system

Abstract

We provide the system. [Solution] A means of receiving input information from the user, A means for calculating the similarity with previously recorded input information, A means for obtaining response information corresponding to past input information whose similarity exceeds a predetermined threshold, A means of providing the acquired response information to the user, A generation means for generating response information when no similar input information exists, A system including means for recording generated response information.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a character of the chatbot, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance that responds to the user utterance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] [[ID=Address]]With the increase in power consumption associated with the use of large language models, it is an important issue from the perspectives of sustainability and efficiency. Due to this problem, there are concerns about the expansion of data centers and the increase in energy consumption, which may increase the environmental and economic burdens. Therefore, there is a need for a system that can efficiently provide high-quality answers to user questions without directly operating a large language model.

Means for Solving the Problems

[0005] This invention provides a system that receives and records user input information, evaluates its similarity to previously collected input information, and then quickly provides a response based on similar past information. If the similarity exceeds a predetermined threshold, the system retrieves already recorded response information and directly sends it back to the user, thereby achieving an efficient response. If no similar input information exists, a new response is generated, and this generated response is recorded and used for subsequent responses. In this way, this invention enables the rapid and high-quality provision of information while suppressing power consumption.

[0006] "Input information" refers to data such as questions and requests sent from the user to the system.

[0007] "Similarity" is an index used to evaluate how similar two pieces of input information or data are.

[0008] A "threshold" is a numerical value used as a criterion for recognizing similar tasks in similarity evaluation.

[0009] "Response information" refers to the answers or information that the system provides in response to the input information.

[0010] "Generation means" refers to a function or process that creates new response information as needed.

[0011] "Means of recording" refers to the function of writing and storing input information and response information in a database or similar system.

[0012] A "system" is a device or program that implements these means in practice and, as a whole, performs information provision services efficiently. [Brief explanation of the drawing]

[0013] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2]It is a conceptual diagram showing an example of the main functions of a data processing device and a smart device according to the first embodiment. [Figure 3] It is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] It is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] It is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] It is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] It is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] It is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] It shows an emotion map to which a plurality of emotions are mapped. [Figure 10] It shows an emotion map to which a plurality of emotions are mapped. [Figure 11] It is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Example 2 when an emotion engine is combined. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when an emotion engine is combined.

Embodiments for Carrying Out the Invention

[0014] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0015] First, the terms used in the following description will be explained.

[0016] In the following embodiments, the numbered processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0017] In the following embodiments, the numbered RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0018] In the following embodiments, the numbered storage is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, etc.

[0019] In the following embodiments, the numbered communication I / F (Interface) is an interface including a communication processor and 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), or Bluetooth (registered trademark), etc.

[0020] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0021] [First Embodiment]

[0022] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0023] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0024] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0025] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0026] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0027] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0028] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0029] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0030] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0031] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0032] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0033] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0034] The present invention is implemented as a software program that runs on multiple computer systems. The central system of the invention aims to effectively process user input information and provide responses in a fast and power-efficient manner.

[0035] First, the server receives input information from the user. This input is data expressed in natural language as the user's questions or requests. The received input information is recorded directly in the database. This database record becomes the foundational data used for future similar task searches and response generation.

[0036] Next, the server compares the received input information with previously recorded data and calculates the similarity. Here, natural language processing techniques are used to evaluate the similarity. If the similarity exceeds a certain threshold, the server retrieves response information linked to past input information from the database and sends it back to the user.

[0037] For example, if a user asks "What is the population of XX city?", and there have been similar questions in the past such as "I want to know the number of residents in XX city," the system will immediately provide the answer from those previous questions. By doing this, the system utilizes existing answers to avoid generating new ones, thereby reducing power consumption.

[0038] If no similar task is found, the server invokes a large language model to generate new response information. The generated response information is recorded in the database and made available for reuse in future queries.

[0039] By implementing this system, redundant processing of the same information can be avoided, leading to efficient resource utilization. Thus, the present invention represents a form that realizes a sustainable and rapid information provision system.

[0040] The following describes the processing flow.

[0041] Step 1:

[0042] Users send input information, such as questions and requests, in natural language through their devices.

[0043] Step 2:

[0044] The server records the received input information in a database. This database stores the entered content, a timestamp, and, if necessary, the user ID.

[0045] Step 3:

[0046] The server uses newly recorded input information to calculate similarity with previously recorded data. This calculation employs natural language processing techniques to vectorize the input information for comparison.

[0047] Step 4:

[0048] The server searches the database for past tasks whose similarity exceeds a predetermined threshold. If a matching past task is found, it retrieves the relevant response information.

[0049] Step 5:

[0050] If a similar task is found, the server immediately provides an answer by sending the previously retrieved response information to the user's terminal.

[0051] Step 6:

[0052] If no similar task is found, the server will invoke a large language model to generate new response information. This process is expected to require significant time and computing resources.

[0053] Step 7:

[0054] The server records the generated response information in a database and stores it as reference data for future queries.

[0055] Step 8:

[0056] Finally, the server sends the newly generated response information to the user's terminal, completing the information provision process.

[0057] (Example 1)

[0058] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0059] In today's information-saturated environment, it is becoming increasingly difficult for users to obtain information quickly and efficiently. Furthermore, redundant processing of the same questions wastes resources, highlighting the growing need for efficient response systems.

[0060] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0061] In this invention, the server includes means for efficiently receiving input information from a user and recording the data, means for calculating the similarity between previously recorded information and the received information using natural language processing technology, and means for obtaining response information related to the past input information when the similarity exceeds a threshold. This makes it possible to provide the user with the necessary information quickly and in a power-efficient manner.

[0062] "Means for efficiently receiving user input and recording data" refers to a function that accurately and quickly receives natural language questions and requests provided by users to the system and systematically stores them in a database.

[0063] "Methods for calculating similarity using natural language processing technology" refers to techniques that analyze input natural language information and compare it with past records to evaluate how similar they are.

[0064] "Means for obtaining response information related to past input information" refers to a function that extracts the most relevant response from previously stored data based on the results of similarity calculations and sends it to the user.

[0065] "A means of activating a generative AI model to generate response information" refers to a function that uses a large-scale language model to create a new response when no similar past data exists.

[0066] "Means for recording generated response information for future use" refers to a function that efficiently stores newly generated response data and reuses it when similar inquiries are made later.

[0067] This invention is realized as an information processing system through the cooperation of a server, terminal, and user. First, the user inputs questions or requests in natural language. This input, specifically as a prompt, may include phrases such as "Please tell me the nearest cafe." This input is then sent from the terminal to the server.

[0068] The server first has the function of recording the received information in a database. This recorded information serves as an important foundation for reuse when similar queries occur later. The server then uses natural language processing techniques to compare the received input information with previously recorded information and calculate the similarity. Commonly used platforms and libraries (e.g., NLTK, spaCy, etc.) are used for natural language processing.

[0069] If the similarity exceeds a predetermined threshold, the server retrieves response information related to past input information from the database and sends it back to the user. This is a power-efficient design that conserves resources by making maximum use of existing information.

[0070] If no similar historical data is found, the server generates a new response using a generative AI model. This AI model could be a general-purpose language model (e.g., BERT, GPT series, etc.). The generated new response is then checked for quality and recorded in the database for future use. This process allows the system to flexibly adapt to new information.

[0071] This system enables rapid and power-efficient information delivery, allowing users to receive accurate and immediate answers. As a result, in this form, the present invention streamlines information management in large-scale systems and supports sustainable operation.

[0072] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0073] Step 1:

[0074] Users input questions and requests in natural language. For example, they might enter the prompt "Please tell me the nearest cafe" into a text box and press the submit button. The entered data is immediately sent to the server via the terminal. This input is natural language text data that includes the user's intent.

[0075] Step 2:

[0076] The server receives input information from the user. The received data is natural language text in string format. The server first records this data in a database. This prepares the foundational data necessary for subsequent similarity calculations and response generation. The output of this process is a new record in the database.

[0077] Step 3:

[0078] The server begins analyzing the received information using natural language processing techniques. First, it tokenizes the text data and extracts the meaning of each word. During this process, it uses a natural language processing library to analyze the text. Next, it calculates the similarity to previously recorded information. The input consists of raw text data and the contents of the past database, and the output is a similarity score.

[0079] Step 4:

[0080] The server retrieves past response information based on the similarity score if it exceeds a predetermined threshold. The input here is the similarity score, and the output is past response information. The retrieved response is then checked to ensure it aligns with the user's intent, and if deemed correct, it is returned to the user.

[0081] Step 5:

[0082] If similar information does not exist in the database, the server activates a generative AI model. Based on the input information, it processes the data to generate a new response. The input is text data containing the user's request, and the output is a new response in natural language text format. The generated response undergoes a quality check process before being provided to the user.

[0083] Step 6:

[0084] The server records the newly generated response information in the database. This allows for faster and more power-efficient responses to subsequent queries. The input in this step is the generated response information, and the output is the updated database information.

[0085] (Application Example 1)

[0086] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0087] In modern society, there is a demand for quick and efficient responses to user questions and requests, and the provision of appropriate information and content. However, conventional technologies require large-scale data processing, consume a lot of power, and make real-time information provision difficult. Furthermore, when using visual output devices, there is a lack of content recommendations that are appropriate to the user's situation. It is necessary to provide a system that solves these problems.

[0088] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0089] In this invention, the server includes means for receiving input information from a user, means for calculating the similarity with previously recorded input information, means for acquiring response information corresponding to past input information whose similarity exceeds a predetermined threshold, means for providing the acquired response information to the user, means for generating response information when no similar input information exists, means for recording the generated response information, means for recommending content to a visual output device based on user information, and means for converting speech into text information using a speech recognition device. This enables rapid and energy-efficient information provision in response to user input, and also enables appropriate content recommendation through a visual output device.

[0090] "Means for receiving user input information" refers to devices or mechanisms for collecting text and audio information provided by users.

[0091] "Means for calculating the similarity with previously recorded input information" refers to algorithms or processes for calculating the similarity between newly received input information and existing information in the database.

[0092] "Means for obtaining response information corresponding to past input information whose similarity exceeds a predetermined threshold" refers to a mechanism that extracts relevant information from a database when the calculated similarity meets a certain standard.

[0093] "Means of providing acquired response information to the user" refers to methods or devices for presenting collected information to the user.

[0094] "Generating means for generating response information when similar input information does not exist" refers to language generation models or systems for constructing and supplying the necessary information in response to new questions.

[0095] "Means for recording generated response information" refers to a technology for saving the generated information in a database for future use.

[0096] "A means of recommending content to a visual output device based on user information" refers to a system that analyzes user input data and presents suitable content through a display or similar device.

[0097] "Means for converting speech into text information using a speech recognition device" refers to technology for analyzing speech signals and converting them into text format.

[0098] To realize this invention, it is necessary to construct a system that receives user input information and performs appropriate information processing. The system is centered around a server and consists of multiple hardware and software components.

[0099] The server receives voice and text input from the user and first converts the voice input into text using a speech recognition device. Specific software options include speech recognition engines such as Google® Speech-to-Text. Then, natural language processing techniques are used to analyze the text data and understand its meaning. At this stage, natural language processing libraries such as spaCy or NLTK are used.

[0100] Next, the server compares the received input information with previously recorded information and calculates the similarity. This similarity calculation is a process that compares the new request with data stored in the existing database and determines the degree of relevance. If the similarity exceeds a predetermined threshold, the corresponding response information is quickly retrieved from the database and the appropriate information is presented to the user's visual output device. This visual output device could be, for example, smart glasses. This allows the user to see appropriate content based on past similar information in real time.

[0101] If no similar input information exists, the server activates a generative AI model to generate new response information. This generated information is recorded in a database and stored in a reusable format for future requests. Effective generative AI models include OpenAI® GPT.

[0102] For example, if a user asks "What movies do you recommend?" via voice, the server converts the voice into text and recommends the most suitable movies based on past history and related information. Based on this prompt, it is possible to input into a generative AI model that says, "When a user asks 'What movies do you recommend?', provide a response based on their browsing history and similar questions, and recommend the most suitable movie title."

[0103] This invention enables improved real-time user experience and enhances energy efficiency in information delivery.

[0104] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0105] Step 1:

[0106] The server receives voice and text input information from the user. In the case of voice input, a speech recognition device is used to convert the voice into text. In this process, the voice data is the input, and the text information is the output. Speech recognition is performed using tools such as Google Speech-to-Text.

[0107] Step 2:

[0108] The server analyzes the converted text information using natural language processing techniques. Specifically, it processes the received text data using libraries such as spaCy and NLTK to understand its intent. In this step, the text information is the input, and the analyzed semantic structure is the output.

[0109] Step 3:

[0110] The server calculates similarity by comparing the analyzed semantic structure with previously recorded data in the database. It retrieves past input information from the database and applies it to a similarity calculation algorithm. The input here consists of the semantic structure and existing data, and the output is a similarity score.

[0111] Step 4:

[0112] If the similarity score exceeds a predetermined threshold, the server retrieves the corresponding response information from the database. In this step, the similarity score is the input, and the retrieved response information is the output. This information is then ready to be presented to the user immediately.

[0113] Step 5:

[0114] If no similar input information exists in the database, the server generates new response information using a generative AI model. Models such as OpenAI GPT are used, and a prompt statement is provided as input. The input in this case is the prompt statement, and the generated response information is the output.

[0115] Step 6:

[0116] The server sends the acquired or generated response information to the user's terminal. If the terminal has a visual output device, it is provided as displayed content. Here, the input is the response information, and the output is the information displayed in the user's field of vision.

[0117] Step 7:

[0118] The generated response information is recorded in a database for later reference. This enables quick responses to similar inputs in the future. The input is the response information, and the output is a new entry stored in the database.

[0119] Through these steps, a system is realized that can respond quickly and efficiently to user requests and provide appropriate content.

[0120] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0121] This invention is a system that incorporates sentiment analysis into the process of generating responses to input information, with the aim of improving the user experience. This system is configured to recognize the user's emotions in real time and provide appropriate responses based on those emotions.

[0122] The system is implemented as a software program running on multiple computers. First, the user sends questions and requests to the system via a terminal using natural language. The input information is then transferred to a server for analysis by an emotion engine.

[0123] The server not only records the received input information in a database, but also analyzes the user's emotions using an emotion engine. This emotion engine uses natural language processing to extract emotional elements from the input information and classify them into emotion categories such as positive, negative, and neutral.

[0124] Once the emotion of the input information is recognized, the server selects or generates an appropriate response based on that result. Specifically, it adjusts the tone and content to match the emotion based on existing response information. For example, if a user inputs "I'm busy and feeling down," the emotion engine recognizes the negative emotion and adjusts the response to provide an encouraging message or helpful information.

[0125] If similar past inputs exist, the server uses those responses to quickly provide a reply. If no similar tasks are found, the server uses a language model to generate a new response and records it in the database to prepare for similar tasks in the future.

[0126] Thus, the present invention enables the provision of responses that take user emotions into consideration, resulting in a more human-like interaction. This aims to provide information efficiently and comfortably.

[0127] The following describes the processing flow.

[0128] Step 1:

[0129] Users send questions and requests to the system in natural language through their terminal. The input information is in standard text format.

[0130] Step 2:

[0131] The server receives the input information submitted by the user and records it in data storage. Here, related information such as the input content, user ID, and timestamp is stored in the database.

[0132] Step 3:

[0133] The server passes the input information to the emotion engine, which analyzes the emotions embedded in the user's input. This analysis uses natural language processing techniques to identify emotions such as positive, negative, and neutral from the input text.

[0134] Step 4:

[0135] After the emotion engine recognizes the user's emotions, the server selects or adjusts its response to take those emotions into account. For example, if negative emotions are detected, the server will select a response that shows encouragement or empathy.

[0136] Step 5:

[0137] The server searches its past database for similar input information and calculates the similarity score. If a similar task exists, it uses the response information from that task.

[0138] Step 6:

[0139] If no similar task can be found, the server uses a language model to generate new response information. This process ensures that the generated response is appropriate to the user's emotional state.

[0140] Step 7:

[0141] The server records the generated response information in data storage and stores it for future inquiries.

[0142] Step 8:

[0143] Finally, the server sends the response information to the user's terminal, completing the information provision to the user.

[0144] (Example 2)

[0145] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0146] Traditional systems often produce formulaic responses to user input that do not take emotions or context into account, making it difficult to improve the user experience. In particular, generating responses that reflect the user's emotions is challenging, which is a factor that lowers user satisfaction.

[0147] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0148] In this invention, the server includes means for analyzing received input information and extracting emotional elements, means for selecting or generating appropriate response information based on emotional categories, and means for acquiring response information corresponding to past input information whose similarity exceeds a predetermined threshold. This enables more humane and personalized responses that respond to the user's emotions.

[0149] A "user" is an entity that provides input information to a system and receives a response.

[0150] "Input information" refers to questions and requests in natural language that users send to the system.

[0151] "Emotional elements" refer to information extracted from the user's emotions and related emotions contained in the input information.

[0152] "Response information" refers to the response content generated or selected by the system based on the user's input information.

[0153] "Natural language processing technology" is a general term for technologies that enable computers to understand, analyze, and generate natural language.

[0154] A "language model" is a model that learns the structure and patterns of natural language and generates new text.

[0155] "Similarity" is an indicator that shows the degree of similarity between the user's current input information and previously recorded input information.

[0156] This invention is a system for improving the user experience, and aims to provide appropriate responses by performing sentiment analysis based on user input information.

[0157] First, the user uses a terminal to input questions or requests in natural language and sends them to the system. The input information is transferred to the server in text format. As a concrete example, consider the case where the user inputs "I'm very tired today."

[0158] Next, the server receives this input information and first records it in a database. Then, it uses an emotion engine to analyze the emotional elements of the input information. The emotion engine uses natural language processing techniques to extract emotions such as positive, negative, and neutral from the input information.

[0159] Once sentiment analysis is complete, the server selects or generates appropriate response information based on the sentiment category. This process utilizes a generative AI model, which has the ability to create responses with appropriate tone and content by comparing them with past data.

[0160] A concrete example of a prompt message is, "My friend canceled our plans, and I'm feeling really down. What can I do to cheer myself up?" Based on this example, the server generates an empathetic response that reflects the user's emotions.

[0161] Finally, the server sends the generated response information to the terminal and displays it to the user. This system allows users to receive information in a way that is sensitive to their emotions, resulting in a more satisfying experience.

[0162] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0163] Step 1:

[0164] The user enters questions and requests in natural language into the terminal and sends them to the system. This input information is transferred to the server in text format. The input might be in the form of, for example, "I'm very tired today." The server receives this input data.

[0165] Step 2:

[0166] The server receives the input information and first records it in the database. By storing the input information in an identifiable format, it prepares for subsequent similarity checks and other processes. The database records not only the input information itself but also metadata such as the input time.

[0167] Step 3:

[0168] The server activates the emotion engine and analyzes the input information. Using natural language processing techniques, it extracts emotional elements from the input text and categorizes them as positive, negative, or neutral. Specifically, it analyzes keywords and context in the text to calculate an emotion score. The output is the extracted emotion category.

[0169] Step 4:

[0170] Based on the analysis results, the server determines response information appropriate to the emotion category. It searches the database for similar past inputs and retrieves responses with high similarity. If no suitable past responses are found, it uses a generative AI model to generate a new response. The model utilizes language building rules to create text that resonates with the user's emotions.

[0171] Step 5:

[0172] The server sends the generated or selected response information to the terminal. The user can receive this response on the terminal. The content of the response is emotionally sensitive to the user's input and may include, for example, "Thank you for your hard work. Please take some time to relax."

[0173] In this way, through a series of processes, it becomes possible to provide users with emotionally considerate responses.

[0174] (Application Example 2)

[0175] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0176] Traditional systems struggled to provide emotionally responsive answers to user inquiries and requests, often resulting in a poor user experience, particularly in online store customer support. Furthermore, the need to quickly provide responses tailored to positive or negative emotions posed a challenge in improving customer satisfaction and maintaining purchasing intent.

[0177] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0178] In this invention, the server includes means for receiving input information from a user, means for analyzing the received input information to recognize emotions, and means for calculating the similarity between the recognized emotions and previously recorded input information. This makes it possible to provide responses in an appropriate tone according to the user's emotions in online store customer support.

[0179] "Means for receiving user input information" refers to a device or program for acquiring information entered by a user via a communication terminal.

[0180] "An emotion analysis means that analyzes received input information to recognize emotions" refers to a program or device that analyzes the emotional state of a user based on the acquired user input information.

[0181] "Means for calculating the similarity with previously recorded input information" refers to algorithms or devices for comparing current input information with past records and calculating the degree of similarity.

[0182] "Means for adjusting and providing response information acquired based on recognized emotions to the user" refers to a device or program that customizes responses with a tone and content appropriate to the analyzed emotions and communicates them to the user.

[0183] "Generating means for generating response information appropriate to an emotion when no similar input information exists" refers to a device or program for automatically generating an appropriate response corresponding to an emotion in response to new input.

[0184] "Means for recording generated response information" refers to a database or program for saving generated responses for later reference or improvement.

[0185] The system that implements this application example first operates by having the user provide input information using a terminal. The information sent from the terminal is transferred to a server via the network. The server utilizes natural language processing technologies such as the Google Cloud Natural Language API to perform sentiment analysis on the received input information.

[0186] The server extracts emotional elements from the input information and classifies them into emotional categories such as positive, negative, and neutral. Next, based on the results of the emotional analysis, it calculates the similarity to similar past input information and performs calculations to obtain the corresponding response information. Machine learning algorithms and database search engines are used to evaluate the similarity.

[0187] If no similar input information exists, the server uses a generative AI model to generate new response information. For example, if a user contacts store support about a product defect, the server can detect negative emotions and provide a tailored response that shows sympathy for the customer and aims to resolve the problem.

[0188] The generated response is delivered to the device in a tone that matches the user's emotions. This allows the user to have a comfortable support experience that is tailored to their feelings. In addition, the generated response information is recorded in a database by the server to improve the accuracy of support in the future.

[0189] For example, if a user sends a message saying, "The product I recently received was broken," the server can analyze this information and generate a response such as, "We apologize for the inconvenience. We will send you a replacement immediately."

[0190] An example of a prompt might be the instruction, "If the customer expresses negative emotions, generate specific solutions and sympathetic responses."

[0191] In this way, the server can generate and provide appropriate responses in real time based on the user's emotions.

[0192] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0193] Step 1:

[0194] The user sends input information via their device. This input information consists of data entered by the user in natural language, such as inquiries or requests, which the device then sends to the server via the network.

[0195] Step 2:

[0196] The server receives input information from the terminal and inputs that data into the sentiment analysis engine. The sentiment analysis engine uses the Google Cloud Natural Language API to extract key emotional elements from the input information and outputs positive, negative, or neutral emotional categories based on that.

[0197] Step 3:

[0198] The server calculates the similarity between the sentiment analysis results and past input information. This process uses a machine learning algorithm to calculate a similarity score between the received input information and past information stored in the database, and retrieves responses with high scores from the database.

[0199] Step 4:

[0200] If similar past information is found on the server, the server adjusts the corresponding response information and formats it to send to the terminal in a tone appropriate to the emotion. This includes the process of adjusting the tone of the response to be positive, negative, or neutral.

[0201] Step 5:

[0202] If no similar information is found on the server, a generative AI model is used to generate new response information. Using sentiment-based prompts, the language model generates a response, and its output is formatted as the default response.

[0203] Step 6:

[0204] The final generated response information is sent from the server to the terminal and presented to the user. Based on the response information received, the user can resolve the problem or take the next action.

[0205] Step 7:

[0206] The generated response and input information are recorded in a database on the server for future queries. This record is used as new data for later analysis and system improvements.

[0207] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0208] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0209] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0210] [Second Embodiment]

[0211] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0212] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0213] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0214] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0215] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0216] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0217] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0218] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0219] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0220] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0221] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0222] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0223] The present invention is implemented as a software program that runs on multiple computer systems. The central system of the invention aims to effectively process user input information and provide responses in a fast and power-efficient manner.

[0224] First, the server receives input information from the user. This input is data expressed in natural language as the user's questions or requests. The received input information is recorded directly in the database. This database record becomes the foundational data used for future similar task searches and response generation.

[0225] Next, the server compares the received input information with previously recorded data and calculates the similarity. Here, natural language processing techniques are used to evaluate the similarity. If the similarity exceeds a certain threshold, the server retrieves response information linked to past input information from the database and sends it back to the user.

[0226] For example, if a user asks "What is the population of XX city?", and there have been similar questions in the past such as "I want to know the number of residents in XX city," the system will immediately provide the answer from those previous questions. By doing this, the system utilizes existing answers to avoid generating new ones, thereby reducing power consumption.

[0227] If no similar task is found, the server invokes a large language model to generate new response information. The generated response information is recorded in the database and made available for reuse in future queries.

[0228] By implementing this system, redundant processing of the same information can be avoided, leading to efficient resource utilization. Thus, the present invention represents a form that realizes a sustainable and rapid information provision system.

[0229] The following describes the processing flow.

[0230] Step 1:

[0231] Users send input information, such as questions and requests, in natural language through their devices.

[0232] Step 2:

[0233] The server records the received input information in a database. This database stores the entered content, a timestamp, and, if necessary, the user ID.

[0234] Step 3:

[0235] The server uses newly recorded input information to calculate similarity with previously recorded data. This calculation employs natural language processing techniques to vectorize the input information for comparison.

[0236] Step 4:

[0237] The server searches the database for past tasks whose similarity exceeds a predetermined threshold. If a matching past task is found, it retrieves the relevant response information.

[0238] Step 5:

[0239] If a similar task is found, the server immediately provides an answer by sending the previously retrieved response information to the user's terminal.

[0240] Step 6:

[0241] If no similar task is found, the server will invoke a large language model to generate new response information. This process is expected to require significant time and computing resources.

[0242] Step 7:

[0243] The server records the generated response information in a database and stores it as reference data for future queries.

[0244] Step 8:

[0245] Finally, the server sends the newly generated response information to the user's terminal, completing the information provision process.

[0246] (Example 1)

[0247] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0248] In today's information-saturated environment, it is becoming increasingly difficult for users to obtain information quickly and efficiently. Furthermore, redundant processing of the same questions wastes resources, highlighting the growing need for efficient response systems.

[0249] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0250] In this invention, the server includes means for efficiently receiving input information from a user and recording the data, means for calculating the similarity between previously recorded information and the received information using natural language processing technology, and means for obtaining response information related to the past input information when the similarity exceeds a threshold. This makes it possible to provide the user with the necessary information quickly and in a power-efficient manner.

[0251] "Means for efficiently receiving user input and recording data" refers to a function that accurately and quickly receives natural language questions and requests provided by users to the system and systematically stores them in a database.

[0252] "Methods for calculating similarity using natural language processing technology" refers to techniques that analyze input natural language information and compare it with past records to evaluate how similar they are.

[0253] "Means for obtaining response information related to past input information" refers to a function that extracts the most relevant response from previously stored data based on the results of similarity calculations and sends it to the user.

[0254] "A means of activating a generative AI model to generate response information" refers to a function that uses a large-scale language model to create a new response when no similar past data exists.

[0255] "Means for recording generated response information for future use" refers to a function that efficiently stores newly generated response data and reuses it when similar inquiries are made later.

[0256] This invention is realized as an information processing system through the cooperation of a server, terminal, and user. First, the user inputs questions or requests in natural language. This input, specifically as a prompt, may include phrases such as "Please tell me the nearest cafe." This input is then sent from the terminal to the server.

[0257] The server first has the function of recording the received information in a database. This recorded information serves as an important foundation for reuse when similar queries occur later. The server then uses natural language processing techniques to compare the received input information with previously recorded information and calculate the similarity. Commonly used platforms and libraries (e.g., NLTK, spaCy, etc.) are used for natural language processing.

[0258] If the similarity exceeds a predetermined threshold, the server retrieves response information related to past input information from the database and sends it back to the user. This is a power-efficient design that conserves resources by making maximum use of existing information.

[0259] If no similar historical data is found, the server generates a new response using a generative AI model. This AI model could be a general-purpose language model (e.g., BERT, GPT series, etc.). The generated new response is then checked for quality and recorded in the database for future use. This process allows the system to flexibly adapt to new information.

[0260] This system enables rapid and power-efficient information delivery, allowing users to receive accurate and immediate answers. As a result, in this form, the present invention streamlines information management in large-scale systems and supports sustainable operation.

[0261] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0262] Step 1:

[0263] Users input questions and requests in natural language. For example, they might enter the prompt "Please tell me the nearest cafe" into a text box and press the submit button. The entered data is immediately sent to the server via the terminal. This input is natural language text data that includes the user's intent.

[0264] Step 2:

[0265] The server receives input information from the user. The received data is natural language text in string format. The server first records this data in a database. This prepares the foundational data necessary for subsequent similarity calculations and response generation. The output of this process is a new record in the database.

[0266] Step 3:

[0267] The server begins analyzing the received information using natural language processing techniques. First, it tokenizes the text data and extracts the meaning of each word. During this process, it uses a natural language processing library to analyze the text. Next, it calculates the similarity to previously recorded information. The input consists of raw text data and the contents of the past database, and the output is a similarity score.

[0268] Step 4:

[0269] The server retrieves past response information based on the similarity score if it exceeds a predetermined threshold. The input here is the similarity score, and the output is past response information. The retrieved response is then checked to ensure it aligns with the user's intent, and if deemed correct, it is returned to the user.

[0270] Step 5:

[0271] If similar information does not exist in the database, the server activates a generative AI model. Based on the input information, it processes the data to generate a new response. The input is text data containing the user's request, and the output is a new response in natural language text format. The generated response undergoes a quality check process before being provided to the user.

[0272] Step 6:

[0273] The server records the newly generated response information in the database. This allows for faster and more power-efficient responses to subsequent queries. The input in this step is the generated response information, and the output is the updated database information.

[0274] (Application Example 1)

[0275] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0276] In modern society, there is a demand for quick and efficient responses to user questions and requests, and the provision of appropriate information and content. However, conventional technologies require large-scale data processing, consume a lot of power, and make real-time information provision difficult. Furthermore, when using visual output devices, there is a lack of content recommendations that are appropriate to the user's situation. It is necessary to provide a system that solves these problems.

[0277] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0278] In this invention, the server includes means for receiving input information from a user, means for calculating the similarity with input information recorded in the past, means for obtaining response information corresponding to past input information whose similarity exceeds a predetermined threshold, means for providing the obtained response information to the user, generating means for generating response information when there is no similar input information, means for recording the generated response information, means for recommending content to a visual output device based on user information, and means for converting voice into character information by a voice recognition device. As a result, it becomes possible to provide information quickly and with high energy efficiency in response to user input, and it also becomes possible to make appropriate content recommendations through the visual output device.

[0279] The "means for receiving input information from a user" is a device or mechanism for collecting text or voice information provided by the user.

[0280] The "means for calculating the similarity with input information recorded in the past" is an algorithm or process for calculating the similarity between newly received input information and existing information in the database.

[0281] The "means for obtaining response information corresponding to past input information whose similarity exceeds a predetermined threshold" is a mechanism for extracting relevant information from the database when the calculated similarity meets a certain criterion.

[0282] The "means for providing the obtained response information to the user" is a method or device for presenting the collected information to the user.

[0283] The "generating means for generating response information when there is no similar input information" is a language generation model or system for constructing and supplying necessary information for a new question.

[0284] The "means for recording the generated response information" is a technology for storing the generated information in the database for future use.

[0285] The means of "recommending content to a visual output device based on user information" is a system that analyzes user input data and presents suitable content through a display or the like.

[0286] The means of "converting speech into character information by a speech recognition device" is a technology that analyzes a speech signal and converts it into text format.

[0287] To implement this invention, it is necessary to construct a system that receives user input information and performs appropriate information processing. The system is centered around a server and is composed of multiple hardware and software components.

[0288] The server receives input information such as speech and text from the user, and first uses a speech recognition device to convert the speech input into character information. As specific software, a speech recognition engine such as Google Speech-to-Text can be used. Then, natural language processing technology is utilized to analyze the text data and understand its meaning. At this stage, natural language processing libraries such as spaCy or NLTK can be used.

[0289] Next, the server compares the received input information with the information recorded in the past and calculates the similarity. This similarity calculation is a process of comparing the data stored in an existing database with a new request and determining the relevance. If the similarity exceeds a predetermined threshold, corresponding response information is quickly retrieved from the database and appropriate information is presented to the user's visual output device. The visual output device may include, for example, smart glasses. As a result, the user can confirm appropriate content based on past similar information in real time.

[0290] If there is no similar input information, the server activates a generative AI model to generate new response information. The generated information is recorded in the database and saved in a reusable form for future requests. As specific generative AI models, OpenAI GPT, etc. are effective.

[0291] For example, if a user asks "What movies do you recommend?" via voice, the server converts the voice into text and recommends the most suitable movies based on past history and related information. Based on this prompt, it is possible to input into a generative AI model that says, "When a user asks 'What movies do you recommend?', provide a response based on their browsing history and similar questions, and recommend the most suitable movie title."

[0292] This invention enables improved real-time user experience and enhances energy efficiency in information delivery.

[0293] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0294] Step 1:

[0295] The server receives voice and text input information from the user. In the case of voice input, a speech recognition device is used to convert the voice into text. In this process, the voice data is the input, and the text information is the output. Speech recognition is performed using tools such as Google Speech-to-Text.

[0296] Step 2:

[0297] The server analyzes the converted text information using natural language processing techniques. Specifically, it processes the received text data using libraries such as spaCy and NLTK to understand its intent. In this step, the text information is the input, and the analyzed semantic structure is the output.

[0298] Step 3:

[0299] The server calculates similarity by comparing the analyzed semantic structure with previously recorded data in the database. It retrieves past input information from the database and applies it to a similarity calculation algorithm. The input here consists of the semantic structure and existing data, and the output is a similarity score.

[0300] Step 4:

[0301] If the similarity score exceeds a predetermined threshold, the server retrieves the corresponding response information from the database. In this step, the similarity score is the input, and the retrieved response information is the output. It is made ready to be presented to the user immediately.

[0302] Step 5:

[0303] If no similar input information exists in the database, the server newly generates response information using a generative AI model. A model such as OpenAI GPT is used, and a prompt sentence is given as the input. At this time, the input is the prompt sentence, and the generated response information is the output.

[0304] Step 6:

[0305] The server transmits the retrieved or generated response information to the user's terminal. If the terminal can use a visual output device, it is provided as the displayed content. Here, the input is the response information, and the output is the information displayed in the user's field of vision.

[0306] Step 7:

[0307] The generated response information is recorded in the database for future reference. This enables a quick response to future similar inputs. The input is the response information, and the output is the new entry stored in the database.

[0308] Through each such step, a system that quickly and efficiently responds to the user's requests and provides appropriate content is realized.

[0309] Furthermore, an emotion engine for estimating the user's emotion may be combined. That is, the specific processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform specific processing using the user's emotion.

[0310] This invention is a system that incorporates sentiment analysis into the process of generating responses to input information, with the aim of improving the user experience. This system is configured to recognize the user's emotions in real time and provide appropriate responses based on those emotions.

[0311] The system is implemented as a software program running on multiple computers. First, the user sends questions and requests to the system via a terminal using natural language. The input information is then transferred to a server for analysis by an emotion engine.

[0312] The server not only records the received input information in a database, but also analyzes the user's emotions using an emotion engine. This emotion engine uses natural language processing to extract emotional elements from the input information and classify them into emotion categories such as positive, negative, and neutral.

[0313] Once the emotion of the input information is recognized, the server selects or generates an appropriate response based on that result. Specifically, it adjusts the tone and content to match the emotion based on existing response information. For example, if a user inputs "I'm busy and feeling down," the emotion engine recognizes the negative emotion and adjusts the response to provide an encouraging message or helpful information.

[0314] If similar past inputs exist, the server uses those responses to quickly provide a reply. If no similar tasks are found, the server uses a language model to generate a new response and records it in the database to prepare for similar tasks in the future.

[0315] Thus, the present invention enables the provision of responses that take user emotions into consideration, resulting in a more human-like interaction. This aims to provide information efficiently and comfortably.

[0316] The following describes the processing flow.

[0317] Step 1:

[0318] Users send questions and requests to the system in natural language through their terminal. The input information is in standard text format.

[0319] Step 2:

[0320] The server receives the input information submitted by the user and records it in data storage. Here, the input content, user ID, timestamp, and other related information are stored in the database.

[0321] Step 3:

[0322] The server passes the input information to the emotion engine, which analyzes the emotions embedded in the user's input. This analysis uses natural language processing techniques to identify emotions such as positive, negative, and neutral from the input text.

[0323] Step 4:

[0324] After the emotion engine recognizes the user's emotions, the server selects or adjusts its response to take those emotions into account. For example, if negative emotions are detected, the server will select a response that shows encouragement or empathy.

[0325] Step 5:

[0326] The server searches its past database for similar input information and calculates the similarity score. If a similar task exists, it uses the response information from that task.

[0327] Step 6:

[0328] If no similar task can be found, the server uses a language model to generate new response information. This process ensures that the generated response is appropriate to the user's emotional state.

[0329] Step 7:

[0330] The server records the generated response information in data storage and stores it for future inquiries.

[0331] Step 8:

[0332] Finally, the server sends the response information to the user's terminal, completing the information provision to the user.

[0333] (Example 2)

[0334] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0335] Traditional systems often produce formulaic responses to user input that do not take emotions or context into account, making it difficult to improve the user experience. In particular, generating responses that reflect the user's emotions is challenging, which is a factor that lowers user satisfaction.

[0336] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0337] In this invention, the server includes means for analyzing received input information and extracting emotional elements, means for selecting or generating appropriate response information based on emotional categories, and means for acquiring response information corresponding to past input information whose similarity exceeds a predetermined threshold. This enables more humane and personalized responses that respond to the user's emotions.

[0338] A "user" is an entity that provides input information to a system and receives a response.

[0339] "Input information" refers to questions and requests in natural language that users send to the system.

[0340] "Emotional elements" refer to information extracted from the user's emotions and related emotions contained in the input information.

[0341] "Response information" refers to the response content generated or selected by the system based on the user's input information.

[0342] "Natural language processing technology" is a general term for technologies that enable computers to understand, analyze, and generate natural language.

[0343] A "language model" is a model that learns the structure and patterns of natural language and generates new text.

[0344] "Similarity" is an indicator that shows the degree of similarity between the user's current input information and previously recorded input information.

[0345] This invention is a system for improving the user experience, and aims to provide appropriate responses by performing sentiment analysis based on user input information.

[0346] First, the user uses a terminal to input questions or requests in natural language and sends them to the system. The input information is transferred to the server in text format. As a concrete example, consider the case where the user inputs "I'm very tired today."

[0347] Next, the server receives this input information and first records it in a database. Then, it uses an emotion engine to analyze the emotional elements of the input information. The emotion engine uses natural language processing techniques to extract emotions such as positive, negative, and neutral from the input information.

[0348] Once sentiment analysis is complete, the server selects or generates appropriate response information based on the sentiment category. This process utilizes a generative AI model, which has the ability to create responses with appropriate tone and content by comparing them with past data.

[0349] A concrete example of a prompt message is, "My friend canceled our plans, and I'm feeling really down. What can I do to cheer myself up?" Based on this example, the server generates an empathetic response that reflects the user's emotions.

[0350] Finally, the server sends the generated response information to the terminal and displays it to the user. This system allows users to receive information in a way that is sensitive to their emotions, resulting in a more satisfying experience.

[0351] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0352] Step 1:

[0353] The user enters questions and requests in natural language into the terminal and sends them to the system. This input information is transferred to the server in text format. The input might be in the form of, for example, "I'm very tired today." The server receives this input data.

[0354] Step 2:

[0355] The server receives the input information and first records it in the database. By storing the input information in an identifiable format, it prepares for subsequent similarity checks and other processes. The database records not only the input information itself but also metadata such as the input time.

[0356] Step 3:

[0357] The server activates the emotion engine and analyzes the input information. Using natural language processing techniques, it extracts emotional elements from the input text and categorizes them as positive, negative, or neutral. Specifically, it analyzes keywords and context in the text to calculate an emotion score. The output is the extracted emotion category.

[0358] Step 4:

[0359] Based on the analysis results, the server determines response information appropriate to the emotion category. It searches the database for similar past inputs and retrieves responses with high similarity. If no suitable past responses are found, it uses a generative AI model to generate a new response. The model utilizes language building rules to create text that resonates with the user's emotions.

[0360] Step 5:

[0361] The server sends the generated or selected response information to the terminal. The user can receive this response on the terminal. The content of the response is emotionally sensitive to the user's input and may include, for example, "Thank you for your hard work. Please take some time to relax."

[0362] In this way, through a series of processes, it becomes possible to provide users with emotionally considerate responses.

[0363] (Application Example 2)

[0364] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0365] Traditional systems struggled to provide emotionally responsive answers to user inquiries and requests, often resulting in a poor user experience, particularly in online store customer support. Furthermore, the need to quickly provide responses tailored to positive or negative emotions posed a challenge in improving customer satisfaction and maintaining purchasing intent.

[0366] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0367] In this invention, the server includes means for receiving input information from a user, means for analyzing the received input information to recognize emotions, and means for calculating the similarity between the recognized emotions and previously recorded input information. This makes it possible to provide responses in an appropriate tone according to the user's emotions in online store customer support.

[0368] "Means for receiving user input information" refers to a device or program for acquiring information entered by a user via a communication terminal.

[0369] "An emotion analysis means that analyzes received input information to recognize emotions" refers to a program or device that analyzes the emotional state of a user based on the acquired user input information.

[0370] "Means for calculating the similarity with previously recorded input information" refers to algorithms or devices for comparing current input information with past records and calculating the similarity.

[0371] "Means for adjusting and providing response information acquired based on recognized emotions to the user" refers to a device or program that customizes responses with a tone and content appropriate to the analyzed emotions and communicates them to the user.

[0372] "Generating means for generating response information appropriate to an emotion when no similar input information exists" refers to a device or program for automatically generating an appropriate response corresponding to an emotion in response to new input.

[0373] "Means for recording generated response information" refers to a database or program for saving generated responses for later reference or improvement.

[0374] The system that implements this application example first operates by having the user provide input information using a terminal. The information sent from the terminal is transferred to a server via the network. The server utilizes natural language processing technologies such as the Google Cloud Natural Language API to perform sentiment analysis on the received input information.

[0375] The server extracts emotional elements from the input information and classifies them into emotional categories such as positive, negative, and neutral. Next, based on the results of the emotional analysis, it calculates the similarity to similar past input information and performs calculations to obtain the corresponding response information. Machine learning algorithms and database search engines are used to evaluate the similarity.

[0376] If no similar input information exists, the server uses a generative AI model to generate new response information. For example, if a user contacts store support about a product defect, the server can detect negative emotions and provide a tailored response that shows sympathy for the customer and aims to resolve the problem.

[0377] The generated response is delivered to the device in a tone that matches the user's emotions. This allows the user to have a comfortable support experience that is tailored to their feelings. In addition, the generated response information is recorded in a database by the server to improve the accuracy of support in the future.

[0378] For example, if a user sends a message saying, "The product I recently received was broken," the server can analyze this information and generate a response such as, "We apologize for the inconvenience. We will send you a replacement immediately."

[0379] An example of a prompt might be the instruction, "If the customer expresses negative emotions, generate specific solutions and sympathetic responses."

[0380] In this way, the server can generate and provide appropriate responses in real time based on the user's emotions.

[0381] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0382] Step 1:

[0383] The user sends input information via their device. This input information consists of data entered by the user in natural language, such as inquiries or requests, which the device then sends to the server via the network.

[0384] Step 2:

[0385] The server receives input information from the terminal and inputs that data into the sentiment analysis engine. The sentiment analysis engine uses the Google Cloud Natural Language API to extract key emotional elements from the input information and outputs positive, negative, or neutral emotional categories based on that.

[0386] Step 3:

[0387] The server calculates the similarity between the sentiment analysis results and past input information. This process uses a machine learning algorithm to calculate a similarity score between the received input information and past information stored in the database, and retrieves responses with high scores from the database.

[0388] Step 4:

[0389] If similar past information is found on the server, the server adjusts the corresponding response information and formats it to send to the terminal in a tone appropriate to the emotion. This includes the process of adjusting the tone of the response to be positive, negative, or neutral.

[0390] Step 5:

[0391] If no similar information is found on the server, a generative AI model is used to generate new response information. Using sentiment-based prompts, the language model generates a response, and its output is formatted as the default response.

[0392] Step 6:

[0393] The final generated response information is sent from the server to the terminal and presented to the user. Based on the response information received, the user can resolve the problem or take the next action.

[0394] Step 7:

[0395] The generated response and input information are recorded in a database on the server for future queries. This record is used as new data for later analysis and system improvements.

[0396] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0397] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0398] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0399] [Third Embodiment]

[0400] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0401] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0402] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0403] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0404] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0405] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0406] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0407] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0408] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0409] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0410] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0411] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0412] The present invention is implemented as a software program that runs on multiple computer systems. The central system of the invention aims to effectively process user input information and provide responses in a fast and power-efficient manner.

[0413] First, the server receives input information from the user. This input is data expressed in natural language as the user's questions or requests. The received input information is recorded directly in the database. This database record becomes the foundational data used for future similar task searches and response generation.

[0414] Next, the server compares the received input information with previously recorded data and calculates the similarity. Here, natural language processing techniques are used to evaluate the similarity. If the similarity exceeds a certain threshold, the server retrieves response information linked to past input information from the database and sends it back to the user.

[0415] For example, if a user asks "What is the population of XX city?", and there have been similar questions in the past such as "I want to know the number of residents in XX city," the system will immediately provide the answer from those previous questions. By doing this, the system utilizes existing answers to avoid generating new ones, thereby reducing power consumption.

[0416] If no similar task is found, the server invokes a large language model to generate new response information. The generated response information is recorded in the database and made available for reuse in future queries.

[0417] By implementing this system, redundant processing of the same information can be avoided, leading to efficient resource utilization. Thus, the present invention represents a form that realizes a sustainable and rapid information provision system.

[0418] The following describes the processing flow.

[0419] Step 1:

[0420] Users send input information, such as questions and requests, in natural language through their devices.

[0421] Step 2:

[0422] The server records the received input information in a database. This database stores the entered content, a timestamp, and, if necessary, the user ID.

[0423] Step 3:

[0424] The server uses newly recorded input information to calculate similarity with previously recorded data. This calculation employs natural language processing techniques to vectorize the input information for comparison.

[0425] Step 4:

[0426] The server searches the database for past tasks whose similarity exceeds a predetermined threshold. If a matching past task is found, it retrieves the relevant response information.

[0427] Step 5:

[0428] If a similar task is found, the server immediately provides an answer by sending the previously retrieved response information to the user's terminal.

[0429] Step 6:

[0430] If no similar task is found, the server will invoke a large language model to generate new response information. This process is expected to require significant time and computing resources.

[0431] Step 7:

[0432] The server records the generated response information in a database and stores it as reference data for future queries.

[0433] Step 8:

[0434] Finally, the server sends the newly generated response information to the user's terminal, completing the information provision process.

[0435] (Example 1)

[0436] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0437] In today's information-saturated environment, it is becoming increasingly difficult for users to obtain information quickly and efficiently. Furthermore, redundant processing of the same questions wastes resources, highlighting the growing need for efficient response systems.

[0438] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0439] In this invention, the server includes means for efficiently receiving input information from a user and recording the data, means for calculating the similarity between previously recorded information and the received information using natural language processing technology, and means for obtaining response information related to the past input information when the similarity exceeds a threshold. This makes it possible to provide the user with the necessary information quickly and in a power-efficient manner.

[0440] "Means for efficiently receiving user input and recording data" refers to a function that accurately and quickly receives natural language questions and requests provided by users to the system and systematically stores them in a database.

[0441] "Methods for calculating similarity using natural language processing technology" refers to techniques that analyze input natural language information and compare it with past records to evaluate how similar they are.

[0442] "Means for obtaining response information related to past input information" refers to a function that extracts the most relevant response from previously stored data based on the results of similarity calculations and sends it to the user.

[0443] "A means of activating a generative AI model to generate response information" refers to a function that uses a large-scale language model to create a new response when no similar past data exists.

[0444] "Means for recording generated response information for future use" refers to a function that efficiently stores newly generated response data and reuses it when similar inquiries are made later.

[0445] This invention is realized as an information processing system through the cooperation of a server, terminal, and user. First, the user inputs questions or requests in natural language. This input, specifically as a prompt, may include phrases such as "Please tell me the nearest cafe." This input is then sent from the terminal to the server.

[0446] The server first has the function of recording the received information in a database. This recorded information serves as an important foundation for reuse when similar queries occur later. The server then uses natural language processing techniques to compare the received input information with previously recorded information and calculate the similarity. Commonly used platforms and libraries (e.g., NLTK, spaCy, etc.) are used for natural language processing.

[0447] If the similarity exceeds a predetermined threshold, the server retrieves response information related to past input information from the database and sends it back to the user. This is a power-efficient design that conserves resources by making maximum use of existing information.

[0448] If no similar historical data is found, the server generates a new response using a generative AI model. This AI model could be a general-purpose language model (e.g., BERT, GPT series, etc.). The generated new response is then checked for quality and recorded in the database for future use. This process allows the system to flexibly adapt to new information.

[0449] This system enables rapid and power-efficient information delivery, allowing users to receive accurate and immediate answers. As a result, in this form, the present invention streamlines information management in large-scale systems and supports sustainable operation.

[0450] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0451] Step 1:

[0452] Users input questions and requests in natural language. For example, they might enter the prompt "Please tell me the nearest cafe" into a text box and press the submit button. The entered data is immediately sent to the server via the terminal. This input is natural language text data that includes the user's intent.

[0453] Step 2:

[0454] The server receives input information from the user. The received data is natural language text in string format. The server first records this data in a database. This prepares the foundational data necessary for subsequent similarity calculations and response generation. The output of this process is a new record in the database.

[0455] Step 3:

[0456] The server begins analyzing the received information using natural language processing techniques. First, it tokenizes the text data and extracts the meaning of each word. During this process, it uses a natural language processing library to analyze the text. Next, it calculates the similarity to previously recorded information. The input consists of raw text data and the contents of the past database, and the output is a similarity score.

[0457] Step 4:

[0458] The server retrieves past response information based on the similarity score if it exceeds a predetermined threshold. The input here is the similarity score, and the output is past response information. The retrieved response is then checked to ensure it aligns with the user's intent, and if deemed correct, it is returned to the user.

[0459] Step 5:

[0460] If similar information does not exist in the database, the server activates a generative AI model. Based on the input information, it processes the data to generate a new response. The input is text data containing the user's request, and the output is a new response in natural language text format. The generated response undergoes a quality check process before being provided to the user.

[0461] Step 6:

[0462] The server records the newly generated response information in the database. This allows for faster and more power-efficient responses to subsequent queries. The input in this step is the generated response information, and the output is the updated database information.

[0463] (Application Example 1)

[0464] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0465] In modern society, there is a demand for quick and efficient responses to user questions and requests, and the provision of appropriate information and content. However, conventional technologies require large-scale data processing, consume a lot of power, and make real-time information provision difficult. Furthermore, when using visual output devices, there is a lack of content recommendations that are appropriate to the user's situation. It is necessary to provide a system that solves these problems.

[0466] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0467] In this invention, the server includes means for receiving input information from a user, means for calculating the similarity with previously recorded input information, means for acquiring response information corresponding to past input information whose similarity exceeds a predetermined threshold, means for providing the acquired response information to the user, means for generating response information when no similar input information exists, means for recording the generated response information, means for recommending content to a visual output device based on user information, and means for converting speech into text information using a speech recognition device. This enables rapid and energy-efficient information provision in response to user input, and also enables appropriate content recommendation through a visual output device.

[0468] "Means for receiving user input information" refers to devices or mechanisms for collecting text and audio information provided by users.

[0469] "Means for calculating the similarity with previously recorded input information" refers to algorithms or processes for calculating the similarity between newly received input information and existing information in the database.

[0470] "Means for obtaining response information corresponding to past input information whose similarity exceeds a predetermined threshold" refers to a mechanism that extracts relevant information from a database when the calculated similarity meets a certain standard.

[0471] "Means of providing acquired response information to the user" refers to methods or devices for presenting collected information to the user.

[0472] "Generating means for generating response information when similar input information does not exist" refers to language generation models or systems for constructing and supplying the necessary information in response to new questions.

[0473] "Means for recording generated response information" refers to a technology for saving the generated information in a database for future use.

[0474] "A means of recommending content to a visual output device based on user information" refers to a system that analyzes user input data and presents suitable content through a display or similar device.

[0475] "Means for converting speech into text information using a speech recognition device" refers to technology for analyzing speech signals and converting them into text format.

[0476] To realize this invention, it is necessary to construct a system that receives user input information and performs appropriate information processing. The system is centered around a server and consists of multiple hardware and software components.

[0477] The server receives voice and text input from the user and first converts the voice input into text using a speech recognition device. Specific software options include speech recognition engines such as Google Speech-to-Text. Then, natural language processing techniques are used to analyze the text data and understand its meaning. At this stage, natural language processing libraries such as spaCy or NLTK are used.

[0478] Next, the server compares the received input information with previously recorded information and calculates the similarity. This similarity calculation is a process that compares the new request with data stored in the existing database and determines the degree of relevance. If the similarity exceeds a predetermined threshold, the corresponding response information is quickly retrieved from the database and the appropriate information is presented to the user's visual output device. This visual output device could be, for example, smart glasses. This allows the user to see appropriate content based on past similar information in real time.

[0479] If no similar input information exists, the server activates a generative AI model to generate new response information. This generated information is recorded in a database and stored in a reusable format for future requests. OpenAI GPT is a suitable example of a generative AI model.

[0480] For example, if a user asks "What movies do you recommend?" via voice, the server converts the voice into text and recommends the most suitable movies based on past history and related information. Based on this prompt, it is possible to input into a generative AI model that says, "When a user asks 'What movies do you recommend?', provide a response based on their browsing history and similar questions, and recommend the most suitable movie title."

[0481] This invention enables improved real-time user experience and enhances energy efficiency in information delivery.

[0482] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0483] Step 1:

[0484] The server receives voice and text input information from the user. In the case of voice input, a speech recognition device is used to convert the voice into text. In this process, the voice data is the input, and the text information is the output. Speech recognition is performed using tools such as Google Speech-to-Text.

[0485] Step 2:

[0486] The server analyzes the converted text information using natural language processing techniques. Specifically, it processes the received text data using libraries such as spaCy and NLTK to understand its intent. In this step, the text information is the input, and the analyzed semantic structure is the output.

[0487] Step 3:

[0488] The server calculates similarity by comparing the analyzed semantic structure with previously recorded data in the database. It retrieves past input information from the database and applies it to a similarity calculation algorithm. The input here consists of the semantic structure and existing data, and the output is a similarity score.

[0489] Step 4:

[0490] If the similarity score exceeds a predetermined threshold, the server retrieves the corresponding response information from the database. In this step, the similarity score is the input, and the retrieved response information is the output. This information is then ready to be presented to the user immediately.

[0491] Step 5:

[0492] If no similar input information exists in the database, the server generates new response information using a generative AI model. Models such as OpenAI GPT are used, and a prompt statement is provided as input. The input in this case is the prompt statement, and the generated response information is the output.

[0493] Step 6:

[0494] The server sends the acquired or generated response information to the user's terminal. If the terminal has a visual output device, it is provided as displayed content. Here, the input is the response information, and the output is the information displayed in the user's field of vision.

[0495] Step 7:

[0496] The generated response information is recorded in a database for later reference. This enables quick responses to similar inputs in the future. The input is the response information, and the output is a new entry stored in the database.

[0497] Through these steps, a system is realized that can respond quickly and efficiently to user requests and provide appropriate content.

[0498] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0499] This invention is a system that incorporates sentiment analysis into the process of generating responses to input information, with the aim of improving the user experience. This system is configured to recognize the user's emotions in real time and provide appropriate responses based on those emotions.

[0500] The system is implemented as a software program running on multiple computers. First, the user sends questions and requests to the system via a terminal using natural language. The input information is then transferred to a server for analysis by an emotion engine.

[0501] The server not only records the received input information in a database, but also analyzes the user's emotions using an emotion engine. This emotion engine uses natural language processing to extract emotional elements from the input information and classify them into emotion categories such as positive, negative, and neutral.

[0502] Once the emotion of the input information is recognized, the server selects or generates an appropriate response based on that result. Specifically, it adjusts the tone and content to match the emotion based on existing response information. For example, if a user inputs "I'm busy and feeling down," the emotion engine recognizes the negative emotion and adjusts the response to provide an encouraging message or helpful information.

[0503] If similar past inputs exist, the server uses those responses to quickly provide a reply. If no similar tasks are found, the server uses a language model to generate a new response and records it in the database to prepare for similar tasks in the future.

[0504] Thus, the present invention enables the provision of responses that take user emotions into consideration, resulting in a more human-like interaction. This aims to provide information efficiently and comfortably.

[0505] The following describes the processing flow.

[0506] Step 1:

[0507] Users send questions and requests to the system in natural language through their terminal. The input information is in standard text format.

[0508] Step 2:

[0509] The server receives the input information submitted by the user and records it in data storage. Here, the input content, user ID, timestamp, and other related information are stored in the database.

[0510] Step 3:

[0511] The server passes the input information to the emotion engine, which analyzes the emotions embedded in the user's input. This analysis uses natural language processing techniques to identify emotions such as positive, negative, and neutral from the input text.

[0512] Step 4:

[0513] After the emotion engine recognizes the user's emotions, the server selects or adjusts its response to take those emotions into account. For example, if negative emotions are detected, the server will select a response that shows encouragement or empathy.

[0514] Step 5:

[0515] The server searches its past database for similar input information and calculates the similarity score. If a similar task exists, it uses the response information from that task.

[0516] Step 6:

[0517] If no similar task can be found, the server uses a language model to generate new response information. This process ensures that the generated response is appropriate to the user's emotional state.

[0518] Step 7:

[0519] The server records the generated response information in data storage and stores it for future inquiries.

[0520] Step 8:

[0521] Finally, the server sends the response information to the user's terminal, completing the information provision to the user.

[0522] (Example 2)

[0523] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0524] Traditional systems often produce formulaic responses to user input that do not take emotions or context into account, making it difficult to improve the user experience. In particular, generating responses that reflect the user's emotions is challenging, which is a factor that lowers user satisfaction.

[0525] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0526] In this invention, the server includes means for analyzing received input information and extracting emotional elements, means for selecting or generating appropriate response information based on emotional categories, and means for acquiring response information corresponding to past input information whose similarity exceeds a predetermined threshold. This enables more humane and personalized responses that respond to the user's emotions.

[0527] A "user" is an entity that provides input information to a system and receives a response.

[0528] "Input information" refers to questions and requests in natural language that users send to the system.

[0529] "Emotional elements" refer to information extracted from the user's emotions and related emotions contained in the input information.

[0530] "Response information" refers to the response content generated or selected by the system based on the user's input information.

[0531] "Natural language processing technology" is a general term for technologies that enable computers to understand, analyze, and generate natural language.

[0532] A "language model" is a model that learns the structure and patterns of natural language and generates new text.

[0533] "Similarity" is an indicator that shows the degree of similarity between the user's current input information and previously recorded input information.

[0534] This invention is a system for improving the user experience, and aims to provide appropriate responses by performing sentiment analysis based on user input information.

[0535] First, the user uses a terminal to input questions or requests in natural language and sends them to the system. The input information is transferred to the server in text format. As a concrete example, consider the case where the user inputs "I'm very tired today."

[0536] Next, the server receives this input information and first records it in a database. Then, it uses an emotion engine to analyze the emotional elements of the input information. The emotion engine uses natural language processing techniques to extract emotions such as positive, negative, and neutral from the input information.

[0537] Once sentiment analysis is complete, the server selects or generates appropriate response information based on the sentiment category. This process utilizes a generative AI model, which has the ability to create responses with appropriate tone and content by comparing them with past data.

[0538] A concrete example of a prompt message is, "My friend canceled our plans, and I'm feeling really down. What can I do to cheer myself up?" Based on this example, the server generates an empathetic response that reflects the user's emotions.

[0539] Finally, the server sends the generated response information to the terminal and displays it to the user. This system allows users to receive information in a way that is sensitive to their emotions, resulting in a more satisfying experience.

[0540] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0541] Step 1:

[0542] The user enters questions and requests in natural language into the terminal and sends them to the system. This input information is transferred to the server in text format. The input might be in the form of, for example, "I'm very tired today." The server receives this input data.

[0543] Step 2:

[0544] The server receives the input information and first records it in the database. By storing the input information in an identifiable format, it prepares for subsequent similarity checks and other processes. The database records not only the input information itself but also metadata such as the input time.

[0545] Step 3:

[0546] The server activates the emotion engine and analyzes the input information. Using natural language processing techniques, it extracts emotional elements from the input text and categorizes them as positive, negative, or neutral. Specifically, it analyzes keywords and context in the text to calculate an emotion score. The output is the extracted emotion category.

[0547] Step 4:

[0548] Based on the analysis results, the server determines response information appropriate to the emotion category. It searches the database for similar past inputs and retrieves responses with high similarity. If no suitable past responses are found, it uses a generative AI model to generate a new response. The model utilizes language building rules to create text that resonates with the user's emotions.

[0549] Step 5:

[0550] The server sends the generated or selected response information to the terminal. The user can receive this response on the terminal. The content of the response is emotionally sensitive to the user's input and may include, for example, "Thank you for your hard work. Please take some time to relax."

[0551] In this way, through a series of processes, it becomes possible to provide users with emotionally considerate responses.

[0552] (Application Example 2)

[0553] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0554] Traditional systems struggled to provide emotionally responsive answers to user inquiries and requests, often resulting in a poor user experience, particularly in online store customer support. Furthermore, the need to quickly provide responses tailored to positive or negative emotions posed a challenge in improving customer satisfaction and maintaining purchasing intent.

[0555] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0556] In this invention, the server includes means for receiving input information from a user, means for analyzing the received input information to recognize emotions, and means for calculating the similarity between the recognized emotions and previously recorded input information. This makes it possible to provide responses in an appropriate tone according to the user's emotions in online store customer support.

[0557] "Means for receiving user input information" refers to a device or program for acquiring information entered by a user via a communication terminal.

[0558] "An emotion analysis means that analyzes received input information to recognize emotions" refers to a program or device that analyzes the emotional state of a user based on the acquired user input information.

[0559] "Means for calculating the similarity with previously recorded input information" refers to algorithms or devices for comparing current input information with past records and calculating the similarity.

[0560] "Means for adjusting and providing response information acquired based on recognized emotions to the user" refers to a device or program that customizes responses with a tone and content appropriate to the analyzed emotions and communicates them to the user.

[0561] "Generating means for generating response information appropriate to an emotion when no similar input information exists" refers to a device or program for automatically generating an appropriate response corresponding to an emotion in response to new input.

[0562] "Means for recording generated response information" refers to a database or program for saving generated responses for later reference or improvement.

[0563] The system that implements this application example first operates by having the user provide input information using a terminal. The information sent from the terminal is transferred to a server via the network. The server utilizes natural language processing technologies such as the Google Cloud Natural Language API to perform sentiment analysis on the received input information.

[0564] The server extracts emotional elements from the input information and classifies them into emotional categories such as positive, negative, and neutral. Next, based on the results of the emotional analysis, it calculates the similarity to similar past input information and performs calculations to obtain the corresponding response information. Machine learning algorithms and database search engines are used to evaluate the similarity.

[0565] If no similar input information exists, the server uses a generative AI model to generate new response information. For example, if a user contacts store support about a product defect, the server can detect negative emotions and provide a tailored response that shows sympathy for the customer and aims to resolve the problem.

[0566] The generated response is delivered to the device in a tone that matches the user's emotions. This allows the user to have a comfortable support experience that is tailored to their feelings. In addition, the generated response information is recorded in a database by the server to improve the accuracy of support in the future.

[0567] For example, if a user sends a message saying, "The product I recently received was broken," the server can analyze this information and generate a response such as, "We apologize for the inconvenience. We will send you a replacement immediately."

[0568] An example of a prompt might be the instruction, "If the customer expresses negative emotions, generate specific solutions and sympathetic responses."

[0569] In this way, the server can generate and provide appropriate responses in real time based on the user's emotions.

[0570] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0571] Step 1:

[0572] The user sends input information via their device. This input information consists of data entered by the user in natural language, such as inquiries or requests, which the device then sends to the server via the network.

[0573] Step 2:

[0574] The server receives input information from the terminal and inputs that data into the sentiment analysis engine. The sentiment analysis engine uses the Google Cloud Natural Language API to extract key emotional elements from the input information and outputs positive, negative, or neutral emotional categories based on that.

[0575] Step 3:

[0576] The server calculates the similarity between the sentiment analysis results and past input information. This process uses a machine learning algorithm to calculate a similarity score between the received input information and past information stored in the database, and retrieves responses with high scores from the database.

[0577] Step 4:

[0578] If similar past information is found on the server, the server adjusts the corresponding response information and formats it to send to the terminal in a tone appropriate to the emotion. This includes the process of adjusting the tone of the response to be positive, negative, or neutral.

[0579] Step 5:

[0580] If no similar information is found on the server, a generative AI model is used to generate new response information. Using sentiment-based prompts, the language model generates a response, and its output is formatted as the default response.

[0581] Step 6:

[0582] The final generated response information is sent from the server to the terminal and presented to the user. Based on the response information received, the user can resolve the problem or take the next action.

[0583] Step 7:

[0584] The generated response and input information are recorded in a database on the server for future queries. This record is used as new data for later analysis and system improvements.

[0585] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0586] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0587] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0588] [Fourth Embodiment]

[0589] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0590] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0591] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0592] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0593] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0594] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0595] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0596] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0597] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0598] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0599] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0600] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0601] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0602] The present invention is implemented as a software program that runs on multiple computer systems. The central system of the invention aims to effectively process user input information and provide responses in a fast and power-efficient manner.

[0603] First, the server receives input information from the user. This input is data expressed in natural language as the user's questions or requests. The received input information is recorded directly in the database. This database record becomes the foundational data used for future similar task searches and response generation.

[0604] Next, the server compares the received input information with previously recorded data and calculates the similarity. Here, natural language processing techniques are used to evaluate the similarity. If the similarity exceeds a certain threshold, the server retrieves response information linked to past input information from the database and sends it back to the user.

[0605] For example, if a user asks "What is the population of XX city?", and there have been similar questions in the past such as "I want to know the number of residents in XX city," the system will immediately provide the answer from those previous questions. By doing this, the system utilizes existing answers to avoid generating new ones, thereby reducing power consumption.

[0606] If no similar task is found, the server invokes a large language model to generate new response information. The generated response information is recorded in the database and made available for reuse in future queries.

[0607] By implementing this system, redundant processing of the same information can be avoided, leading to efficient resource utilization. Thus, the present invention represents a form that realizes a sustainable and rapid information provision system.

[0608] The following describes the processing flow.

[0609] Step 1:

[0610] Users send input information, such as questions and requests, in natural language through their devices.

[0611] Step 2:

[0612] The server records the received input information in a database. This database stores the entered content, a timestamp, and, if necessary, the user ID.

[0613] Step 3:

[0614] The server uses newly recorded input information to calculate similarity with previously recorded data. This calculation employs natural language processing techniques to vectorize the input information for comparison.

[0615] Step 4:

[0616] The server searches the database for past tasks whose similarity exceeds a predetermined threshold. If a matching past task is found, it retrieves the relevant response information.

[0617] Step 5:

[0618] If a similar task is found, the server immediately provides an answer by sending the previously retrieved response information to the user's terminal.

[0619] Step 6:

[0620] If no similar task is found, the server will invoke a large language model to generate new response information. This process is expected to require significant time and computing resources.

[0621] Step 7:

[0622] The server records the generated response information in a database and stores it as reference data for future queries.

[0623] Step 8:

[0624] Finally, the server sends the newly generated response information to the user's terminal, completing the information provision process.

[0625] (Example 1)

[0626] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0627] In today's information-saturated environment, it is becoming increasingly difficult for users to obtain information quickly and efficiently. Furthermore, redundant processing of the same questions wastes resources, highlighting the growing need for efficient response systems.

[0628] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0629] In this invention, the server includes means for efficiently receiving input information from a user and recording the data, means for calculating the similarity between previously recorded information and the received information using natural language processing technology, and means for obtaining response information related to the past input information when the similarity exceeds a threshold. This makes it possible to provide the user with the necessary information quickly and in a power-efficient manner.

[0630] "Means for efficiently receiving user input and recording data" refers to a function that accurately and quickly receives natural language questions and requests provided by users to the system and systematically stores them in a database.

[0631] "Methods for calculating similarity using natural language processing technology" refers to techniques that analyze input natural language information and compare it with past records to evaluate how similar they are.

[0632] "Means for obtaining response information related to past input information" refers to a function that extracts the most relevant response from previously stored data based on the results of similarity calculations and sends it to the user.

[0633] "A means of activating a generative AI model to generate response information" refers to a function that uses a large-scale language model to create a new response when no similar past data exists.

[0634] "Means for recording generated response information for future use" refers to a function that efficiently stores newly generated response data and reuses it when similar inquiries are made later.

[0635] This invention is realized as an information processing system through the cooperation of a server, terminal, and user. First, the user inputs questions or requests in natural language. This input, specifically as a prompt, may include phrases such as "Please tell me the nearest cafe." This input is then sent from the terminal to the server.

[0636] The server first has the function of recording the received information in a database. This recorded information serves as an important foundation for reuse when similar queries occur later. The server then uses natural language processing techniques to compare the received input information with previously recorded information and calculate the similarity. Commonly used platforms and libraries (e.g., NLTK, spaCy, etc.) are used for natural language processing.

[0637] If the similarity exceeds a predetermined threshold, the server retrieves response information related to past input information from the database and sends it back to the user. This is a power-efficient design that conserves resources by making maximum use of existing information.

[0638] If no similar historical data is found, the server generates a new response using a generative AI model. This AI model could be a general-purpose language model (e.g., BERT, GPT series, etc.). The generated new response is then checked for quality and recorded in the database for future use. This process allows the system to flexibly adapt to new information.

[0639] This system enables rapid and power-efficient information delivery, allowing users to receive accurate and immediate answers. As a result, in this form, the present invention streamlines information management in large-scale systems and supports sustainable operation.

[0640] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0641] Step 1:

[0642] Users input questions and requests in natural language. For example, they might enter the prompt "Please tell me the nearest cafe" into a text box and press the submit button. The entered data is immediately sent to the server via the terminal. This input is natural language text data that includes the user's intent.

[0643] Step 2:

[0644] The server receives input information from the user. The received data is natural language text in string format. The server first records this data in a database. This prepares the foundational data necessary for subsequent similarity calculations and response generation. The output of this process is a new record in the database.

[0645] Step 3:

[0646] The server begins analyzing the received information using natural language processing techniques. First, it tokenizes the text data and extracts the meaning of each word. During this process, it uses a natural language processing library to analyze the text. Next, it calculates the similarity to previously recorded information. The input consists of raw text data and the contents of the past database, and the output is a similarity score.

[0647] Step 4:

[0648] The server retrieves past response information based on the similarity score if it exceeds a predetermined threshold. The input here is the similarity score, and the output is past response information. The retrieved response is then checked to ensure it aligns with the user's intent, and if deemed correct, it is returned to the user.

[0649] Step 5:

[0650] If similar information does not exist in the database, the server activates a generative AI model. Based on the input information, it processes the data to generate a new response. The input is text data containing the user's request, and the output is a new response in natural language text format. The generated response undergoes a quality check process before being provided to the user.

[0651] Step 6:

[0652] The server records the newly generated response information in the database. This allows for faster and more power-efficient responses to subsequent queries. The input in this step is the generated response information, and the output is the updated database information.

[0653] (Application Example 1)

[0654] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0655] In modern society, there is a demand for quick and efficient responses to user questions and requests, and the provision of appropriate information and content. However, conventional technologies require large-scale data processing, consume a lot of power, and make real-time information provision difficult. Furthermore, when using visual output devices, there is a lack of content recommendations that are appropriate to the user's situation. It is necessary to provide a system that solves these problems.

[0656] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0657] In this invention, the server includes means for receiving input information from a user, means for calculating the similarity with previously recorded input information, means for acquiring response information corresponding to past input information whose similarity exceeds a predetermined threshold, means for providing the acquired response information to the user, means for generating response information when no similar input information exists, means for recording the generated response information, means for recommending content to a visual output device based on user information, and means for converting speech into text information using a speech recognition device. This enables rapid and energy-efficient information provision in response to user input, and also enables appropriate content recommendation through a visual output device.

[0658] "Means for receiving user input information" refers to devices or mechanisms for collecting text and audio information provided by users.

[0659] "Means for calculating the similarity with previously recorded input information" refers to algorithms or processes for calculating the similarity between newly received input information and existing information in the database.

[0660] "Means for obtaining response information corresponding to past input information whose similarity exceeds a predetermined threshold" refers to a mechanism that extracts relevant information from a database when the calculated similarity meets a certain standard.

[0661] "Means of providing acquired response information to the user" refers to methods or devices for presenting collected information to the user.

[0662] "Generating means for generating response information when similar input information does not exist" refers to language generation models or systems for constructing and supplying the necessary information in response to new questions.

[0663] "Means for recording generated response information" refers to a technology for saving the generated information in a database for future use.

[0664] "A means of recommending content to a visual output device based on user information" refers to a system that analyzes user input data and presents suitable content through a display or similar device.

[0665] "Means for converting speech into text information using a speech recognition device" refers to technology for analyzing speech signals and converting them into text format.

[0666] To realize this invention, it is necessary to construct a system that receives user input information and performs appropriate information processing. The system is centered around a server and consists of multiple hardware and software components.

[0667] The server receives voice and text input from the user and first converts the voice input into text using a speech recognition device. Specific software options include speech recognition engines such as Google Speech-to-Text. Then, natural language processing techniques are used to analyze the text data and understand its meaning. At this stage, natural language processing libraries such as spaCy or NLTK are used.

[0668] Next, the server compares the received input information with previously recorded information and calculates the similarity. This similarity calculation is a process that compares the new request with data stored in the existing database and determines the degree of relevance. If the similarity exceeds a predetermined threshold, the corresponding response information is quickly retrieved from the database and the appropriate information is presented to the user's visual output device. This visual output device could be, for example, smart glasses. This allows the user to see appropriate content based on past similar information in real time.

[0669] If no similar input information exists, the server activates a generative AI model to generate new response information. This generated information is recorded in a database and stored in a reusable format for future requests. OpenAI GPT is a suitable example of a generative AI model.

[0670] For example, if a user asks "What movies do you recommend?" via voice, the server converts the voice into text and recommends the most suitable movies based on past history and related information. Based on this prompt, it is possible to input into a generative AI model that says, "When a user asks 'What movies do you recommend?', provide a response based on their browsing history and similar questions, and recommend the most suitable movie title."

[0671] This invention enables improved real-time user experience and enhances energy efficiency in information delivery.

[0672] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0673] Step 1:

[0674] The server receives voice and text input information from the user. In the case of voice input, a speech recognition device is used to convert the voice into text. In this process, the voice data is the input, and the text information is the output. Speech recognition is performed using tools such as Google Speech-to-Text.

[0675] Step 2:

[0676] The server analyzes the converted text information using natural language processing techniques. Specifically, it processes the received text data using libraries such as spaCy and NLTK to understand its intent. In this step, the text information is the input, and the analyzed semantic structure is the output.

[0677] Step 3:

[0678] The server calculates similarity by comparing the analyzed semantic structure with previously recorded data in the database. It retrieves past input information from the database and applies it to a similarity calculation algorithm. The input here consists of the semantic structure and existing data, and the output is a similarity score.

[0679] Step 4:

[0680] If the similarity score exceeds a predetermined threshold, the server retrieves the corresponding response information from the database. In this step, the similarity score is the input, and the retrieved response information is the output. This information is then ready to be presented to the user immediately.

[0681] Step 5:

[0682] If no similar input information exists in the database, the server generates new response information using a generative AI model. Models such as OpenAI GPT are used, and a prompt statement is provided as input. The input in this case is the prompt statement, and the generated response information is the output.

[0683] Step 6:

[0684] The server sends the acquired or generated response information to the user's terminal. If the terminal has a visual output device, it is provided as displayed content. Here, the input is the response information, and the output is the information displayed in the user's field of vision.

[0685] Step 7:

[0686] The generated response information is recorded in a database for later reference. This enables quick responses to similar inputs in the future. The input is the response information, and the output is a new entry stored in the database.

[0687] Through these steps, a system is realized that can respond quickly and efficiently to user requests and provide appropriate content.

[0688] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0689] This invention is a system that incorporates sentiment analysis into the process of generating responses to input information, with the aim of improving the user experience. This system is configured to recognize the user's emotions in real time and provide appropriate responses based on those emotions.

[0690] The system is implemented as a software program running on multiple computers. First, the user sends questions and requests to the system via a terminal using natural language. The input information is then transferred to a server for analysis by an emotion engine.

[0691] The server not only records the received input information in a database, but also analyzes the user's emotions using an emotion engine. This emotion engine uses natural language processing to extract emotional elements from the input information and classify them into emotion categories such as positive, negative, and neutral.

[0692] Once the emotion of the input information is recognized, the server selects or generates an appropriate response based on that result. Specifically, it adjusts the tone and content to match the emotion based on existing response information. For example, if a user inputs "I'm busy and feeling down," the emotion engine recognizes the negative emotion and adjusts the response to provide an encouraging message or helpful information.

[0693] If similar past inputs exist, the server uses those responses to quickly provide a reply. If no similar tasks are found, the server uses a language model to generate a new response and records it in the database to prepare for similar tasks in the future.

[0694] Thus, the present invention enables the provision of responses that take user emotions into consideration, resulting in a more human-like interaction. This aims to provide information efficiently and comfortably.

[0695] The following describes the processing flow.

[0696] Step 1:

[0697] Users send questions and requests to the system in natural language through their terminal. The input information is in standard text format.

[0698] Step 2:

[0699] The server receives the input information submitted by the user and records it in data storage. Here, the input content, user ID, timestamp, and other related information are stored in the database.

[0700] Step 3:

[0701] The server passes the input information to the emotion engine, which analyzes the emotions embedded in the user's input. This analysis uses natural language processing techniques to identify emotions such as positive, negative, and neutral from the input text.

[0702] Step 4:

[0703] After the emotion engine recognizes the user's emotions, the server selects or adjusts its response to take those emotions into account. For example, if negative emotions are detected, the server will select a response that shows encouragement or empathy.

[0704] Step 5:

[0705] The server searches its past database for similar input information and calculates the similarity score. If a similar task exists, it uses the response information from that task.

[0706] Step 6:

[0707] If no similar task can be found, the server uses a language model to generate new response information. This process ensures that the generated response is appropriate to the user's emotional state.

[0708] Step 7:

[0709] The server records the generated response information in data storage and stores it for future inquiries.

[0710] Step 8:

[0711] Finally, the server sends the response information to the user's terminal, completing the information provision to the user.

[0712] (Example 2)

[0713] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0714] Traditional systems often produce formulaic responses to user input that do not take emotions or context into account, making it difficult to improve the user experience. In particular, generating responses that reflect the user's emotions is challenging, which is a factor that lowers user satisfaction.

[0715] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0716] In this invention, the server includes means for analyzing received input information and extracting emotional elements, means for selecting or generating appropriate response information based on emotional categories, and means for acquiring response information corresponding to past input information whose similarity exceeds a predetermined threshold. This enables more humane and personalized responses that respond to the user's emotions.

[0717] A "user" is an entity that provides input information to a system and receives a response.

[0718] "Input information" refers to questions and requests in natural language that users send to the system.

[0719] "Emotional elements" refer to information extracted from the user's emotions and related emotions contained in the input information.

[0720] "Response information" refers to the response content generated or selected by the system based on the user's input information.

[0721] "Natural language processing technology" is a general term for technologies that enable computers to understand, analyze, and generate natural language.

[0722] A "language model" is a model that learns the structure and patterns of natural language and generates new text.

[0723] "Similarity" is an indicator that shows the degree of similarity between the user's current input information and previously recorded input information.

[0724] This invention is a system for improving the user experience, and aims to provide appropriate responses by performing sentiment analysis based on user input information.

[0725] First, the user uses a terminal to input questions or requests in natural language and sends them to the system. The input information is transferred to the server in text format. As a concrete example, consider the case where the user inputs "I'm very tired today."

[0726] Next, the server receives this input information and first records it in a database. Then, it uses an emotion engine to analyze the emotional elements of the input information. The emotion engine uses natural language processing techniques to extract emotions such as positive, negative, and neutral from the input information.

[0727] Once sentiment analysis is complete, the server selects or generates appropriate response information based on the sentiment category. This process utilizes a generative AI model, which has the ability to create responses with appropriate tone and content by comparing them with past data.

[0728] A concrete example of a prompt message is, "My friend canceled our plans, and I'm feeling really down. What can I do to cheer myself up?" Based on this example, the server generates an empathetic response that reflects the user's emotions.

[0729] Finally, the server sends the generated response information to the terminal and displays it to the user. This system allows users to receive information in a way that is sensitive to their emotions, resulting in a more satisfying experience.

[0730] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0731] Step 1:

[0732] The user enters questions and requests in natural language into the terminal and sends them to the system. This input information is transferred to the server in text format. The input might be in the form of, for example, "I'm very tired today." The server receives this input data.

[0733] Step 2:

[0734] The server receives the input information and first records it in the database. By storing the input information in an identifiable format, it prepares for subsequent similarity checks and other processes. The database records not only the input information itself but also metadata such as the input time.

[0735] Step 3:

[0736] The server activates the emotion engine and analyzes the input information. Using natural language processing techniques, it extracts emotional elements from the input text and categorizes them as positive, negative, or neutral. Specifically, it analyzes keywords and context in the text to calculate an emotion score. The output is the extracted emotion category.

[0737] Step 4:

[0738] Based on the analysis results, the server determines response information appropriate to the emotion category. It searches the database for similar past inputs and retrieves responses with high similarity. If no suitable past responses are found, it uses a generative AI model to generate a new response. The model utilizes language building rules to create text that resonates with the user's emotions.

[0739] Step 5:

[0740] The server sends the generated or selected response information to the terminal. The user can receive this response on the terminal. The content of the response is emotionally sensitive to the user's input and may include, for example, "Thank you for your hard work. Please take some time to relax."

[0741] In this way, through a series of processes, it becomes possible to provide users with emotionally considerate responses.

[0742] (Application Example 2)

[0743] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0744] Traditional systems struggled to provide emotionally responsive answers to user inquiries and requests, often resulting in a poor user experience, particularly in online store customer support. Furthermore, the need to quickly provide responses tailored to positive or negative emotions posed a challenge in improving customer satisfaction and maintaining purchasing intent.

[0745] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0746] In this invention, the server includes means for receiving input information from a user, means for analyzing the received input information to recognize emotions, and means for calculating the similarity between the recognized emotions and previously recorded input information. This makes it possible to provide responses in an appropriate tone according to the user's emotions in online store customer support.

[0747] "Means for receiving user input information" refers to a device or program for acquiring information entered by a user via a communication terminal.

[0748] "An emotion analysis means that analyzes received input information to recognize emotions" refers to a program or device that analyzes the emotional state of a user based on the acquired user input information.

[0749] "Means for calculating the similarity with previously recorded input information" refers to algorithms or devices for comparing current input information with past records and calculating the degree of similarity.

[0750] "Means for adjusting and providing response information acquired based on recognized emotions to the user" refers to a device or program that customizes responses with a tone and content appropriate to the analyzed emotions and communicates them to the user.

[0751] "Generating means for generating response information appropriate to an emotion when no similar input information exists" refers to a device or program for automatically generating an appropriate response corresponding to an emotion in response to new input.

[0752] "Means for recording generated response information" refers to a database or program for saving generated responses for later reference or improvement.

[0753] The system that implements this application example first operates by having the user provide input information using a terminal. The information sent from the terminal is transferred to a server via the network. The server utilizes natural language processing technologies such as the Google Cloud Natural Language API to perform sentiment analysis on the received input information.

[0754] The server extracts emotional elements from the input information and classifies them into emotional categories such as positive, negative, and neutral. Next, based on the results of the emotional analysis, it calculates the similarity to similar past input information and performs calculations to obtain the corresponding response information. Machine learning algorithms and database search engines are used to evaluate the similarity.

[0755] If no similar input information exists, the server uses a generative AI model to generate new response information. For example, if a user contacts store support about a product defect, the server can detect negative emotions and provide a tailored response that shows sympathy for the customer and aims to resolve the problem.

[0756] The generated response is delivered to the device in a tone that matches the user's emotions. This allows the user to have a comfortable support experience that is tailored to their feelings. In addition, the generated response information is recorded in a database by the server to improve the accuracy of support in the future.

[0757] For example, if a user sends a message saying, "The product I recently received was broken," the server can analyze this information and generate a response such as, "We apologize for the inconvenience. We will send you a replacement immediately."

[0758] An example of a prompt might be the instruction, "If the customer expresses negative emotions, generate specific solutions and sympathetic responses."

[0759] In this way, the server can generate and provide appropriate responses in real time based on the user's emotions.

[0760] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0761] Step 1:

[0762] The user sends input information via their device. This input information consists of data entered by the user in natural language, such as inquiries or requests, which the device then sends to the server via the network.

[0763] Step 2:

[0764] The server receives input information from the terminal and inputs that data into the sentiment analysis engine. The sentiment analysis engine uses the Google Cloud Natural Language API to extract key emotional elements from the input information and outputs positive, negative, or neutral emotional categories based on that.

[0765] Step 3:

[0766] The server calculates the similarity between the sentiment analysis results and past input information. This process uses a machine learning algorithm to calculate a similarity score between the received input information and past information stored in the database, and retrieves responses with high scores from the database.

[0767] Step 4:

[0768] If similar past information is found on the server, the server adjusts the corresponding response information and formats it to send to the terminal in a tone appropriate to the emotion. This includes the process of adjusting the tone of the response to be positive, negative, or neutral.

[0769] Step 5:

[0770] If no similar information is found on the server, a generative AI model is used to generate new response information. Using sentiment-based prompts, the language model generates a response, and its output is formatted as the default response.

[0771] Step 6:

[0772] The final generated response information is sent from the server to the terminal and presented to the user. Based on the response information received, the user can resolve the problem or take the next action.

[0773] Step 7:

[0774] The generated response and input information are recorded in a database on the server for future queries. This record is used as new data for later analysis and system improvements.

[0775] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0776] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0777] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0778] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0779] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0780] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0781] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0782] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0783] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0784] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0785] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0786] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0787] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0788] 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.

[0789] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0790] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0791] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0792] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0793] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0794] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0795] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[0796] The following is further disclosed regarding the embodiments described above.

[0797] (Claim 1)

[0798] A means of receiving input information from the user,

[0799] A means for calculating the similarity with previously recorded input information,

[0800] A means for obtaining response information corresponding to past input information whose similarity exceeds a predetermined threshold,

[0801] A means of providing the acquired response information to the user,

[0802] A generation means for generating response information when no similar input information exists,

[0803] A system including means for recording generated response information.

[0804] (Claim 2)

[0805] The system according to claim 1, which uses natural language processing techniques in similarity calculation.

[0806] (Claim 3)

[0807] The system according to claim 1, which uses a language model in generating response information.

[0808] "Example 1"

[0809] (Claim 1)

[0810] A means of efficiently receiving user input information and recording the data,

[0811] A means for calculating the similarity between previously recorded information and received information using natural language processing technology,

[0812] A means for obtaining response information related to past input information when the similarity exceeds a threshold,

[0813] A means of providing the acquired response information to the user in an energy-efficient manner,

[0814] A means of generating response information by activating a generative AI model when no similar information is available,

[0815] A system including means for recording generated response information for future use.

[0816] (Claim 2)

[0817] The system according to claim 1, which calculates the similarity of input information using natural language processing technology.

[0818] (Claim 3)

[0819] The system according to claim 1, which utilizes a generation AI model when generating response information.

[0820] "Application Example 1"

[0821] (Claim 1)

[0822] A means of receiving input information from the user,

[0823] A means for calculating the similarity with previously recorded input information,

[0824] A means for obtaining response information corresponding to past input information whose similarity exceeds a predetermined threshold,

[0825] A means of providing the acquired response information to the user,

[0826] A generation means for generating response information when no similar input information exists,

[0827] A means for recording the generated response information,

[0828] A means of recommending content to a visual output device based on user information,

[0829] A system including means for converting speech into text information using a speech recognition device.

[0830] (Claim 2)

[0831] The system according to claim 1, which uses natural language processing techniques in similarity calculation.

[0832] (Claim 3)

[0833] The system according to claim 1, which uses a language model in generating response information.

[0834] "Example 2 of combining an emotion engine"

[0835] (Claim 1)

[0836] A means of receiving input information from the user,

[0837] A means of analyzing received input information and extracting emotional elements,

[0838] Means for selecting or generating appropriate response information based on emotion categories,

[0839] A means for calculating the similarity with previously recorded input information,

[0840] A means for obtaining response information corresponding to past input information whose similarity exceeds a predetermined threshold,

[0841] A means of providing the acquired response information to the user,

[0842] A means for generating response information using a generation means when no similar input information exists,

[0843] A system including means for recording generated response information.

[0844] (Claim 2)

[0845] The system according to claim 1, which extracts emotional elements using natural language processing technology.

[0846] (Claim 3)

[0847] The system according to claim 1, which generates response information using a language model.

[0848] "Application example 2 when combining with an emotional engine"

[0849] (Claim 1)

[0850] A means of receiving input information from the user,

[0851] An emotion analysis means that analyzes received input information to recognize emotions,

[0852] A means for calculating the similarity to previously recorded input information based on recognized emotions,

[0853] A means for obtaining response information corresponding to past input information whose similarity exceeds a predetermined threshold,

[0854] A means of adjusting response information obtained based on recognized emotions and providing it to the user,

[0855] A generation means for generating response information appropriate to an emotion when no similar input information exists,

[0856] A system including means for recording generated response information.

[0857] (Claim 2)

[0858] The system according to claim 1, which uses natural language processing techniques in similarity calculation and sentiment recognition.

[0859] (Claim 3)

[0860] The system according to claim 1, which uses a language model to generate response information and generates response information in a tone adjusted based on the user's emotions. [Explanation of Symbols]

[0861] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of receiving input information from the user, A means for calculating the similarity with previously recorded input information, A means for obtaining response information corresponding to past input information whose similarity exceeds a predetermined threshold, A means of providing the acquired response information to the user, A generation means for generating response information when no similar input information exists, A system including means for recording generated response information.

2. The system according to claim 1, which uses natural language processing technology in similarity calculation.

3. The system according to claim 1, which uses a language model in generating response information.

Images