the reply message selected by the user, and the output is the message delivered to the recipient's device. This completes fast and appropriate communication.

[1368] (Application example 1)

[1369] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1370] In recent years, security systems have been required to process warning messages quickly and accurately and to take appropriate action. However, the volume and variety of warning messages received by users has made it difficult to respond quickly. Therefore, it is necessary to develop a system that can instantly understand the content of warning messages and generate appropriate responses.

[1371] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1372] In this invention, the server includes a receiving means, a text analyzing means, a reply generating means, a sending means, and a security analyzing means, which enable the server to quickly and accurately analyze the received warning message, generate appropriate reply candidates, and present them to the user.

[1373] "Receiving means" refers to devices or software for receiving data or messages from the outside.

[1374] "Text analysis means" refers to a device or software for analyzing received text data and extracting key elements.

[1375] A "reply generator" is a device or software that generates appropriate reply candidates based on a user's past message history and communication style.

[1376] "Transmission means" refers to a device or software for transmitting the generated reply candidates to the user terminal.

[1377] "Security analysis means" refers to devices or software that analyze received warning messages and provide appropriate response candidates.

[1378] The present invention relates to a system that improves the efficiency of users' message replies and supports appropriate communication. The system includes a receiving unit, a text analyzing unit, a reply generating unit, a sending unit, and a security analyzing unit.

[1379] The user uses the security assistant application to receive warning messages sent from the security system or camera. This reception can be performed using an application installed on a device such as a smartphone. The received messages are then sent to the server.

[1380] The server analyzes the received text data using natural language processing technology based on Hugging Face's Transformers library. The text analysis method extracts key elements of the received warning message and uses this to understand the message's content and intent.

[1381] Based on the analysis results, the reply generator evaluates the user's past message history and communication style to generate appropriate reply candidates, using the user's response history database as well.

[1382] The generated reply candidates are sent to the user terminal by a sending means, and the user selects an appropriate reply from the presented reply candidates through the application interface.

[1383] For example, if a warning message is received stating, "The window sensor has detected an abnormality," the server analyzes the message and generates appropriate response candidates, such as, "We will call the police," or "Please check the security camera footage." These response candidates are displayed on the user's device, and the user selects an appropriate response, which is then sent via the server to the security system.

[1384] An example prompt is:

[1385] Alert message: "Window sensor has detected an abnormality"

[1386] Potential responses generated: "Call the police." "Check your security camera footage." "I need your help right away."

[1387] In this way, the present invention can quickly and accurately analyze received warning messages, generate appropriate response candidates, and present them to the user, thereby streamlining the user's security response and providing a safer environment.

[1388] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1389] Step 1:

[1390] The user uses the security assistant application to receive an alert message. This input is an alert message sent from a security system or camera in a home or office. The device receives this and sends it to the next step.

[1391] Step 2:

[1392] The terminal sends the received message to the server. In this process, the warning message is transferred to the server as text data. The input data is the warning message sent via the terminal, and the output data is the text data received by the server.

[1393] Step 3:

[1394] The server uses the Hugging Face Transformers library to analyze the received text data. The input data is the received text data, and the analyzed results include the main elements of the message (e.g., "sensor abnormality" and "window").

[1395] Step 4:

[1396] The server uses security analysis means to understand the contents of the warning message based on the analysis results and determines an appropriate response. This process determines what response is appropriate based on the analysis results (e.g., "sensor abnormality"). The input data is the analysis results, and the output data is information about the intended response (e.g., "warning response").

[1397] Step 5:

[1398] The reply generator evaluates the user's past message history and communication style to generate appropriate reply candidates. The input data are the analysis results and user history data, and the output data are the generated reply candidates (e.g., "I will call the police." "Please check the security camera footage.").

[1399] Step 6:

[1400] The server transmits the generated reply candidates to the user terminal, where the input data are the generated reply candidates and the output data are the multiple reply candidates displayed on the user terminal.

[1401] Step 7:

[1402] The user selects an appropriate reply from the presented reply candidates. The input data are the reply candidates presented to the user, and the output data is the reply selected by the user.

[1403] Step 8:

[1404] The terminal sends the selected reply message to the server, which delivers the reply to the security system. The input data is the reply selected by the user, and the output data is the final reply sent to the security system.

[1405] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1406] The present invention relates to a system and method for improving the efficiency of users' message replies and supporting appropriate communication, and is characterized in that reply candidates are generated taking into account the user's emotions by combining an emotion engine. Detailed embodiments are described below.

[1407] This system consists of a user terminal, a server, a text analysis engine, an AI model, and an emotion engine. The operation flow of the system will be explained in detail below.

[1408] Overview of program processing flow

[1409] Receiving and parsing messages

[1410] 1. A user receives a message using a chat application, for example, from a friend, "Want to go see a movie with me this weekend?"

[1411] 2. The terminal displays the received message and sends the message as text data to the server.

[1412] 3. The server sends the received message as a request to the text analysis engine and emotion engine. The analysis engine uses natural language processing technology to analyze the content of the message and extract key elements.

[1413] Emotion Analysis

[1414] 4. The emotion engine analyzes the message and recognizes the user's emotions. For example, emotions such as "happiness," "sadness," and "surprise" are identified by the emotion engine. In this case, the message "Would you like to go see a movie with me?" is analyzed as "curiosity" and "anticipation."

[1415] Generate reply suggestions

[1416] 5. The server accesses the user's past message history database to obtain the user's communication style, taking into account the analyzed emotional information.

[1417] 6. The server's AI model generates appropriate reply suggestions based on the text analysis results, the user's communication style, and emotional information. For example, the following suggestions might be generated:

[1418] "Thanks, I'd love to go! Sounds interesting."

[1419] "Sorry, I have plans this weekend so I can't make it. See you next time."

[1420] I haven't decided on a schedule yet, so I'll think about it for a bit and get back to you.

[1421] Sending and selecting suggested replies

[1422] 7. The server sends the generated reply candidates to the device, where they are displayed on the user interface (UI).

[1423] 8. The user selects the appropriate reply from the suggested replies displayed. For example, the user selects the reply "Thanks, I'd love to go! Sounds interesting."

[1424] 9. The terminal sends the selected reply message to the server.

[1425] 10. The server delivers the final reply message to the recipient friend, who then displays the message on their device.

[1426] Specific examples

[1427] For example, if a user receives a message from a friend saying, "Want to go see a movie with me this weekend?", the system works as follows:

[1428] While a user has a chat app open, they receive a message from a friend. The device sends this message to a server, which then asks the text analysis engine and emotion engine to analyze the message. The text analysis engine analyzes the message and classifies it as a "request to confirm plans," and the emotion engine extracts the emotions of "curiosity" and "anticipation." The server generates reply candidates based on the user's past message history and the results of these analyses. For example, a candidate might be "Thanks, I'd love to go! Sounds interesting." The server sends this to the device so the user can select one. After the user selects and sends an appropriate reply, the server distributes the reply to the friend.

[1429] In this way, the present invention provides a system that takes into account the user's emotions to create more personalized replies, making users' communications more efficient.

[1430] The processing flow will be explained below.

[1431] Step 1:

[1432] A user is using a chat app and receives a message from a friend saying, "Want to go see a movie with me this weekend?"

[1433] Step 2:

[1434] The terminal sends this message as text data to the server.

[1435] Step 3:

[1436] The server sends the received message as a request to the text analysis engine, which uses natural language processing technology to analyze the content of the message and extract key elements.

[1437] Step 4:

[1438] A text analysis engine analyzes the message, extracts key elements (e.g., "weekend," "movie," "going together"), and classifies the intent of the message as "a request to confirm plans."

[1439] Step 5:

[1440] The server simultaneously sends the message to the emotion engine, which uses natural language processing technology to analyze the emotions in the message.

[1441] Step 6:

[1442] The emotion engine analyzes messages and identifies emotions. For example, for the message "Would you like to go see a movie with me?", it extracts the emotions "curiosity" and "anticipation."

[1443] Step 7:

[1444] The server accesses a database of the user's past messages to obtain the user's communication style, for example, whether the user previously preferred polite replies.

[1445] Step 8:

[1446] The server's AI model generates multiple reply suggestions based on the text analysis results, the user's communication style, and emotional information, such as:

[1447] "Thanks, I'd love to go! Sounds interesting."

[1448] "Sorry, I have plans this weekend so I can't make it. Let's go next time."

[1449] I haven't decided on a schedule yet, so I'll think about it for a bit and get back to you.

[1450] Step 9:

[1451] The server sends these reply suggestions to the device, which then displays them on the user interface (UI).

[1452] Step 10:

[1453] The user selects the appropriate reply from the suggested replies, for example, "Thanks, I'd love to go! Sounds interesting."

[1454] Step 11:

[1455] The terminal sends the selected reply message to the server.

[1456] Step 12:

[1457] The server delivers the final reply message to the friend's device, which displays "Thanks, I'd love to go! Sounds interesting."

[1458] This series of steps allows users to quickly respond appropriately after receiving a message. By taking into account the user's emotions, the system provides more personalized replies and achieves effective communication.

[1459] Example 2

[1460] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1461] Conventional message reply systems have difficulty generating personalized and appropriate replies because they do not take into account the user's emotions or past communication styles. As a result, improvements to communication efficiency and user experience have been limited.

[1462] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1463] In this invention, the server includes a receiving means, a text analysis means for analyzing the received text data and extracting key elements, a sentiment analysis means for recognizing the user's sentiment based on the analyzed data, a reply generation means for generating appropriate reply candidates based on the user's past message history and communication style, and a transmission means for transmitting the generated reply candidates to the user terminal. This makes it possible to generate and provide personalized and appropriate reply candidates that take into account the user's sentiment and past communication style.

[1464] "Receiving means" means a piece of hardware or software designed to receive messages or data from a user.

[1465] "Text analysis means" means a piece of hardware or software that has the function of analyzing received text data and extracting key elements or information.

[1466] "Emotion analysis means" means a piece of hardware or software that employs technology to recognize and identify a user's emotions based on analyzed data.

[1467] A "reply generator" is a piece of hardware or software that takes into account a user's past message history and communication style to generate appropriate reply candidates.

[1468] The "transmitting means" is a piece of hardware or software for transferring the generated reply candidates to the user terminal.

[1469] "Natural language processing technology" is a technology that enables computers to understand, interpret, and generate human language.

[1470] A "user terminal" is a device that is directly operated by a user, and includes computers, smartphones, tablets, etc.

[1471] "Communication style" refers to characteristics such as tone, vocabulary, and expression used in messages a user has used in the past.

[1472] "Suggested replies" are system-generated suggested messages for a user to send.

[1473] A "database" is a structured collection of data for efficient storage and retrieval of specific information.

[1474] A "generative AI model" is a model that uses artificial intelligence technology to generate appropriate output based on user input.

[1475] The present invention relates to a system and method for improving the efficiency of users' message replies and supporting appropriate communication. The system includes a user terminal, a server, a text analysis engine, an AI model, and a sentiment analysis engine. The operational flow of this system is specifically shown below.

[1476] Hardware and Software

[1477] The system uses the following components:

[1478] User devices: computers, smartphones, tablets, etc.

[1479] Server: Cloud service or on-premise server.

[1480] Text analysis engines: Natural language processing libraries (e.g., SpaCy, NLTK).

[1481] Sentiment Analysis Engine: A sentiment analysis library (e.g., TextBlob, VADER).

[1482] Generative AI models: Models that use artificial intelligence techniques (e.g., TensorFlow, PyTorch).

[1483] Receiving and parsing messages

[1484] A user uses a chat application and receives a message, for example, "Would you like to go see a movie with me this weekend?" The device displays this message and sends it as text data to the server. The server then sends the received message as a request to the text analysis engine and sentiment analysis engine. The text analysis engine analyzes the received message and extracts key elements.

[1485] Emotion Analysis

[1486] The emotion analysis engine analyzes messages to recognize the user's emotions. For example, in the message "Would you like to go see a movie with me this weekend?", it identifies the emotions "curiosity" and "anticipation."

[1487] Generate reply suggestions

[1488] The server accesses the user's past message history database to obtain the user's consistent communication style. Based on the analyzed text data and emotional information, the generative AI model generates reply candidates. For example, the following candidates are generated:

[1489] "Thanks, I'd love to go! Sounds interesting."

[1490] "Sorry, I have plans this weekend so I can't make it. See you next time."

[1491] I haven't decided on a schedule yet, so I'll think about it for a bit and get back to you.

[1492] Sending and selecting suggested replies

[1493] The server generates reply candidates and sends them to the device, which displays them on the user interface (UI). The user selects the most appropriate reply from the displayed candidates. For example, if the user selects "Thanks, I'd love to go! Sounds interesting," the device sends the selected reply message to the server. The server then distributes the final reply message to the friend's device, where the message is displayed.

[1494] Specific examples

[1495] For example, if a user receives a message from a friend saying, "Want to go see a movie with me this weekend?", the system works as follows:

[1496] While a user has a chat app open, the user receives a message from a friend. The device sends this message to a server, which then asks the text analysis engine and sentiment analysis engine to analyze the message. The text analysis engine analyzes the message and classifies it as a "request to confirm plans," and the sentiment analysis engine extracts the emotions of "curiosity" and "anticipation." The server generates reply candidates based on the user's past message history and the results of these analyses. For example, a candidate might be "Thanks, I'd love to go! Sounds interesting." The server sends this to the device so the user can select one. After the user selects and sends an appropriate reply, the server distributes the reply to the friend.

[1497] In this way, the present invention provides a system that takes into account the user's emotions to create more personalized replies, making users' communications more efficient.

[1498] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1499] Processing Steps

[1500] Step 1: Receiving a message

[1501] Input: A user receives a message using a chat application (e.g., "Want to go see a movie with me this weekend?").

[1502] Operation:

[1503] The user opens a chat app.

[1504] A message is sent from a friend and appears on your device.

[1505] Output: The received message displayed.

[1506] Step 2: Sending the message to the server

[1507] Input: The received message displayed on the terminal.

[1508] Operation:

[1509] The terminal transmits the received message to the server as text data.

[1510] Messages are sent to the server via HTTP POST requests.

[1511] Output: The text data received by the server.

[1512] Step 3: Send a text analysis request

[1513] Input: Text data received by the server.

[1514] Operation:

[1515] The server sends a request to the text analysis engine to analyze the message.

[1516] A text analysis engine (e.g., SpaCy, NLTK) analyzes the message content and extracts key elements.

[1517] Output: Data containing the parsed key elements.

[1518] Step 4: Send a sentiment analysis request

[1519] Input: Text data received by the server.

[1520] Operation:

[1521] The server sends a request to the sentiment analysis engine to analyze the message.

[1522] A sentiment analysis engine (e.g., TextBlob, VADER) analyzes the message and identifies the user's sentiment.

[1523] Output: Parsed emotion information.

[1524] Step 5: Obtaining user's past message data

[1525] Input: The server accesses the database based on the user ID.

[1526] Operation:

[1527] The server executes a database query to retrieve the user's past messaging data and communication style.

[1528] A database (e.g. SQL, NoSQL) returns the user history.

[1529] Output: User's past messaging data and communication style.

[1530] Step 6: Generate reply suggestions

[1531] Input: analyzed key elements, sentiment information, user's previous messages and communication style.

[1532] Operation:

[1533] The server uses this data to input prompt sentences into the generative AI model.

[1534] A generative AI model (e.g., TensorFlow, PyTorch) generates potential replies (example prompt: "Generate a reply when a friend invites you to the movies.").

[1535] Output: Multiple reply suggestions.

[1536] Step 7: Send reply suggestions to your device

[1537] Input: The generated multiple reply suggestions.

[1538] Operation:

[1539] The server transmits the generated reply candidates to the terminal.

[1540] Messages are sent to the device via HTTP POST requests.

[1541] Output: Suggested replies displayed on the device's user interface (UI).

[1542] Step 8: Select a reply suggestion

[1543] Input: Multiple reply suggestions displayed in the device's user interface.

[1544] Operation:

[1545] The user selects the appropriate reply suggestion.

[1546] The terminal sends the selected reply message to the server.

[1547] Output: The selected reply message.

[1548] Step 9: Sending a final reply

[1549] Input: The selected reply message received by the server.

[1550] Operation:

[1551] The server delivers the received message to the friend's terminal.

[1552] Messages are transferred using a real-time communication framework (e.g., WebSocket, Socket.IO).

[1553] Output: The final reply message displayed on your friend's device.

[1554] The above is a concrete explanation of the processing of the system program, broken down into steps, and shows the flow including input and output at each processing step.

[1555] (Application example 2)

[1556] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1557] Modern self-driving vehicles lack dialogue systems that take user emotions into account, resulting in a lack of a satisfactory user experience. Furthermore, navigation and environmental adjustments do not provide optimal support based on the user's mood and state, resulting in low user satisfaction. To address these issues, a system incorporating user emotion analysis is needed.

[1558] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a receiving means, a text analysis means that analyzes received text data and extracts key elements, a reply generation means that generates reply candidates based on the user's past message history and communication style, a transmission means that transmits the generated reply candidates to the user terminal, and in an in-car dialogue system, a sentiment analysis means that analyzes the sentiment of the user's message, a model generation means that uses a generative AI model to generate optimal reply candidates based on the analysis results, and an execution means that performs navigation and environmental adjustment using prompt sentences. This makes it possible to analyze the user's sentiment and adjust the navigation and environment optimally.

[1559] "Receiving means" refers to a device or system that receives messages or data from a user.

[1560] "Text analysis means" refers to technology or devices that analyze received text data and extract key elements and intent.

[1561] A "reply generator" is a system or algorithm that generates appropriate reply suggestions based on a user's past messaging history and communication style.

[1562] "Transmission means" refers to a technique or device for transmitting the generated reply candidates to a user terminal.

[1563] An "interactive system" is a system or device that enables two-way communication between a user and the system.

[1564] "Emotion analysis means" refers to technology or devices that analyze and recognize emotions from user messages.

[1565] A "generative AI model" is a model or algorithm that uses artificial intelligence to automatically generate appropriate reply suggestions or actions.

[1566] A "prompt" is a sentence or phrase that can be input into a generative AI model to elicit a specific outcome or response.

[1567] "Navigation" refers to systems and functions that present and guide vehicles and users on the optimal route to their destination.

[1568] "Environmental control" refers to technologies and functions that adjust in-car conditions (temperature, music, lighting, etc.) to improve user comfort.

[1569] The present invention relates to a system and method for analyzing user sentiment in dialogue systems within autonomous vehicles, and providing appropriate responses and navigation support. The system aims to provide a more personalized experience by combining user sentiment analysis with generative AI models.

[1570] System configuration

[1571] 1. User Device:

[1572] A device that allows users to input information, such as a smartphone or an in-car display.

[1573] 2. Server:

[1574] It is a high-performance server that uses an instance of a cloud service provider (e.g., a virtual machine of a cloud service). The server is configured to include a text analysis engine, a sentiment analysis engine, and a generative AI model.

[1575] 3. Text Analysis Engine:

[1576] The received message is analyzed and its main elements and intent are extracted. Specifically, an engine that uses natural language processing technology (for example, the API of a natural language processing cloud service provider) is used.

[1577] 4. Sentiment Analysis Engine:

[1578] An engine for analyzing and recognizing emotions from user messages. For example, we use a cloud-based emotion analysis engine.

[1579] 5. Generative AI Models:

[1580] An AI model that generates appropriate reply candidates and actions based on the results of user sentiment analysis and past data. For example, it is possible to use the latest model of generative AI.

[1581] 6. Dialogue Systems:

[1582] It is an interface for realizing two-way communication within the vehicle, and includes execution means for executing generated responses and actions.

[1583] Specific operation of the system

[1584] In this system, the user inputs a message into the in-car dialogue system using voice or text. The user's device then sends this message to the server. The server first analyzes the message content using a text analysis engine to extract key elements. It then uses an emotion analysis engine to recognize the user's emotions. Once the emotion analysis is complete, the server's generative AI model generates optimal reply candidates based on the analysis results and past data.

[1585] The generated reply candidates are sent to the user's device and displayed on the in-car display. The user can select an appropriate reply from the displayed candidates. The selected reply or action is then executed by the dialogue system, adjusting the navigation or the in-car environment.

[1586] Examples and prompts

[1587] Specific examples

[1588] If a user is driving and says to the in-car system, "I'm tired today and want to get home early," the system will:

[1589] 1. The user device sends a message to the server.

[1590] 2. The text analysis engine analyzes "I want to go home early" and extracts the main intent.

[1591] 3. The emotion analysis engine detects "fatigue."

[1592] 4. The generative AI model generates the following reply suggestions:

[1593] "Thank you for your hard work. I'm calculating the quickest route home."

[1594] "Play some relaxing music?"

[1595] 5. Candidates are displayed on the in-car display and the user selects one.

[1596] 6. The selected reply is executed and navigation changes to the shortest route.

[1597] Prompt Sentence Examples

[1598] If user fatigue is detected:

[1599] "User fatigue detected. Suggest navigation for the quickest route home."

[1600] "The user is tired, so please suggest three relaxing music pieces."

[1601] Using the above configuration, the present invention can provide a more personalized driving experience that takes into account the user's emotions, and realize comfortable communication and navigation.

[1602] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1603] Step 1:

[1604] The user inputs a message to the in-car dialogue system.

[1605] Input: User's voice or text message

[1606] Output: Message data

[1607] Action: The user tells the dialogue system, "I'm tired today and want to go home early."

[1608] Step 2:

[1609] The terminal sends the received message to the server.

[1610] Input: Message data

[1611] Output: Message data sent to the server

[1612] How it works: The device digitizes the user's message and sends it to the server.

[1613] Step 3:

[1614] The server sends the received message to a text analysis engine to extract key elements.

[1615] Input: Message data

[1616] Output: Key element data (e.g., "I want to go home early")

[1617] How it works: A text analysis engine in the server analyzes the message and extracts the main idea.

[1618] Step 4:

[1619] The server sends the results of the text analysis engine to the sentiment analysis engine to identify the user's sentiment.

[1620] Input: Main element data

[1621] Output: Emotion data (e.g., "fatigue")

[1622] How it works: The sentiment analysis engine analyzes and identifies the user's sentiment based on the text analysis results.

[1623] Step 5:

[1624] The server generates reply suggestions using a generative AI model based on the sentiment analysis results and past user data.

[1625] Input: Emotion data, past user data

[1626] Output: Reply candidate data

[1627] How it works: A generative AI model combines sentiment data with past user data to generate appropriate reply suggestions.

[1628] Step 6:

[1629] The server sends the generated reply candidates to the terminal and presents them to the user.

[1630] Input: Reply candidate data

[1631] Output: Suggested replies displayed on the user's device

[1632] How it works: The server sends reply candidates to the device, which are then displayed on the in-car display.

[1633] Step 7:

[1634] The user selects from the suggested replies.

[1635] Type: Show suggested replies

[1636] Output: Selected reply suggestion

[1637] What happens: The user selects the reply suggestion "Great work! Calculating the quickest route home."

[1638] Step 8:

[1639] The terminal sends the selected reply to the server.

[1640] Enter: Selected reply suggestion

[1641] Output: Selected replies sent to the server

[1642] How it works: The device sends the user's selection to the server.

[1643] Step 9:

[1644] The server performs navigation and environmental adjustments based on the selected reply.

[1645] Input: Selected reply data

[1646] Output: Navigation route, environment settings

[1647] How it works: The server sends commands to the navigation system to calculate and set the shortest route, and also makes appropriate environmental adjustments (e.g., playing music).

[1648] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1649] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1650] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1651] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1652] FIG. 9 illustrates an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and behaviors arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1653] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1654] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1655] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1656] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1657] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1658] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1659] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1660] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1661] 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.

[1662] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1663] The hardware resource for executing a specific process can be any of the following types of processors: A processor, for example, is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1664] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1665] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1666] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1667] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1668] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1669] The following is further disclosed regarding the above embodiment.

[1670] (Claim 1)

[1671] receiving means;

[1672] a text analysis means for analyzing the received text data to extract key elements;

[1673] A reply generation means for generating reply candidates based on the user's past message history and communication style;

[1674] a transmitting means for transmitting the generated reply candidates to a user terminal;

[1675] A system including:

[1676] (Claim 2)

[1677] 10. The system of claim 1, wherein the received text data is analyzed using natural language processing techniques.

[1678] (Claim 3)

[1679] 2. The system according to claim 1, wherein when generating reply candidates, a plurality of candidates are generated and the candidates are displayed on the user terminal.

[1680] "Example 1"

[1681] (Claim 1)

[1682] receiving means;

[1683] a text analysis means for analyzing the received text data to extract key elements;

[1684] A reply generation means for generating reply candidates based on the user's past message history and communication style;

[1685] a transmitting means for transmitting the generated reply candidates to a user terminal;

[1686] The system includes a distribution means for receiving a selected reply message from a user terminal and transmitting it to a receiving terminal.

[1687] (Claim 2)

[1688] 10. The system of claim 1, wherein the received text data is analyzed using natural language processing techniques.

[1689] (Claim 3)

[1690] 2. The system according to claim 1, wherein when generating reply candidates, a plurality of candidates are generated and the candidates are displayed on the user terminal.

[1691] "Application Example 1"

[1692] (Claim 1)

[1693] receiving means;

[1694] a text analysis means for analyzing the received text data to extract key elements;

[1695] A reply generation means for generating reply candidates based on the user's past message history and communication style;

[1696] a transmitting means for transmitting the generated reply candidates to a user terminal;

[1697] a security analysis means for analyzing the received warning message and providing appropriate candidate responses;

[1698] A system including:

[1699] (Claim 2)

[1700] 10. The system of claim 1, wherein the received text data is analyzed using natural language processing techniques.

[1701] (Claim 3)

[1702] 2. The system according to claim 1, wherein when generating reply candidates, a plurality of candidates are generated and the candidates are displayed on the user terminal.

[1703] "Example 2: Combining Emotion Engines"

[1704] (Claim 1)

[1705] receiving means;

[1706] a text analysis means for analyzing the received text data to extract key elements;

[1707] an emotion analysis means for recognizing the user's emotion based on the analyzed data;

[1708] A reply generation means for generating appropriate reply candidates based on the user's past message history and communication style;

[1709] a transmitting means for transmitting the generated reply candidates to a user terminal;

[1710] A system including:

[1711] (Claim 2)

[1712] 10. The system of claim 1, wherein the received text data is analyzed using natural language processing techniques.

[1713] (Claim 3)

[1714] The system according to claim 1, wherein when generating reply candidates, the system generates multiple appropriate reply candidates based on the user's emotional information and past message history, and displays the candidates on the user's terminal.

[1715] "Application example 2 when combining emotion engines"

[1716] (Claim 1)

[1717] receiving means;

[1718] a text analysis means for analyzing the received text data to extract key elements;

[1719] A reply generation means for generating reply candidates based on the user's past message history and communication style;

[1720] a transmitting means for transmitting the generated reply candidates to a user terminal;

[1721] In an in-car dialogue system, an emotion analysis means for analyzing emotions in a user message;

[1722] A model generation means that uses a generative AI model to generate optimal reply candidates based on the analysis results;

[1723] an execution means for performing navigation and environment adjustment using prompt sentences;

[1724] A system including:

[1725] (Claim 2)

[1726] 2. The system according to claim 1, further comprising emotion analysis means for analyzing the received text data using natural language processing techniques and extracting emotions.

[1727] (Claim 3)

[1728] The system according to claim 1, wherein when generating reply candidates, a plurality of candidates are generated, the candidates are displayed on the user terminal, and the most suitable candidate is used to perform navigation or environmental adjustment. [Explanation of symbols]

[1729] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. receiving means; a text analysis means for analyzing the received text data to extract key elements; A reply generation means for generating reply candidates based on the user's past message history and communication style; a transmitting means for transmitting the generated reply candidates to a user terminal; A system including:

2. 10. The system of claim 1, wherein the received text data is analyzed using natural language processing techniques.

3. 2. The system according to claim 1, wherein when generating reply candidates, a plurality of candidates are generated and displayed on the user terminal.

Images