system

Abstract

We provide the system. [Solution] A means of collecting information on the user's past instructions, A means for performing natural language processing based on the aforementioned instruction information to preprocess the data, A means for automatically generating optimal suggestions for home appliances based on user preferences, A means for automatically inputting the generated proposal into the interface of a home appliance, A system that includes means for retraining to improve the accuracy of suggestions based on feedback.

Description

Technical Field

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

Background Art

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

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] Conventionally, when answering a questionnaire, users have to manually input the same basic information every time, which is time-consuming and reduces the efficiency of point acquisition. In particular, for users who complete a large number of questionnaires in a short period of time for the purpose of point activities, the time and effort required for answering have been a significant burden.

Means for Solving the Problems

[0005] To address this challenge, the present invention provides a system that collects users' past response data, preprocesses it using natural language processing, and then uses a machine learning model to predict the optimal answer to a question. Specifically, the system has a function that automatically generates answers to new questions using a model trained on past survey data and automatically inputs those answers into the interface. Furthermore, this system has a function to improve the accuracy of the model by incorporating feedback from users. As a result, users can significantly reduce the effort required to answer surveys and efficiently earn points.

[0006] "User" refers to an individual or group that uses the system to answer a survey.

[0007] "Response information" refers to user response data to survey questions.

[0008] "Natural language processing" refers to the technology used to process, analyze, and understand human language using computers.

[0009] A "machine learning model" refers to a collection of algorithms and models that learn from data and make predictions and decisions based on new information.

[0010] "Training" refers to the process by which a machine learning model adjusts its parameters using data to improve its prediction accuracy.

[0011] An "interface" refers to the screen or means by which a user interacts with a computer system.

[0012] "Feedback" refers to the opinions and evaluations provided by users after they have used the system. [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] Shows an emotion map to which multiple emotions are mapped. [Figure 10] Shows an emotion map to which multiple emotions are mapped. [Figure 11] It is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Example 2 when an emotion engine is combined. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when an emotion engine is combined.

Mode for Carrying Out the Invention

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

[0019] In the following embodiments, a numbered communication I / F (Interface) is an interface including a communication processor, an antenna, and the like. 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).

[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] This invention provides an AI agent system that streamlines the process of answering questionnaires. Based on the user's past questionnaire responses, this system utilizes natural language processing and machine learning techniques to automatically generate optimal answers to new questions.

[0035] The server first collects past survey response data from users and stores it in a database. This data includes the questions and the users' actual responses. The server uses this data to perform natural language processing, analyzing and preprocessing it. The processed data is then used to train a machine learning model, which learns question and answer patterns.

[0036] When a user receives a new survey, the device sends the question information to the server. The server uses a trained model to predict the best answer to the question and sends the result back to the device. The device automatically enters the returned answer into the survey form, significantly reducing the user's response time.

[0037] For example, suppose a user is asked, "What did you eat for dinner last night?" Based on past data, the server predicts the answer "pasta" and sends it to the device. The device automatically enters this prediction into the survey's answer field, allowing the user to answer the survey with minimal effort.

[0038] This system also includes a feature to improve model accuracy based on user feedback. The server analyzes the feedback information and retrains the model as needed, enabling the generation of more accurate responses. As a result, users can enjoy the benefit of answering surveys more efficiently and earning more points.

[0039] The following describes the processing flow.

[0040] Step 1:

[0041] The server collects users' past survey responses and stores them in a database. This data includes survey questions and their answers.

[0042] Step 2:

[0043] The server preprocesses the stored data using natural language processing techniques. Specifically, it performs text normalization, removes unnecessary information, and tokenizes the data.

[0044] Step 3:

[0045] The server trains a machine learning model using pre-processed data. Through this training, the model learns question-and-answer patterns.

[0046] Step 4:

[0047] When a user receives a new survey, the device sends the question content to the server.

[0048] Step 5:

[0049] The server uses a trained model to predict the best answer to the question it receives.

[0050] Step 6:

[0051] The server sends the predicted answer to the device. The device then automatically fills in this answer into the survey form.

[0052] Step 7:

[0053] The user reviews the answers automatically entered by the device and makes adjustments as needed.

[0054] Step 8:

[0055] The device sends user feedback to the server. The server analyzes this feedback and uses it to improve the machine learning model.

[0056] (Example 1)

[0057] 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."

[0058] Current survey systems present challenges because users must individually answer every question each time, which is time-consuming and labor-intensive. Furthermore, it is difficult for users to effectively reuse past answers to answer new questions, highlighting the need for greater efficiency. Additionally, mechanisms for continuous system improvement based on user feedback are insufficient.

[0059] 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.

[0060] In this invention, the server includes means for accumulating past user response data, means for performing language processing based on the response data to organize the information, and means for training a learning model using the organized information. This enables users to obtain optimal answers to new questions efficiently and quickly. Furthermore, by continuously improving the system based on user feedback, answers with higher accuracy can be provided.

[0061] "User" refers to an individual or organization that provides answers to questions using the survey system.

[0062] "Response data" refers to a set of information containing responses to questions previously asked by the user.

[0063] "Language processing" refers to the technology of analyzing and transforming text data expressed in natural language, and preparing it in a format that can be used by machine learning models.

[0064] "Organizing information" refers to the process of analyzing and classifying collected data to reconstruct it into a format necessary for learning.

[0065] A "learning model" refers to an algorithm that uses past data to discover new information patterns and predict answers to future questions.

[0066] "Accumulating" refers to the continuous collection and storage of data, specifically the act of saving it to a database for later use as needed.

[0067] "Feedback" refers to evaluations and opinions provided by users regarding the system's performance and the responses it generates.

[0068] This invention is an AI agent system designed to enable users to efficiently answer questionnaires. The specific method for implementing this system is described below.

[0069] The server first collects users' past response data and stores it in a database. General-purpose database software is used for data management. This data includes previously answered questions and their responses. To analyze the text data, the server employs "spaCy" as natural language processing software to perform preprocessing such as data cleaning, tokenization, and tagging.

[0070] The preprocessed data is used to train a model using the machine learning framework "TENSORFLOW®". This model has the ability to learn past question and answer patterns and predict the optimal answer to new questions. For example, data where a user previously answered "apple" to the question "What is your favorite fruit?" is analyzed by the model, and an appropriate answer is generated for a new question with a similar pattern.

[0071] When a new survey question arrives, the user's device sends it to the server. The server then uses a trained model to predict the best answer to the question and sends the result back to the device. The device automatically enters the received answer into the survey form. This allows users to answer surveys quickly, saving them time and effort.

[0072] This system also includes a feature that allows the server to improve the accuracy of the model based on user feedback. By analyzing the feedback information, the server retrains the model as needed to improve the accuracy of the answers. In this process, a generative AI model is used, and prompts such as "Predict the best answer to the survey question 'What is your favorite fruit?'" are used.

[0073] In this way, users can efficiently answer surveys and enjoy more benefits through the system.

[0074] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0075] Step 1:

[0076] The server collects the user's past response data. The input consists of the user's previous survey questions and their responses. The server stores this data in a database for later analysis. Specifically, a database management system is used for data storage and organization. The output is the stored response data in an accessible format.

[0077] Step 2:

[0078] The server preprocesses the response data collected from the database using natural language processing (NLP). The input consists of stored user questions and answers. The server uses "spaCy" to clean up, tokenize, and tag the text. Specifically, this involves removing unnecessary symbols and spaces, splitting words, and tagging parts of speech. The output is the refined text data.

[0079] Step 3:

[0080] The server trains a machine learning model using pre-processed data. The input is pre-processed question and answer data. The server uses TensorFlow to learn patterns based on the data. Specifically, the model associates questions with past answers and learns patterns for what answers to predict for new questions. The output is the trained model.

[0081] Step 4:

[0082] The terminal receives a new survey question from the user and sends it to the server. The input is the newly provided survey question. The terminal simply sends this to the server. The output is the transmission of the accurate question information to the server.

[0083] Step 5:

[0084] The server uses a trained model to predict the best answer to a new question. The input is new question information received from the terminal. The server uses the model to generate the best answer based on past data. Specifically, it applies the received question to the model and estimates the most likely answer. The output is the predicted answer.

[0085] Step 6:

[0086] The terminal receives predicted responses sent from the server and automatically enters them into the survey form. The input is the predicted response data from the server. The terminal then reflects the received responses in the user interface. The output is the responses automatically entered into the form.

[0087] Step 7:

[0088] Users provide feedback on the system-generated answers. The input consists of the user's evaluation and opinion. The user then sends this feedback to the server. The output is the feedback information received by the server.

[0089] Step 8:

[0090] The server analyzes user feedback and retrains the machine learning model as needed. The input is feedback information based on user evaluations. The server uses this feedback to improve the model's accuracy. Specifically, it identifies incorrect response patterns and adjusts the model. The output is the improved trained model.

[0091] (Application Example 1)

[0092] 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."

[0093] In home appliances, a challenge is to improve the convenience of daily life by providing quick and accurate optimal suggestions based on the user's past instructions. In particular, improving the accuracy of suggestions through relearning using feedback is required.

[0094] 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.

[0095] In this invention, the server includes means for collecting past instruction information of the user, means for performing natural language processing on the instruction information to preprocess the data, and means for automatically generating optimal suggestions for home appliances based on the user's preferences. This makes it possible to improve the efficiency of operating and setting up home appliances.

[0096] "Instruction information" refers to data that records past requests and wishes made by users regarding home appliances.

[0097] "Natural language processing" is a technology that uses computers to process, understand, and generate human language.

[0098] "Household appliances" refer to electrical and electronic devices used in the home, and are devices that support various tasks in daily life.

[0099] A "suggestion" is advice regarding optimal actions and settings generated based on the user's instructions and preferences.

[0100] "Retraining" is an additional training process performed to improve the accuracy of a machine learning model based on feedback from users.

[0101] The system for implementing this invention is designed to provide convenient operation and setting support for household appliances. The system collects information on the user's past instructions and generates optimal suggestions based on that information, thereby streamlining the user's daily life.

[0102] The server stores the user's past instruction information in a database and analyzes it using natural language processing techniques. Specifically, it uses Python and natural language processing libraries such as NLTK and spaCy to preprocess the data. The processed data is then trained using machine learning models, such as TensorFlow or PyTorch. During this training process, the model is built to automatically generate optimal suggestions for the current situation, taking into account the instruction information and the user's preferences.

[0103] Terminals that control home appliances receive suggestions sent from a server and automatically input them into the user's interface. For example, based on past data, a suggestion such as "play relaxing music at night" might be made. This allows users to save time on operating the devices and enjoy a more comfortable life.

[0104] Furthermore, based on user feedback, the server retrains the model to improve the system's accuracy. This feedback allows the model to learn a wider variety of patterns, enabling more personalized suggestions.

[0105] For example, if you instruct a household robot to "clean up after dinner," the robot will suggest and execute an efficient cleaning method based on reliable data. An example of a prompt to the generative AI model would be, "Please tell me what I should do to create a comfortable environment in the living room."

[0106] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0107] Step 1:

[0108] The server collects past user instruction information and stores it in a database. The input is user instruction data, and the output is structured database input. This process allows the server to secure the basic data necessary for subsequent analysis.

[0109] Step 2:

[0110] The server uses natural language processing on the collected instruction information as input to perform data preprocessing. This involves data cleansing and tokenization, generating data in a format suitable for input to the model as output. Specifically, this is done using NLTK or spaCy.

[0111] Step 3:

[0112] The server trains a machine learning model using pre-processed data. The input is the transformed data, and the output is a trained model that links instruction information with preferences. TensorFlow or PyTorch is used to learn patterns based on user preferences.

[0113] Step 4:

[0114] The server uses a pre-trained model to generate optimal suggestions for new user requests. The model takes prompts as input and outputs suggested content. A concrete example would be a suggestion like "Create a list of recommended music for after dinner."

[0115] Step 5:

[0116] The terminal receives the generated suggestions and automatically inputs the output into the interface of the home appliance. The input is suggestion data from the server, and the output is the configuration changes or tasks performed on the appliance. This automation reduces user effort and improves convenience.

[0117] Step 6:

[0118] The user provides feedback on the proposal to the server. The input is the user's feedback information, and based on this, the server obtains data as output to retrain the model. Feedback is a crucial element for further improving the accuracy of the proposal.

[0119] 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.

[0120] This invention provides a system that further personalizes and improves the accuracy of survey response predictions by incorporating an emotion engine into an AI agent system. This system combines the user's past survey response data and emotion data, and utilizes natural language processing and machine learning techniques to automatically generate the optimal response.

[0121] The server collects users' past survey responses and related sentiment data and stores them in a database. Sentiment data is extracted from the user's voice and text input, and the sentiment engine identifies their emotional state. The server preprocesses the collected data using natural language processing techniques and analyzes the data. Subsequently, it trains a machine learning model to learn past sentiment patterns.

[0122] The device interacts with the server when the user receives a survey. The server uses a trained model to predict the best answer to a new question, and this prediction reflects the user's emotional state. The device automatically fills in the predicted answer into the survey form, improving user convenience.

[0123] For example, when a user is asked "How are you feeling today?", the server predicts an answer by referring to past data on how users have answered similar questions in the past, while also considering their current emotional state (for example, the emotion "happy" extracted from a text message). The terminal displays the prediction result for the user to review, allowing them to complete the answer with minimal effort.

[0124] Furthermore, user feedback is sent to the server and used to improve the sentiment engine and machine learning models. This process allows the system to continuously learn, enabling it to generate more accurate and personalized responses.

[0125] The following describes the processing flow.

[0126] Step 1:

[0127] The server collects users' past survey responses and emotional data and stores it in a database. The emotional data includes emotional states extracted from users' voice and text.

[0128] Step 2:

[0129] The server uses natural language processing techniques to preprocess the stored data. Specifically, it normalizes the data and performs sentiment analysis as needed to transform it into a format suitable for machine learning models.

[0130] Step 3:

[0131] The server uses pre-processed data to train a machine learning model. This model learns past question and answer patterns, as well as the user's emotional state.

[0132] Step 4:

[0133] When a user receives a new survey, the device sends the question and current sentiment data to the server. The sentiment data reflects the user's real-time emotional state.

[0134] Step 5:

[0135] The server uses a trained model to predict the best answer to an incoming question. This prediction is influenced by the user's past response history and current sentiment state.

[0136] Step 6:

[0137] The server sends the predicted answer to the device. The device then automatically fills in this answer into the survey form and makes it available for the user to review.

[0138] Step 7:

[0139] Users will have the opportunity to review the answers entered by their device and correct them if necessary.

[0140] Step 8:

[0141] The device receives feedback from the user and sends it to the server. The server uses this feedback to further improve the accuracy of the emotion engine and machine learning models.

[0142] (Example 2)

[0143] 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 will be referred to as the "terminal."

[0144] The present invention aims to provide a system that reduces the burden on users when answering questionnaires and enables more accurate and personalized response predictions based on individual emotional states. The goal is to improve the accuracy of questionnaires and user satisfaction.

[0145] 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.

[0146] In this invention, the server includes means for collecting past response information and sentiment information of users; means for performing natural language processing on the response information and sentiment information to preprocess the data; and means for training a machine learning model using the preprocessed data to learn sentiment patterns. This enables accurate and personalized response predictions that take into account the user's emotional state.

[0147] "User" refers to an individual or legal entity that uses the system to answer a survey.

[0148] "Response information" refers to information and data that users have selected or entered in past surveys.

[0149] "Emotional information" refers to data that indicates a user's psychological or emotional state, extracted from their voice or text.

[0150] "Natural language processing" refers to the technologies and methods that enable computers to understand, interpret, and generate human language.

[0151] "Data preprocessing" refers to the process of shaping or transforming data before performing analysis or model training.

[0152] A "machine learning model" refers to an algorithm or structure used to learn patterns from data and perform predictions or classifications.

[0153] "Emotional patterns" refer to a series of tendencies and characteristics extracted based on the user's emotional information.

[0154] "Prediction" refers to the process by which a machine learning model uses training data to infer or estimate results for new data.

[0155] An "interface" refers to the screen or control panel that allows a user to interact with a system.

[0156] "Feedback" refers to information such as evaluations, opinions, and suggestions for improvement that users provide to the system.

[0157] "Accuracy" is an indicator that shows how accurately a system can respond to the user's intentions and requests.

[0158] This system provides a function that optimizes user survey responses through collaboration between the server and terminals. It primarily utilizes user sentiment information to achieve personalized response predictions.

[0159] server

[0160] The server collects users' past survey responses and sentiment information and stores it in a database. Sentiment information is extracted using speech recognition software and text analysis tools. Specifically, a general speech analysis service is used for speech recognition, and Python's natural language processing libraries (e.g., NLTK and spaCy) are used for text analysis.

[0161] The server preprocesses the collected data, performing tasks such as noise reduction and tokenization, and then formats the data for machine learning models. Machine learning is performed using platforms such as Scikit-learn and TensorFlow to train the models. These models are used to learn user sentiment patterns and predict answers to new questions.

[0162] terminal

[0163] The terminal receives predicted answers sent from the server and automatically inputs them into the user interface. This allows users to complete surveys with minimal effort. User feedback is also sent to the server via the terminal, contributing to the continuous improvement of the system.

[0164] Specific example

[0165] For example, when a user receives a survey question such as "How are you feeling today?", the server can predict the answer "I'm having fun today" based on past data and the user's current emotional state. As a result, the terminal automatically inputs this prediction, and after user confirmation, the survey is completed easily.

[0166] Example of a prompt

[0167] Based on the user's past survey responses and sentiment data, and taking into account their current emotional state, generate a predictive answer to the following question: 'How are you feeling today?'

[0168] This process allows the system to automatically generate more appropriate responses for each user, improving the accuracy of the survey and user satisfaction.

[0169] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0170] Step 1:

[0171] The server collects users' past survey responses and sentiment information and stores it in a database. Specifically, the server uses speech recognition software to extract text from audio data and a text analysis tool to extract sentiment information. It receives past audio and text data as input and stores the response information and sentiment information in the database as output.

[0172] Step 2:

[0173] The server preprocesses the collected data using natural language processing techniques. It performs noise reduction and tokenization on the data, converting the text data into a parseable format. It receives collected response information and sentiment information as input and generates preprocessed data as output. Specifically, it uses Python's NLTK to format the text data.

[0174] Step 3:

[0175] The server trains a machine learning model using preprocessed data. It learns patterns based on the features of the input data and builds a model. It receives preprocessed data as input and generates a trained machine learning model as output. This process utilizes Scikit-learn and TensorFlow to iteratively adjust the model's parameters.

[0176] Step 4:

[0177] The server uses a trained model to predict the best answer to a new survey question. It inputs current sentiment information into the model and calculates the best answer. It accepts a new question and current sentiment information as input and provides a predicted answer as output. Specifically, it uses prompt statements to supply the model with the question text and related data, taking sentiment states into account.

[0178] Step 5:

[0179] The terminal automatically populates the survey form with predicted answers sent from the server and allows the user to confirm them. It receives the predicted results from the server as input and displays the automatically populated state on the user interface as output. The user can then review and modify the data. Specifically, the terminal interface reflects the predicted answers and presents them to the user as editable fields.

[0180] Step 6:

[0181] The user sends feedback to the server via their device. The server receives the user feedback after review as input and provides data as output that is used for the system's continuous learning. The server, upon receiving the feedback, applies it to improve the sentiment engine and machine learning model to improve the accuracy of future predictions. Specifically, a process is executed to retrain the model using the new feedback data as input.

[0182] (Application Example 2)

[0183] 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 device 14 will be referred to as the "terminal."

[0184] In today's world, there is a need to optimize users' purchasing intentions and provide highly accurate product recommendations. However, conventional systems have been unable to reflect the individual emotional state of users in their recommendations, resulting in a decrease in the usefulness of the suggestions. To address this situation, there is a need to provide a system that considers past response information and current emotional state to provide personalized recommendations.

[0185] 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.

[0186] In this invention, the server includes means for collecting past response information and emotional data of the user, means for performing natural language processing on the response information and emotional data to preprocess the data, and means for training a machine learning model using the preprocessed data. This enables personalized suggestions based on the user's emotional tendencies.

[0187] A "user" is an individual or legal entity that uses the system or service.

[0188] "Response information" refers to the content of responses that users have given to past surveys or questions.

[0189] "Emotional data" refers to data that indicates the user's emotional state, and is extracted from voice and text input.

[0190] "Natural language processing" is a technology that enables computers to understand and process human language.

[0191] "Data preprocessing" is the process of preparing data to be suitable for machine learning algorithms.

[0192] A "machine learning model" is a statistical model that learns patterns from data to perform predictions and classifications.

[0193] "Training methods" refer to the process of providing data to a machine learning model and allowing it to learn patterns and relationships.

[0194] "Emotional state" refers to information that indicates the user's current psychological or emotional condition.

[0195] "Personalized recommendations" refer to recommendations for products and services tailored to the specific needs and emotions of the user.

[0196] A "connection device" is a device used by users to access the system.

[0197] The system for implementing the present invention includes a program that collects the user's past response information and emotional data, and utilizes natural language processing and machine learning techniques. The server runs on a platform such as Python or TensorFlow and extracts emotional data from the user's text and voice data. The extracted data is stored in a database and then used to train a machine learning model after preprocessing.

[0198] The server preprocesses data using a natural language processing toolkit (e.g., NLTK) and optimizes machine learning models. Based on the user's response patterns and emotional tendencies, it can generate personalized suggestions. These suggestions, which take into account the user's current emotional state, are provided on the user's device.

[0199] When a user accesses the system using a connected device such as a smartphone, the device presents the user with predicted responses and suggestions. The user's feedback is then sent back to the server and used to improve the model's performance. This feedback process allows the system to continuously improve the accuracy of the suggestions it provides.

[0200] For example, if a user enters "I'm feeling good today," the server will recommend appropriate products from those that the user has considered purchasing in similar emotional states in the past. An example of a prompt for the generative AI model used in this process could be in the form of, "What products would you recommend if the user's emotional data indicates they are feeling good?"

[0201] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0202] Step 1:

[0203] The server receives the user's past survey responses and emotional data collected from voice, text, etc. The input data includes the user's response history and information indicating their emotional state. The server stores this data in a database for later analysis. During this process, data cleansing is performed to remove inaccurate or invalid data.

[0204] Step 2:

[0205] The server preprocesses the collected response information and emotional data using a natural language processing toolkit (such as NLTK). This step converts the input data into a format that can be easily handled by machine learning models. Specifically, it performs tasks such as text tokenization, removal of special characters, and extraction of sentiment words to extract meaningful features.

[0206] Step 3:

[0207] The server trains a machine learning model using pre-processed data. The input data consists of past responses and emotional patterns extracted as features. The output is a model that generates responses and product suggestions based on the user's emotional state. This process utilizes libraries such as TensorFlow to train a neural network.

[0208] Step 4:

[0209] The server receives the user's current questions and emotional state as input and uses a trained model to predict the best response and product suggestion. The input data includes real-time user emotional data. As output, it generates personalized suggestions and responses and sends them to the terminal.

[0210] Step 5:

[0211] The terminal presents the user with predicted responses and suggestions received from the server. In this step, the suggestions are visually displayed through the user interface for the user to review. The user can then make decisions, select, or change based on the suggestions.

[0212] Step 6:

[0213] The device receives user feedback and sends it to the server. The input data consists of user opinions regarding the usefulness and satisfaction level of the suggestions. The server uses this feedback to retrain the machine learning model to improve its performance. This makes it possible to further improve the accuracy of future suggestions.

[0214] 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.

[0215] 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.

[0216] 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.

[0217] [Second Embodiment]

[0218] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0219] 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.

[0220] 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).

[0221] 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.

[0222] 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.

[0223] 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).

[0224] 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.

[0225] 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.

[0226] 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.

[0227] 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.

[0228] 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.

[0229] 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".

[0230] This invention provides an AI agent system that streamlines the process of answering questionnaires. Based on the user's past questionnaire responses, this system utilizes natural language processing and machine learning techniques to automatically generate optimal answers to new questions.

[0231] The server first collects past survey response data from users and stores it in a database. This data includes the questions and the users' actual responses. The server uses this data to perform natural language processing, analyzing and preprocessing it. The processed data is then used to train a machine learning model, which learns question and answer patterns.

[0232] When a user receives a new survey, the device sends the question information to the server. The server uses a trained model to predict the best answer to the question and sends the result back to the device. The device automatically enters the returned answer into the survey form, significantly reducing the user's response time.

[0233] For example, suppose a user is asked, "What did you eat for dinner last night?" Based on past data, the server predicts the answer "pasta" and sends it to the device. The device automatically enters this prediction into the survey's answer field, allowing the user to answer the survey with minimal effort.

[0234] This system also includes a feature to improve model accuracy based on user feedback. The server analyzes the feedback information and retrains the model as needed, enabling the generation of more accurate responses. As a result, users can enjoy the benefit of answering surveys more efficiently and earning more points.

[0235] The following describes the processing flow.

[0236] Step 1:

[0237] The server collects users' past survey responses and stores them in a database. This data includes survey questions and their answers.

[0238] Step 2:

[0239] The server preprocesses the stored data using natural language processing techniques. Specifically, it performs text normalization, removes unnecessary information, and tokenizes the data.

[0240] Step 3:

[0241] The server trains a machine learning model using pre-processed data. Through this training, the model learns question-and-answer patterns.

[0242] Step 4:

[0243] When a user receives a new survey, the device sends the question content to the server.

[0244] Step 5:

[0245] The server uses a trained model to predict the best answer to the question it receives.

[0246] Step 6:

[0247] The server sends the predicted answer to the device. The device then automatically fills in this answer into the survey form.

[0248] Step 7:

[0249] The user reviews the answers automatically entered by the device and makes adjustments as needed.

[0250] Step 8:

[0251] The device sends user feedback to the server. The server analyzes this feedback and uses it to improve the machine learning model.

[0252] (Example 1)

[0253] 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".

[0254] Current survey systems present challenges because users must individually answer every question each time, which is time-consuming and labor-intensive. Furthermore, it is difficult for users to effectively reuse past answers to answer new questions, highlighting the need for greater efficiency. Additionally, mechanisms for continuous system improvement based on user feedback are insufficient.

[0255] 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.

[0256] In this invention, the server includes means for accumulating past user response data, means for performing language processing based on the response data to organize the information, and means for training a learning model using the organized information. This enables users to obtain optimal answers to new questions efficiently and quickly. Furthermore, by continuously improving the system based on user feedback, answers with higher accuracy can be provided.

[0257] "User" refers to an individual or organization that provides answers to questions using the survey system.

[0258] "Response data" refers to a set of information containing responses to questions previously asked by the user.

[0259] "Language processing" refers to the technology of analyzing and transforming text data expressed in natural language, and preparing it in a format that can be used by machine learning models.

[0260] "Organizing information" refers to the process of analyzing and classifying collected data to reconstruct it into a format necessary for learning.

[0261] A "learning model" refers to an algorithm that uses past data to discover new information patterns and predict answers to future questions.

[0262] "Accumulating" refers to the continuous collection and storage of data, specifically the act of saving it to a database for later use as needed.

[0263] "Feedback" refers to evaluations and opinions provided by users regarding the system's performance and the responses it generates.

[0264] This invention is an AI agent system designed to enable users to efficiently answer questionnaires. The specific method for implementing this system is described below.

[0265] The server first collects users' past response data and stores it in a database. General-purpose database software is used for data management. This data includes previously answered questions and their responses. To analyze the text data, the server employs "spaCy" as natural language processing software to perform preprocessing such as data cleaning, tokenization, and tagging.

[0266] The preprocessed data is used to train a model using the machine learning framework "TensorFlow." This model has the ability to learn past question and answer patterns and predict the best answer to new questions. For example, data where a user previously answered "apple" to the question "What is your favorite fruit?" is analyzed by the model, and it generates an appropriate answer to a new question with a similar pattern.

[0267] When a new survey question arrives, the user's device sends it to the server. The server then uses a trained model to predict the best answer to the question and sends the result back to the device. The device automatically enters the received answer into the survey form. This allows users to answer surveys quickly, saving them time and effort.

[0268] This system also includes a feature that allows the server to improve the accuracy of the model based on user feedback. By analyzing the feedback information, the server retrains the model as needed to improve the accuracy of the answers. In this process, a generative AI model is used, and prompts such as "Predict the best answer to the survey question 'What is your favorite fruit?'" are used.

[0269] In this way, users can efficiently answer surveys and enjoy more benefits through the system.

[0270] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0271] Step 1:

[0272] The server collects the user's past response data. The input consists of the user's previous survey questions and their responses. The server stores this data in a database for later analysis. Specifically, a database management system is used for data storage and organization. The output is the stored response data in an accessible format.

[0273] Step 2:

[0274] The server preprocesses the response data collected from the database using natural language processing (NLP). The input consists of stored user questions and answers. The server uses "spaCy" to clean up, tokenize, and tag the text. Specifically, this involves removing unnecessary symbols and spaces, splitting words, and tagging parts of speech. The output is the refined text data.

[0275] Step 3:

[0276] The server trains a machine learning model using pre-processed data. The input is pre-processed question and answer data. The server uses TensorFlow to learn patterns based on the data. Specifically, the model associates questions with past answers and learns patterns for what answers to predict for new questions. The output is the trained model.

[0277] Step 4:

[0278] The terminal receives a new survey question from the user and sends it to the server. The input is the newly provided survey question. The terminal simply sends this to the server. The output is the transmission of the accurate question information to the server.

[0279] Step 5:

[0280] The server uses a trained model to predict the best answer to a new question. The input is new question information received from the terminal. The server uses the model to generate the best answer based on past data. Specifically, it applies the received question to the model and estimates the most likely answer. The output is the predicted answer.

[0281] Step 6:

[0282] The terminal receives the predicted answer sent from the server and automatically enters it into the questionnaire form. The input is the predicted answer data from the server. The terminal performs an operation to reflect the received answer on the user interface. The output is the answer automatically entered into the form.

[0283] Step 7:

[0284] The user provides feedback on the answer generated by the system. The input is the user's evaluation and opinion. The user performs an operation to send the feedback to the server. The output is the feedback information received by the server.

[0285] Step 8:

[0286] The server analyzes the feedback from the user and, if necessary, re-implements the training of the machine learning model. The input is the feedback information based on the user's evaluation. The server attempts to improve the accuracy of the model based on this feedback. Specifically, operations such as identifying incorrect answer patterns and adjusting the model are performed. The output is the improved learning model.

[0287] (Application Example 1)

[0288] Next, Application Example 1 will be described. In the following description, the data processing device 12 is referred to as the "server", and the smart glasses 214 are referred to as the "terminal".

[0289] In household appliances, it is a challenge to improve the convenience of daily life by quickly and accurately making optimal proposals based on the user's past instructions. In particular, it is required to improve the accuracy of the proposals through re-learning using feedback.

[0290] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0291] In this invention, the server includes means for collecting past instruction information of the user, means for performing natural language processing on the instruction information to preprocess the data, and means for automatically generating optimal suggestions for home appliances based on the user's preferences. This makes it possible to improve the efficiency of operating and setting up home appliances.

[0292] "Instruction information" refers to data that records past requests and wishes made by users regarding home appliances.

[0293] "Natural language processing" is a technology that uses computers to process, understand, and generate human language.

[0294] "Household appliances" refer to electrical and electronic devices used in the home, and are devices that support various tasks in daily life.

[0295] A "suggestion" is advice regarding optimal actions and settings generated based on the user's instructions and preferences.

[0296] "Retraining" is an additional training process performed to improve the accuracy of a machine learning model based on feedback from users.

[0297] The system for implementing this invention is designed to provide convenient operation and setting support for household appliances. The system collects information on the user's past instructions and generates optimal suggestions based on that information, thereby streamlining the user's daily life.

[0298] The server stores the user's past instruction information in a database and analyzes it using natural language processing techniques. Specifically, it uses Python and natural language processing libraries such as NLTK and spaCy to preprocess the data. The processed data is then trained using machine learning models, such as TensorFlow or PyTorch. During this training process, the model is built to automatically generate optimal suggestions for the current situation, taking into account the instruction information and the user's preferences.

[0299] Terminals that control home appliances receive suggestions sent from a server and automatically input them into the user's interface. For example, based on past data, a suggestion such as "play relaxing music at night" might be made. This allows users to save time on operating the devices and enjoy a more comfortable life.

[0300] Furthermore, based on user feedback, the server retrains the model to improve the system's accuracy. This feedback allows the model to learn a wider variety of patterns, enabling more personalized suggestions.

[0301] For example, if you instruct a household robot to "clean up after dinner," the robot will suggest and execute an efficient cleaning method based on reliable data. An example of a prompt to the generative AI model would be, "Please tell me what I should do to create a comfortable environment in the living room."

[0302] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0303] Step 1:

[0304] The server collects past user instruction information and stores it in a database. The input is user instruction data, and the output is structured database input. This process allows the server to secure the basic data necessary for subsequent analysis.

[0305] Step 2:

[0306] The server uses natural language processing on the collected instruction information as input to perform data preprocessing. This involves data cleansing and tokenization, generating data in a format suitable for input to the model as output. Specifically, this is done using NLTK or spaCy.

[0307] Step 3:

[0308] The server trains a machine learning model using the preprocessed data. The input is the converted data, and the output is a trained model that correlates instruction information with preferences. Using TensorFlow or PyTorch, it learns patterns based on the user's preferences.

[0309] Step 4:

[0310] The server uses the trained model to generate an optimal proposal for the requests of new users. A prompt is input into the model, and the proposal content is generated as the output. As a specific example, a proposal such as "Create a recommended music list after dinner" is made.

[0311] Step 5:

[0312] The terminal receives the generated proposal and automatically inputs the output to the interface of the household device. The input is the proposal data from the server, and the output is the setting changes on the device and the tasks executed. This automation saves the user's effort and improves convenience.

[0313] Step 6:

[0314] The user provides feedback on the proposal to the server. The input is the feedback information from the user, and based on this, the server obtains data for re-training the model as the output. Feedback is an important factor for further improving the proposal accuracy.

[0315] 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 recognition model 59 and perform specific processing using the user's emotion.

[0316] This invention provides a system that further personalizes and improves the accuracy of survey response predictions by incorporating an emotion engine into an AI agent system. This system combines the user's past survey response data and emotion data, and utilizes natural language processing and machine learning techniques to automatically generate the optimal response.

[0317] The server collects users' past survey responses and related sentiment data and stores them in a database. Sentiment data is extracted from the user's voice and text input, and the sentiment engine identifies their emotional state. The server preprocesses the collected data using natural language processing techniques and analyzes the data. Subsequently, it trains a machine learning model to learn past sentiment patterns.

[0318] The device interacts with the server when the user receives a survey. The server uses a trained model to predict the best answer to a new question, and this prediction reflects the user's emotional state. The device automatically fills in the predicted answer into the survey form, improving user convenience.

[0319] For example, when a user is asked "How are you feeling today?", the server predicts an answer by referring to past data on how users have answered similar questions in the past, while also considering their current emotional state (for example, the emotion "happy" extracted from a text message). The terminal displays the prediction result for the user to review, allowing them to complete the answer with minimal effort.

[0320] Furthermore, user feedback is sent to the server and used to improve the sentiment engine and machine learning models. This process allows the system to continuously learn, enabling it to generate more accurate and personalized responses.

[0321] The following describes the processing flow.

[0322] Step 1:

[0323] The server collects users' past survey responses and emotional data and stores it in a database. The emotional data includes emotional states extracted from users' voice and text.

[0324] Step 2:

[0325] The server uses natural language processing techniques to preprocess the stored data. Specifically, it normalizes the data and performs sentiment analysis as needed to transform it into a format suitable for machine learning models.

[0326] Step 3:

[0327] The server uses pre-processed data to train a machine learning model. This model learns past question and answer patterns, as well as the user's emotional state.

[0328] Step 4:

[0329] When a user receives a new survey, the device sends the question and current sentiment data to the server. The sentiment data reflects the user's real-time emotional state.

[0330] Step 5:

[0331] The server uses a trained model to predict the best answer to an incoming question. This prediction is influenced by the user's past response history and current sentiment state.

[0332] Step 6:

[0333] The server sends the predicted answer to the device. The device then automatically fills in this answer into the survey form and makes it available for the user to review.

[0334] Step 7:

[0335] Users will have the opportunity to review the answers entered by their device and correct them if necessary.

[0336] Step 8:

[0337] The device receives feedback from the user and sends it to the server. The server uses this feedback to further improve the accuracy of the emotion engine and machine learning models.

[0338] (Example 2)

[0339] 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".

[0340] The present invention aims to provide a system that reduces the burden on users when answering questionnaires and enables more accurate and personalized response predictions based on individual emotional states. The goal is to improve the accuracy of questionnaires and user satisfaction.

[0341] 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.

[0342] In this invention, the server includes means for collecting past response information and sentiment information of users; means for performing natural language processing on the response information and sentiment information to preprocess the data; and means for training a machine learning model using the preprocessed data to learn sentiment patterns. This enables accurate and personalized response predictions that take into account the user's emotional state.

[0343] "User" refers to an individual or legal entity that uses the system to answer a survey.

[0344] "Response information" refers to information and data that users have selected or entered in past surveys.

[0345] "Emotional information" refers to data that indicates a user's psychological or emotional state, extracted from their voice or text.

[0346] "Natural language processing" refers to the technologies and methods that enable computers to understand, interpret, and generate human language.

[0347] "Data preprocessing" refers to the process of shaping or transforming data before performing analysis or model training.

[0348] A "machine learning model" refers to an algorithm or structure used to learn patterns from data and perform predictions or classifications.

[0349] "Emotional patterns" refer to a series of tendencies and characteristics extracted based on the user's emotional information.

[0350] "Prediction" refers to the process by which a machine learning model uses training data to infer or estimate results for new data.

[0351] An "interface" refers to the screen or control panel that allows a user to interact with a system.

[0352] "Feedback" refers to information such as evaluations, opinions, and suggestions for improvement that users provide to the system.

[0353] "Accuracy" is an indicator that shows how accurately a system can respond to the user's intentions and requests.

[0354] This system provides a function that optimizes user survey responses through collaboration between the server and terminals. It primarily utilizes user sentiment information to achieve personalized response predictions.

[0355] server

[0356] The server collects users' past survey responses and sentiment information and stores it in a database. Sentiment information is extracted using speech recognition software and text analysis tools. Specifically, a general speech analysis service is used for speech recognition, and Python's natural language processing libraries (e.g., NLTK and spaCy) are used for text analysis.

[0357] The server preprocesses the collected data, performing tasks such as noise reduction and tokenization, and then formats the data for machine learning models. Machine learning is performed using platforms such as Scikit-learn and TensorFlow to train the models. These models are used to learn user sentiment patterns and predict answers to new questions.

[0358] terminal

[0359] The terminal receives predicted answers sent from the server and automatically inputs them into the user interface. This allows users to complete surveys with minimal effort. User feedback is also sent to the server via the terminal, contributing to the continuous improvement of the system.

[0360] Specific example

[0361] For example, when a user receives a survey question such as "How are you feeling today?", the server can predict the answer "I'm having fun today" based on past data and the user's current emotional state. As a result, the terminal automatically inputs this prediction, and after user confirmation, the survey is completed easily.

[0362] Example of a prompt

[0363] Based on the user's past survey responses and sentiment data, and taking into account their current emotional state, generate a predictive answer to the following question: 'How are you feeling today?'

[0364] This process allows the system to automatically generate more appropriate responses for each user, improving the accuracy of the survey and user satisfaction.

[0365] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0366] Step 1:

[0367] The server collects users' past survey responses and sentiment information and stores it in a database. Specifically, the server uses speech recognition software to extract text from audio data and a text analysis tool to extract sentiment information. It receives past audio and text data as input and stores the response information and sentiment information in the database as output.

[0368] Step 2:

[0369] The server preprocesses the collected data using natural language processing techniques. It performs noise reduction and tokenization on the data, converting the text data into a parseable format. It receives collected response information and sentiment information as input and generates preprocessed data as output. Specifically, it uses Python's NLTK to format the text data.

[0370] Step 3:

[0371] The server trains a machine learning model using preprocessed data. It learns patterns based on the features of the input data and builds a model. It receives preprocessed data as input and generates a trained machine learning model as output. This process utilizes Scikit-learn and TensorFlow to iteratively adjust the model's parameters.

[0372] Step 4:

[0373] The server uses a trained model to predict the best answer to a new survey question. It inputs current sentiment information into the model and calculates the best answer. It accepts a new question and current sentiment information as input and provides a predicted answer as output. Specifically, it uses prompt statements to supply the model with the question text and related data, taking sentiment states into account.

[0374] Step 5:

[0375] The terminal automatically populates the survey form with predicted answers sent from the server and allows the user to confirm them. It receives the predicted results from the server as input and displays the automatically populated state on the user interface as output. The user can then review and modify the data. Specifically, the terminal interface reflects the predicted answers and presents them to the user as editable fields.

[0376] Step 6:

[0377] The user sends feedback to the server via their device. The server receives the user feedback after review as input and provides data as output that is used for the system's continuous learning. The server, upon receiving the feedback, applies it to improve the sentiment engine and machine learning model to improve the accuracy of future predictions. Specifically, a process is executed to retrain the model using the new feedback data as input.

[0378] (Application Example 2)

[0379] 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."

[0380] In today's world, there is a need to optimize users' purchasing intentions and provide highly accurate product recommendations. However, conventional systems have been unable to reflect the individual emotional state of users in their recommendations, resulting in a decrease in the usefulness of the suggestions. To address this situation, there is a need to provide a system that considers past response information and current emotional state to provide personalized recommendations.

[0381] 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.

[0382] In this invention, the server includes means for collecting past response information and emotional data of the user, means for performing natural language processing on the response information and emotional data to preprocess the data, and means for training a machine learning model using the preprocessed data. This enables personalized suggestions based on the user's emotional tendencies.

[0383] A "user" is an individual or legal entity that uses the system or service.

[0384] "Response information" refers to the content of responses that users have given to past surveys or questions.

[0385] "Emotional data" refers to data that indicates the user's emotional state, and is extracted from voice and text input.

[0386] "Natural language processing" is a technology that enables computers to understand and process human language.

[0387] "Data preprocessing" is the process of preparing data to be suitable for machine learning algorithms.

[0388] A "machine learning model" is a statistical model that learns patterns from data to perform predictions and classifications.

[0389] "Training methods" refer to the process of providing data to a machine learning model and allowing it to learn patterns and relationships.

[0390] "Emotional state" refers to information that indicates the user's current psychological or emotional condition.

[0391] "Personalized recommendations" refer to recommendations for products and services tailored to the specific needs and emotions of the user.

[0392] A "connection device" is a device used by users to access the system.

[0393] The system for implementing the present invention includes a program that collects the user's past response information and emotional data, and utilizes natural language processing and machine learning techniques. The server runs on a platform such as Python or TensorFlow and extracts emotional data from the user's text and voice data. The extracted data is stored in a database and then used to train a machine learning model after preprocessing.

[0394] The server preprocesses data using a natural language processing toolkit (e.g., NLTK) and optimizes machine learning models. Based on the user's response patterns and emotional tendencies, it can generate personalized suggestions. These suggestions, which take into account the user's current emotional state, are provided on the user's device.

[0395] When a user accesses the system using a connected device such as a smartphone, the device presents the user with predicted responses and suggestions. The user's feedback is then sent back to the server and used to improve the model's performance. This feedback process allows the system to continuously improve the accuracy of the suggestions it provides.

[0396] For example, if a user enters "I'm feeling good today," the server will recommend appropriate products from those that the user has considered purchasing in similar emotional states in the past. An example of a prompt for the generative AI model used in this process could be in the form of, "What products would you recommend if the user's emotional data indicates they are feeling good?"

[0397] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0398] Step 1:

[0399] The server receives the user's past survey responses and emotional data collected from voice, text, etc. The input data includes the user's response history and information indicating their emotional state. The server stores this data in a database for later analysis. During this process, data cleansing is performed to remove inaccurate or invalid data.

[0400] Step 2:

[0401] The server preprocesses the collected response information and emotional data using a natural language processing toolkit (such as NLTK). This step converts the input data into a format that can be easily handled by machine learning models. Specifically, it performs tasks such as text tokenization, removal of special characters, and extraction of sentiment words to extract meaningful features.

[0402] Step 3:

[0403] The server trains a machine learning model using pre-processed data. The input data consists of past responses and emotional patterns extracted as features. The output is a model that generates responses and product suggestions based on the user's emotional state. This process utilizes libraries such as TensorFlow to train a neural network.

[0404] Step 4:

[0405] The server receives the user's current questions and emotional state as input and uses a trained model to predict the best response and product suggestion. The input data includes real-time user emotional data. As output, it generates personalized suggestions and responses and sends them to the terminal.

[0406] Step 5:

[0407] The terminal presents the user with predicted responses and suggestions received from the server. In this step, the suggestions are visually displayed through the user interface for the user to review. The user can then make decisions, select, or change based on the suggestions.

[0408] Step 6:

[0409] The device receives user feedback and sends it to the server. The input data consists of user opinions regarding the usefulness and satisfaction level of the suggestions. The server uses this feedback to retrain the machine learning model to improve its performance. This makes it possible to further improve the accuracy of future suggestions.

[0410] 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.

[0411] 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.

[0412] 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.

[0413] [Third Embodiment]

[0414] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0415] 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.

[0416] 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).

[0417] 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.

[0418] 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.

[0419] 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).

[0420] 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.

[0421] 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.

[0422] 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.

[0423] 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.

[0424] 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.

[0425] 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".

[0426] This invention provides an AI agent system that streamlines the process of answering questionnaires. Based on the user's past questionnaire responses, this system utilizes natural language processing and machine learning techniques to automatically generate optimal answers to new questions.

[0427] The server first collects past survey response data from users and stores it in a database. This data includes the questions and the users' actual responses. The server uses this data to perform natural language processing, analyzing and preprocessing it. The processed data is then used to train a machine learning model, which learns question and answer patterns.

[0428] When a user receives a new survey, the device sends the question information to the server. The server uses a trained model to predict the best answer to the question and sends the result back to the device. The device automatically enters the returned answer into the survey form, significantly reducing the user's response time.

[0429] For example, suppose a user is asked, "What did you eat for dinner last night?" Based on past data, the server predicts the answer "pasta" and sends it to the device. The device automatically enters this prediction into the survey's answer field, allowing the user to answer the survey with minimal effort.

[0430] This system also includes a feature to improve model accuracy based on user feedback. The server analyzes the feedback information and retrains the model as needed, enabling the generation of more accurate responses. As a result, users can enjoy the benefit of answering surveys more efficiently and earning more points.

[0431] The following describes the processing flow.

[0432] Step 1:

[0433] The server collects users' past survey responses and stores them in a database. This data includes survey questions and their answers.

[0434] Step 2:

[0435] The server preprocesses the stored data using natural language processing techniques. Specifically, it performs text normalization, removes unnecessary information, and tokenizes the data.

[0436] Step 3:

[0437] The server trains a machine learning model using pre-processed data. Through this training, the model learns question-and-answer patterns.

[0438] Step 4:

[0439] When a user receives a new survey, the device sends the question content to the server.

[0440] Step 5:

[0441] The server uses a trained model to predict the best answer to the question it receives.

[0442] Step 6:

[0443] The server sends the predicted answer to the device. The device then automatically fills in this answer into the survey form.

[0444] Step 7:

[0445] The user reviews the answers automatically entered by the device and makes adjustments as needed.

[0446] Step 8:

[0447] The device sends user feedback to the server. The server analyzes this feedback and uses it to improve the machine learning model.

[0448] (Example 1)

[0449] 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."

[0450] Current survey systems present challenges because users must individually answer every question each time, which is time-consuming and labor-intensive. Furthermore, it is difficult for users to effectively reuse past answers to answer new questions, highlighting the need for greater efficiency. Additionally, mechanisms for continuous system improvement based on user feedback are insufficient.

[0451] 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.

[0452] In this invention, the server includes means for accumulating past user response data, means for performing language processing based on the response data to organize the information, and means for training a learning model using the organized information. This enables users to obtain optimal answers to new questions efficiently and quickly. Furthermore, by continuously improving the system based on user feedback, answers with higher accuracy can be provided.

[0453] "User" refers to an individual or organization that provides answers to questions using the survey system.

[0454] "Response data" refers to a set of information containing responses to questions previously asked by the user.

[0455] "Language processing" refers to the technology of analyzing and transforming text data expressed in natural language, and preparing it in a format that can be used by machine learning models.

[0456] "Organizing information" refers to the process of analyzing and classifying collected data to reconstruct it into a format necessary for learning.

[0457] A "learning model" refers to an algorithm that uses past data to discover new information patterns and predict answers to future questions.

[0458] "Accumulating" refers to the continuous collection and storage of data, specifically the act of saving it to a database for later use as needed.

[0459] "Feedback" refers to evaluations and opinions provided by users regarding the system's performance and the responses it generates.

[0460] This invention is an AI agent system designed to enable users to efficiently answer questionnaires. The specific method for implementing this system is described below.

[0461] The server first collects users' past response data and stores it in a database. General-purpose database software is used for data management. This data includes previously answered questions and their responses. To analyze the text data, the server employs "spaCy" as natural language processing software to perform preprocessing such as data cleaning, tokenization, and tagging.

[0462] The preprocessed data is used to train a model using the machine learning framework "TensorFlow." This model has the ability to learn past question and answer patterns and predict the best answer to new questions. For example, data where a user previously answered "apple" to the question "What is your favorite fruit?" is analyzed by the model, and it generates an appropriate answer to a new question with a similar pattern.

[0463] When a new survey question arrives, the user's device sends it to the server. The server then uses a trained model to predict the best answer to the question and sends the result back to the device. The device automatically enters the received answer into the survey form. This allows users to answer surveys quickly, saving them time and effort.

[0464] This system also includes a feature that allows the server to improve the accuracy of the model based on user feedback. By analyzing the feedback information, the server retrains the model as needed to improve the accuracy of the answers. In this process, a generative AI model is used, and prompts such as "Predict the best answer to the survey question 'What is your favorite fruit?'" are used.

[0465] In this way, users can efficiently answer surveys and enjoy more benefits through the system.

[0466] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0467] Step 1:

[0468] The server collects the user's past response data. The input consists of the user's previous survey questions and their responses. The server stores this data in a database for later analysis. Specifically, a database management system is used for data storage and organization. The output is the stored response data in an accessible format.

[0469] Step 2:

[0470] The server preprocesses the response data collected from the database using natural language processing (NLP). The input consists of stored user questions and answers. The server uses "spaCy" to clean up, tokenize, and tag the text. Specifically, this involves removing unnecessary symbols and spaces, splitting words, and tagging parts of speech. The output is the refined text data.

[0471] Step 3:

[0472] The server trains a machine learning model using pre-processed data. The input is pre-processed question and answer data. The server uses TensorFlow to learn patterns based on the data. Specifically, the model associates questions with past answers and learns patterns for what answers to predict for new questions. The output is the trained model.

[0473] Step 4:

[0474] The terminal receives a new survey question from the user and sends it to the server. The input is the newly provided survey question. The terminal simply sends this to the server. The output is the transmission of the accurate question information to the server.

[0475] Step 5:

[0476] The server uses a trained model to predict the best answer to a new question. The input is new question information received from the terminal. The server uses the model to generate the best answer based on past data. Specifically, it applies the received question to the model and estimates the most likely answer. The output is the predicted answer.

[0477] Step 6:

[0478] The terminal receives predicted responses sent from the server and automatically enters them into the survey form. The input is the predicted response data from the server. The terminal then reflects the received responses in the user interface. The output is the responses automatically entered into the form.

[0479] Step 7:

[0480] Users provide feedback on the system-generated answers. The input consists of the user's evaluation and opinion. The user then sends this feedback to the server. The output is the feedback information received by the server.

[0481] Step 8:

[0482] The server analyzes user feedback and retrains the machine learning model as needed. The input is feedback information based on user evaluations. The server uses this feedback to improve the model's accuracy. Specifically, it identifies incorrect response patterns and adjusts the model. The output is the improved trained model.

[0483] (Application Example 1)

[0484] 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."

[0485] In home appliances, a challenge is to improve the convenience of daily life by providing quick and accurate optimal suggestions based on the user's past instructions. In particular, improving the accuracy of suggestions through relearning using feedback is required.

[0486] 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.

[0487] In this invention, the server includes means for collecting past instruction information of the user, means for performing natural language processing on the instruction information to preprocess the data, and means for automatically generating optimal suggestions for home appliances based on the user's preferences. This makes it possible to improve the efficiency of operating and setting up home appliances.

[0488] "Instruction information" refers to data that records past requests and wishes made by users regarding home appliances.

[0489] "Natural language processing" is a technology that uses computers to process, understand, and generate human language.

[0490] "Household appliances" refer to electrical and electronic devices used in the home, and are devices that support various tasks in daily life.

[0491] A "suggestion" is advice regarding optimal actions and settings generated based on the user's instructions and preferences.

[0492] "Retraining" is an additional training process performed to improve the accuracy of a machine learning model based on feedback from users.

[0493] The system for implementing this invention is designed to provide convenient operation and setting support for household appliances. The system collects information on the user's past instructions and generates optimal suggestions based on that information, thereby streamlining the user's daily life.

[0494] The server stores the user's past instruction information in a database and analyzes it using natural language processing techniques. Specifically, it uses Python and natural language processing libraries such as NLTK and spaCy to preprocess the data. The processed data is then trained using machine learning models, such as TensorFlow or PyTorch. During this training process, the model is built to automatically generate optimal suggestions for the current situation, taking into account the instruction information and the user's preferences.

[0495] Terminals that control home appliances receive suggestions sent from a server and automatically input them into the user's interface. For example, based on past data, a suggestion such as "play relaxing music at night" might be made. This allows users to save time on operating the devices and enjoy a more comfortable life.

[0496] Furthermore, based on user feedback, the server retrains the model to improve the system's accuracy. This feedback allows the model to learn a wider variety of patterns, enabling more personalized suggestions.

[0497] For example, if you instruct a household robot to "clean up after dinner," the robot will suggest and execute an efficient cleaning method based on reliable data. An example of a prompt to the generative AI model would be, "Please tell me what I should do to create a comfortable environment in the living room."

[0498] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0499] Step 1:

[0500] The server collects past user instruction information and stores it in a database. The input is user instruction data, and the output is structured database input. This process allows the server to secure the basic data necessary for subsequent analysis.

[0501] Step 2:

[0502] The server uses natural language processing on the collected instruction information as input to perform data preprocessing. This involves data cleansing and tokenization, generating data in a format suitable for input to the model as output. Specifically, this is done using NLTK or spaCy.

[0503] Step 3:

[0504] The server trains a machine learning model using pre-processed data. The input is the transformed data, and the output is a trained model that links instruction information with preferences. TensorFlow or PyTorch is used to learn patterns based on user preferences.

[0505] Step 4:

[0506] The server uses a pre-trained model to generate optimal suggestions for new user requests. The model takes prompts as input and outputs suggested content. A concrete example would be a suggestion like "Create a list of recommended music for after dinner."

[0507] Step 5:

[0508] The terminal receives the generated suggestions and automatically inputs the output into the interface of the home appliance. The input is suggestion data from the server, and the output is the configuration changes or tasks performed on the appliance. This automation reduces user effort and improves convenience.

[0509] Step 6:

[0510] The user provides feedback on the proposal to the server. The input is the user's feedback information, and based on this, the server obtains data as output to retrain the model. Feedback is a crucial element for further improving the accuracy of the proposal.

[0511] 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.

[0512] This invention provides a system that further personalizes and improves the accuracy of survey response predictions by incorporating an emotion engine into an AI agent system. This system combines the user's past survey response data and emotion data, and utilizes natural language processing and machine learning techniques to automatically generate the optimal response.

[0513] The server collects users' past survey responses and related sentiment data and stores them in a database. Sentiment data is extracted from the user's voice and text input, and the sentiment engine identifies their emotional state. The server preprocesses the collected data using natural language processing techniques and analyzes the data. Subsequently, it trains a machine learning model to learn past sentiment patterns.

[0514] The device interacts with the server when the user receives a survey. The server uses a trained model to predict the best answer to a new question, and this prediction reflects the user's emotional state. The device automatically fills in the predicted answer into the survey form, improving user convenience.

[0515] For example, when a user is asked "How are you feeling today?", the server predicts an answer by referring to past data on how users have answered similar questions in the past, while also considering their current emotional state (for example, the emotion "happy" extracted from a text message). The terminal displays the prediction result for the user to review, allowing them to complete the answer with minimal effort.

[0516] Furthermore, user feedback is sent to the server and used to improve the sentiment engine and machine learning models. This process allows the system to continuously learn, enabling it to generate more accurate and personalized responses.

[0517] The following describes the processing flow.

[0518] Step 1:

[0519] The server collects users' past survey responses and emotional data and stores it in a database. The emotional data includes emotional states extracted from users' voice and text.

[0520] Step 2:

[0521] The server uses natural language processing techniques to preprocess the stored data. Specifically, it normalizes the data and performs sentiment analysis as needed to transform it into a format suitable for machine learning models.

[0522] Step 3:

[0523] The server uses pre-processed data to train a machine learning model. This model learns past question and answer patterns, as well as the user's emotional state.

[0524] Step 4:

[0525] When a user receives a new survey, the device sends the question and current sentiment data to the server. The sentiment data reflects the user's real-time emotional state.

[0526] Step 5:

[0527] The server uses a trained model to predict the best answer to an incoming question. This prediction is influenced by the user's past response history and current sentiment state.

[0528] Step 6:

[0529] The server sends the predicted answer to the device. The device then automatically fills in this answer into the survey form and makes it available for the user to review.

[0530] Step 7:

[0531] Users will have the opportunity to review the answers entered by their device and correct them if necessary.

[0532] Step 8:

[0533] The device receives feedback from the user and sends it to the server. The server uses this feedback to further improve the accuracy of the emotion engine and machine learning models.

[0534] (Example 2)

[0535] 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."

[0536] The present invention aims to provide a system that reduces the burden on users when answering questionnaires and enables more accurate and personalized response predictions based on individual emotional states. The goal is to improve the accuracy of questionnaires and user satisfaction.

[0537] 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.

[0538] In this invention, the server includes means for collecting past response information and sentiment information of users; means for performing natural language processing on the response information and sentiment information to preprocess the data; and means for training a machine learning model using the preprocessed data to learn sentiment patterns. This enables accurate and personalized response predictions that take into account the user's emotional state.

[0539] "User" refers to an individual or legal entity that uses the system to answer a survey.

[0540] "Response information" refers to information and data that users have selected or entered in past surveys.

[0541] "Emotional information" refers to data that indicates a user's psychological or emotional state, extracted from their voice or text.

[0542] "Natural language processing" refers to the technologies and methods that enable computers to understand, interpret, and generate human language.

[0543] "Data preprocessing" refers to the process of shaping or transforming data before performing analysis or model training.

[0544] A "machine learning model" refers to an algorithm or structure used to learn patterns from data and perform predictions or classifications.

[0545] "Emotional patterns" refer to a series of tendencies and characteristics extracted based on the user's emotional information.

[0546] "Prediction" refers to the process by which a machine learning model uses training data to infer or estimate results for new data.

[0547] An "interface" refers to the screen or control panel that allows a user to interact with a system.

[0548] "Feedback" refers to information such as evaluations, opinions, and suggestions for improvement that users provide to the system.

[0549] "Accuracy" is an indicator that shows how accurately a system can respond to the user's intentions and requests.

[0550] This system provides a function that optimizes user survey responses through collaboration between the server and terminals. It primarily utilizes user sentiment information to achieve personalized response predictions.

[0551] server

[0552] The server collects users' past survey responses and sentiment information and stores it in a database. Sentiment information is extracted using speech recognition software and text analysis tools. Specifically, a general speech analysis service is used for speech recognition, and Python's natural language processing libraries (e.g., NLTK and spaCy) are used for text analysis.

[0553] The server preprocesses the collected data, performing tasks such as noise reduction and tokenization, and then formats the data for machine learning models. Machine learning is performed using platforms such as Scikit-learn and TensorFlow to train the models. These models are used to learn user sentiment patterns and predict answers to new questions.

[0554] terminal

[0555] The terminal receives predicted answers sent from the server and automatically inputs them into the user interface. This allows users to complete surveys with minimal effort. User feedback is also sent to the server via the terminal, contributing to the continuous improvement of the system.

[0556] Specific example

[0557] For example, when a user receives a survey question such as "How are you feeling today?", the server can predict the answer "I'm having fun today" based on past data and the user's current emotional state. As a result, the terminal automatically inputs this prediction, and after user confirmation, the survey is completed easily.

[0558] Example of a prompt

[0559] Based on the user's past survey responses and sentiment data, and taking into account their current emotional state, generate a predictive answer to the following question: 'How are you feeling today?'

[0560] This process allows the system to automatically generate more appropriate responses for each user, improving the accuracy of the survey and user satisfaction.

[0561] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0562] Step 1:

[0563] The server collects users' past survey responses and sentiment information and stores it in a database. Specifically, the server uses speech recognition software to extract text from audio data and a text analysis tool to extract sentiment information. It receives past audio and text data as input and stores the response information and sentiment information in the database as output.

[0564] Step 2:

[0565] The server preprocesses the collected data using natural language processing techniques. It performs noise reduction and tokenization on the data, converting the text data into a parseable format. It receives collected response information and sentiment information as input and generates preprocessed data as output. Specifically, it uses Python's NLTK to format the text data.

[0566] Step 3:

[0567] The server trains a machine learning model using preprocessed data. It learns patterns based on the features of the input data and builds a model. It receives preprocessed data as input and generates a trained machine learning model as output. This process utilizes Scikit-learn and TensorFlow to iteratively adjust the model's parameters.

[0568] Step 4:

[0569] The server uses a trained model to predict the best answer to a new survey question. It inputs current sentiment information into the model and calculates the best answer. It accepts a new question and current sentiment information as input and provides a predicted answer as output. Specifically, it uses prompt statements to supply the model with the question text and related data, taking sentiment states into account.

[0570] Step 5:

[0571] The terminal automatically populates the survey form with predicted answers sent from the server and allows the user to confirm them. It receives the predicted results from the server as input and displays the automatically populated state on the user interface as output. The user can then review and modify the data. Specifically, the terminal interface reflects the predicted answers and presents them to the user as editable fields.

[0572] Step 6:

[0573] The user sends feedback to the server via their device. The server receives the user feedback after review as input and provides data as output that is used for the system's continuous learning. The server, upon receiving the feedback, applies it to improve the sentiment engine and machine learning model to improve the accuracy of future predictions. Specifically, a process is executed to retrain the model using the new feedback data as input.

[0574] (Application Example 2)

[0575] 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."

[0576] In today's world, there is a need to optimize users' purchasing intentions and provide highly accurate product recommendations. However, conventional systems have been unable to reflect the individual emotional state of users in their recommendations, resulting in a decrease in the usefulness of the suggestions. To address this situation, there is a need to provide a system that considers past response information and current emotional state to provide personalized recommendations.

[0577] 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.

[0578] In this invention, the server includes means for collecting past response information and emotional data of the user, means for performing natural language processing on the response information and emotional data to preprocess the data, and means for training a machine learning model using the preprocessed data. This enables personalized suggestions based on the user's emotional tendencies.

[0579] A "user" is an individual or legal entity that uses the system or service.

[0580] "Response information" refers to the content of responses that users have given to past surveys or questions.

[0581] "Emotional data" refers to data that indicates the user's emotional state, and is extracted from voice and text input.

[0582] "Natural language processing" is a technology that enables computers to understand and process human language.

[0583] "Data preprocessing" is the process of preparing data to be suitable for machine learning algorithms.

[0584] A "machine learning model" is a statistical model that learns patterns from data to perform predictions and classifications.

[0585] "Training methods" refer to the process of providing data to a machine learning model and allowing it to learn patterns and relationships.

[0586] "Emotional state" refers to information that indicates the user's current psychological or emotional condition.

[0587] "Personalized recommendations" refer to recommendations for products and services tailored to the specific needs and emotions of the user.

[0588] A "connection device" is a device used by users to access the system.

[0589] The system for implementing the present invention includes a program that collects the user's past response information and emotional data, and utilizes natural language processing and machine learning techniques. The server runs on a platform such as Python or TensorFlow and extracts emotional data from the user's text and voice data. The extracted data is stored in a database and then used to train a machine learning model after preprocessing.

[0590] The server preprocesses data using a natural language processing toolkit (e.g., NLTK) and optimizes machine learning models. Based on the user's response patterns and emotional tendencies, it can generate personalized suggestions. These suggestions, which take into account the user's current emotional state, are provided on the user's device.

[0591] When a user accesses the system using a connected device such as a smartphone, the device presents the user with predicted responses and suggestions. The user's feedback is then sent back to the server and used to improve the model's performance. This feedback process allows the system to continuously improve the accuracy of the suggestions it provides.

[0592] For example, if a user enters "I'm feeling good today," the server will recommend appropriate products from those that the user has considered purchasing in similar emotional states in the past. An example of a prompt for the generative AI model used in this process could be in the form of, "What products would you recommend if the user's emotional data indicates they are feeling good?"

[0593] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0594] Step 1:

[0595] The server receives the user's past survey responses and emotional data collected from voice, text, etc. The input data includes the user's response history and information indicating their emotional state. The server stores this data in a database for later analysis. During this process, data cleansing is performed to remove inaccurate or invalid data.

[0596] Step 2:

[0597] The server preprocesses the collected response information and emotional data using a natural language processing toolkit (such as NLTK). This step converts the input data into a format that can be easily handled by machine learning models. Specifically, it performs tasks such as text tokenization, removal of special characters, and extraction of sentiment words to extract meaningful features.

[0598] Step 3:

[0599] The server trains a machine learning model using pre-processed data. The input data consists of past responses and emotional patterns extracted as features. The output is a model that generates responses and product suggestions based on the user's emotional state. This process utilizes libraries such as TensorFlow to train a neural network.

[0600] Step 4:

[0601] The server receives the user's current questions and emotional state as input and uses a trained model to predict the best response and product suggestion. The input data includes real-time user emotional data. As output, it generates personalized suggestions and responses and sends them to the terminal.

[0602] Step 5:

[0603] The terminal presents the user with predicted responses and suggestions received from the server. In this step, the suggestions are visually displayed through the user interface for the user to review. The user can then make decisions, select, or change based on the suggestions.

[0604] Step 6:

[0605] The device receives user feedback and sends it to the server. The input data consists of user opinions regarding the usefulness and satisfaction level of the suggestions. The server uses this feedback to retrain the machine learning model to improve its performance. This makes it possible to further improve the accuracy of future suggestions.

[0606] 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.

[0607] 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.

[0608] 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.

[0609] [Fourth Embodiment]

[0610] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0611] 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.

[0612] 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).

[0613] 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.

[0614] 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.

[0615] 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).

[0616] 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.

[0617] 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.

[0618] 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.

[0619] 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.

[0620] 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.

[0621] 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.

[0622] 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".

[0623] This invention provides an AI agent system that streamlines the process of answering questionnaires. Based on the user's past questionnaire responses, this system utilizes natural language processing and machine learning techniques to automatically generate optimal answers to new questions.

[0624] The server first collects past survey response data from users and stores it in a database. This data includes the questions and the users' actual responses. The server uses this data to perform natural language processing, analyzing and preprocessing it. The processed data is then used to train a machine learning model, which learns question and answer patterns.

[0625] When a user receives a new survey, the device sends the question information to the server. The server uses a trained model to predict the best answer to the question and sends the result back to the device. The device automatically enters the returned answer into the survey form, significantly reducing the user's response time.

[0626] For example, suppose a user is asked, "What did you eat for dinner last night?" Based on past data, the server predicts the answer "pasta" and sends it to the device. The device automatically enters this prediction into the survey's answer field, allowing the user to answer the survey with minimal effort.

[0627] This system also includes a feature to improve model accuracy based on user feedback. The server analyzes the feedback information and retrains the model as needed, enabling the generation of more accurate responses. As a result, users can enjoy the benefit of answering surveys more efficiently and earning more points.

[0628] The following describes the processing flow.

[0629] Step 1:

[0630] The server collects users' past survey responses and stores them in a database. This data includes survey questions and their answers.

[0631] Step 2:

[0632] The server preprocesses the stored data using natural language processing techniques. Specifically, it performs text normalization, removes unnecessary information, and tokenizes the data.

[0633] Step 3:

[0634] The server trains a machine learning model using pre-processed data. Through this training, the model learns question-and-answer patterns.

[0635] Step 4:

[0636] When a user receives a new survey, the device sends the question content to the server.

[0637] Step 5:

[0638] The server uses a trained model to predict the best answer to the question it receives.

[0639] Step 6:

[0640] The server sends the predicted answer to the device. The device then automatically fills in this answer into the survey form.

[0641] Step 7:

[0642] The user reviews the answers automatically entered by the device and makes adjustments as needed.

[0643] Step 8:

[0644] The device sends user feedback to the server. The server analyzes this feedback and uses it to improve the machine learning model.

[0645] (Example 1)

[0646] 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".

[0647] Current survey systems present challenges because users must individually answer every question each time, which is time-consuming and labor-intensive. Furthermore, it is difficult for users to effectively reuse past answers to answer new questions, highlighting the need for greater efficiency. Additionally, mechanisms for continuous system improvement based on user feedback are insufficient.

[0648] 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.

[0649] In this invention, the server includes means for accumulating past user response data, means for performing language processing based on the response data to organize the information, and means for training a learning model using the organized information. This enables users to obtain optimal answers to new questions efficiently and quickly. Furthermore, by continuously improving the system based on user feedback, answers with higher accuracy can be provided.

[0650] "User" refers to an individual or organization that provides answers to questions using the survey system.

[0651] "Response data" refers to a set of information containing responses to questions previously asked by the user.

[0652] "Language processing" refers to the technology of analyzing and transforming text data expressed in natural language, and preparing it in a format that can be used by machine learning models.

[0653] "Organizing information" refers to the process of analyzing and classifying collected data to reconstruct it into a format necessary for learning.

[0654] A "learning model" refers to an algorithm that uses past data to discover new information patterns and predict answers to future questions.

[0655] "Accumulating" refers to the continuous collection and storage of data, specifically the act of saving it to a database for later use as needed.

[0656] "Feedback" refers to evaluations and opinions provided by users regarding the system's performance and the responses it generates.

[0657] This invention is an AI agent system designed to enable users to efficiently answer questionnaires. The specific method for implementing this system is described below.

[0658] The server first collects users' past response data and stores it in a database. General-purpose database software is used for data management. This data includes previously answered questions and their responses. To analyze the text data, the server employs "spaCy" as natural language processing software to perform preprocessing such as data cleaning, tokenization, and tagging.

[0659] The preprocessed data is used to train a model using the machine learning framework "TensorFlow." This model has the ability to learn past question and answer patterns and predict the best answer to new questions. For example, data where a user previously answered "apple" to the question "What is your favorite fruit?" is analyzed by the model, and it generates an appropriate answer to a new question with a similar pattern.

[0660] When a new survey question arrives, the user's device sends it to the server. The server then uses a trained model to predict the best answer to the question and sends the result back to the device. The device automatically enters the received answer into the survey form. This allows users to answer surveys quickly, saving them time and effort.

[0661] This system also includes a feature that allows the server to improve the accuracy of the model based on user feedback. By analyzing the feedback information, the server retrains the model as needed to improve the accuracy of the answers. In this process, a generative AI model is used, and prompts such as "Predict the best answer to the survey question 'What is your favorite fruit?'" are used.

[0662] In this way, users can efficiently answer surveys and enjoy more benefits through the system.

[0663] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0664] Step 1:

[0665] The server collects the user's past response data. The input consists of the user's previous survey questions and their responses. The server stores this data in a database for later analysis. Specifically, a database management system is used for data storage and organization. The output is the stored response data in an accessible format.

[0666] Step 2:

[0667] The server preprocesses the response data collected from the database using natural language processing (NLP). The input consists of stored user questions and answers. The server uses "spaCy" to clean up, tokenize, and tag the text. Specifically, this involves removing unnecessary symbols and spaces, splitting words, and tagging parts of speech. The output is the refined text data.

[0668] Step 3:

[0669] The server trains a machine learning model using pre-processed data. The input is pre-processed question and answer data. The server uses TensorFlow to learn patterns based on the data. Specifically, the model associates questions with past answers and learns patterns for what answers to predict for new questions. The output is the trained model.

[0670] Step 4:

[0671] The terminal receives a new survey question from the user and sends it to the server. The input is the newly provided survey question. The terminal simply sends this to the server. The output is the transmission of the accurate question information to the server.

[0672] Step 5:

[0673] The server uses a trained model to predict the best answer to a new question. The input is new question information received from the terminal. The server uses the model to generate the best answer based on past data. Specifically, it applies the received question to the model and estimates the most likely answer. The output is the predicted answer.

[0674] Step 6:

[0675] The terminal receives predicted responses sent from the server and automatically enters them into the survey form. The input is the predicted response data from the server. The terminal then reflects the received responses in the user interface. The output is the responses automatically entered into the form.

[0676] Step 7:

[0677] Users provide feedback on the system-generated answers. The input consists of the user's evaluation and opinion. The user then sends this feedback to the server. The output is the feedback information received by the server.

[0678] Step 8:

[0679] The server analyzes user feedback and retrains the machine learning model as needed. The input is feedback information based on user evaluations. The server uses this feedback to improve the model's accuracy. Specifically, it identifies incorrect response patterns and adjusts the model. The output is the improved trained model.

[0680] (Application Example 1)

[0681] 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".

[0682] In home appliances, a challenge is to improve the convenience of daily life by providing quick and accurate optimal suggestions based on the user's past instructions. In particular, improving the accuracy of suggestions through relearning using feedback is required.

[0683] 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.

[0684] In this invention, the server includes means for collecting past instruction information of the user, means for performing natural language processing on the instruction information to preprocess the data, and means for automatically generating optimal suggestions for home appliances based on the user's preferences. This makes it possible to improve the efficiency of operating and setting up home appliances.

[0685] "Instruction information" refers to data that records past requests and wishes made by users regarding home appliances.

[0686] "Natural language processing" is a technology that uses computers to process, understand, and generate human language.

[0687] "Household appliances" refer to electrical and electronic devices used in the home, and are devices that support various tasks in daily life.

[0688] A "suggestion" is advice regarding optimal actions and settings generated based on the user's instructions and preferences.

[0689] "Retraining" is an additional training process performed to improve the accuracy of a machine learning model based on feedback from users.

[0690] The system for implementing this invention is designed to provide convenient operation and setting support for household appliances. The system collects information on the user's past instructions and generates optimal suggestions based on that information, thereby streamlining the user's daily life.

[0691] The server stores the user's past instruction information in a database and analyzes it using natural language processing techniques. Specifically, it uses Python and natural language processing libraries such as NLTK and spaCy to preprocess the data. The processed data is then trained using machine learning models, such as TensorFlow or PyTorch. During this training process, the model is built to automatically generate optimal suggestions for the current situation, taking into account the instruction information and the user's preferences.

[0692] Terminals that control home appliances receive suggestions sent from a server and automatically input them into the user's interface. For example, based on past data, a suggestion such as "play relaxing music at night" might be made. This allows users to save time on operating the devices and enjoy a more comfortable life.

[0693] Furthermore, based on user feedback, the server retrains the model to improve the system's accuracy. This feedback allows the model to learn a wider variety of patterns, enabling more personalized suggestions.

[0694] For example, if you instruct a household robot to "clean up after dinner," the robot will suggest and execute an efficient cleaning method based on reliable data. An example of a prompt to the generative AI model would be, "Please tell me what I should do to create a comfortable environment in the living room."

[0695] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0696] Step 1:

[0697] The server collects past user instruction information and stores it in a database. The input is user instruction data, and the output is structured database input. This process allows the server to secure the basic data necessary for subsequent analysis.

[0698] Step 2:

[0699] The server uses natural language processing on the collected instruction information as input to perform data preprocessing. This involves data cleansing and tokenization, generating data in a format suitable for input to the model as output. Specifically, this is done using NLTK or spaCy.

[0700] Step 3:

[0701] The server trains a machine learning model using pre-processed data. The input is the transformed data, and the output is a trained model that links instruction information with preferences. TensorFlow or PyTorch is used to learn patterns based on user preferences.

[0702] Step 4:

[0703] The server uses a pre-trained model to generate optimal suggestions for new user requests. The model takes prompts as input and outputs suggested content. A concrete example would be a suggestion like "Create a list of recommended music for after dinner."

[0704] Step 5:

[0705] The terminal receives the generated suggestions and automatically inputs the output into the interface of the home appliance. The input is suggestion data from the server, and the output is the configuration changes or tasks performed on the appliance. This automation reduces user effort and improves convenience.

[0706] Step 6:

[0707] The user provides feedback on the proposal to the server. The input is the user's feedback information, and based on this, the server obtains data as output to retrain the model. Feedback is a crucial element for further improving the accuracy of the proposal.

[0708] 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.

[0709] This invention provides a system that further personalizes and improves the accuracy of survey response predictions by incorporating an emotion engine into an AI agent system. This system combines the user's past survey response data and emotion data, and utilizes natural language processing and machine learning techniques to automatically generate the optimal response.

[0710] The server collects users' past survey responses and related sentiment data and stores them in a database. Sentiment data is extracted from the user's voice and text input, and the sentiment engine identifies their emotional state. The server preprocesses the collected data using natural language processing techniques and analyzes the data. Subsequently, it trains a machine learning model to learn past sentiment patterns.

[0711] The device interacts with the server when the user receives a survey. The server uses a trained model to predict the best answer to a new question, and this prediction reflects the user's emotional state. The device automatically fills in the predicted answer into the survey form, improving user convenience.

[0712] For example, when a user is asked "How are you feeling today?", the server predicts an answer by referring to past data on how users have answered similar questions in the past, while also considering their current emotional state (for example, the emotion "happy" extracted from a text message). The terminal displays the prediction result for the user to review, allowing them to complete the answer with minimal effort.

[0713] Furthermore, user feedback is sent to the server and used to improve the sentiment engine and machine learning models. This process allows the system to continuously learn, enabling it to generate more accurate and personalized responses.

[0714] The following describes the processing flow.

[0715] Step 1:

[0716] The server collects users' past survey responses and emotional data and stores it in a database. The emotional data includes emotional states extracted from users' voice and text.

[0717] Step 2:

[0718] The server uses natural language processing techniques to preprocess the stored data. Specifically, it normalizes the data and performs sentiment analysis as needed to transform it into a format suitable for machine learning models.

[0719] Step 3:

[0720] The server uses pre-processed data to train a machine learning model. This model learns past question and answer patterns, as well as the user's emotional state.

[0721] Step 4:

[0722] When a user receives a new survey, the device sends the question and current sentiment data to the server. The sentiment data reflects the user's real-time emotional state.

[0723] Step 5:

[0724] The server uses a trained model to predict the best answer to an incoming question. This prediction is influenced by the user's past response history and current sentiment state.

[0725] Step 6:

[0726] The server sends the predicted answer to the device. The device then automatically fills in this answer into the survey form and makes it available for the user to review.

[0727] Step 7:

[0728] Users will have the opportunity to review the answers entered by their device and correct them if necessary.

[0729] Step 8:

[0730] The device receives feedback from the user and sends it to the server. The server uses this feedback to further improve the accuracy of the emotion engine and machine learning models.

[0731] (Example 2)

[0732] 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".

[0733] The present invention aims to provide a system that reduces the burden on users when answering questionnaires and enables more accurate and personalized response predictions based on individual emotional states. The goal is to improve the accuracy of questionnaires and user satisfaction.

[0734] 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.

[0735] In this invention, the server includes means for collecting past response information and sentiment information of users; means for performing natural language processing on the response information and sentiment information to preprocess the data; and means for training a machine learning model using the preprocessed data to learn sentiment patterns. This enables accurate and personalized response predictions that take into account the user's emotional state.

[0736] "User" refers to an individual or legal entity that uses the system to answer a survey.

[0737] "Response information" refers to information and data that users have selected or entered in past surveys.

[0738] "Emotional information" refers to data that indicates a user's psychological or emotional state, extracted from their voice or text.

[0739] "Natural language processing" refers to the technologies and methods that enable computers to understand, interpret, and generate human language.

[0740] "Data preprocessing" refers to the process of shaping or transforming data before performing analysis or model training.

[0741] A "machine learning model" refers to an algorithm or structure used to learn patterns from data and perform predictions or classifications.

[0742] "Emotional patterns" refer to a series of tendencies and characteristics extracted based on the user's emotional information.

[0743] "Prediction" refers to the process by which a machine learning model uses training data to infer or estimate results for new data.

[0744] An "interface" refers to the screen or control panel that allows a user to interact with a system.

[0745] "Feedback" refers to information such as evaluations, opinions, and suggestions for improvement that users provide to the system.

[0746] "Accuracy" is an indicator that shows how accurately a system can respond to the user's intentions and requests.

[0747] This system provides a function that optimizes user survey responses through collaboration between the server and terminals. It primarily utilizes user sentiment information to achieve personalized response predictions.

[0748] server

[0749] The server collects users' past survey responses and sentiment information and stores it in a database. Sentiment information is extracted using speech recognition software and text analysis tools. Specifically, a general speech analysis service is used for speech recognition, and Python's natural language processing libraries (e.g., NLTK and spaCy) are used for text analysis.

[0750] The server preprocesses the collected data, performing tasks such as noise reduction and tokenization, and then formats the data for machine learning models. Machine learning is performed using platforms such as Scikit-learn and TensorFlow to train the models. These models are used to learn user sentiment patterns and predict answers to new questions.

[0751] terminal

[0752] The terminal receives predicted answers sent from the server and automatically inputs them into the user interface. This allows users to complete surveys with minimal effort. User feedback is also sent to the server via the terminal, contributing to the continuous improvement of the system.

[0753] Specific example

[0754] For example, when a user receives a survey question such as "How are you feeling today?", the server can predict the answer "I'm having fun today" based on past data and the user's current emotional state. As a result, the terminal automatically inputs this prediction, and after user confirmation, the survey is completed easily.

[0755] Example of a prompt

[0756] Based on the user's past survey responses and sentiment data, and taking into account their current emotional state, generate a predictive answer to the following question: 'How are you feeling today?'

[0757] This process allows the system to automatically generate more appropriate responses for each user, improving the accuracy of the survey and user satisfaction.

[0758] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0759] Step 1:

[0760] The server collects users' past survey responses and sentiment information and stores it in a database. Specifically, the server uses speech recognition software to extract text from audio data and a text analysis tool to extract sentiment information. It receives past audio and text data as input and stores the response information and sentiment information in the database as output.

[0761] Step 2:

[0762] The server preprocesses the collected data using natural language processing techniques. It performs noise reduction and tokenization on the data, converting the text data into a parseable format. It receives collected response information and sentiment information as input and generates preprocessed data as output. Specifically, it uses Python's NLTK to format the text data.

[0763] Step 3:

[0764] The server trains a machine learning model using preprocessed data. It learns patterns based on the features of the input data and builds a model. It receives preprocessed data as input and generates a trained machine learning model as output. This process utilizes Scikit-learn and TensorFlow to iteratively adjust the model's parameters.

[0765] Step 4:

[0766] The server uses a trained model to predict the best answer to a new survey question. It inputs current sentiment information into the model and calculates the best answer. It accepts a new question and current sentiment information as input and provides a predicted answer as output. Specifically, it uses prompt statements to supply the model with the question text and related data, taking sentiment states into account.

[0767] Step 5:

[0768] The terminal automatically populates the survey form with predicted answers sent from the server and allows the user to confirm them. It receives the predicted results from the server as input and displays the automatically populated state on the user interface as output. The user can then review and modify the data. Specifically, the terminal interface reflects the predicted answers and presents them to the user as editable fields.

[0769] Step 6:

[0770] The user sends feedback to the server via their device. The server receives the user feedback after review as input and provides data as output that is used for the system's continuous learning. The server, upon receiving the feedback, applies it to improve the sentiment engine and machine learning model to improve the accuracy of future predictions. Specifically, a process is executed to retrain the model using the new feedback data as input.

[0771] (Application Example 2)

[0772] 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".

[0773] In today's world, there is a need to optimize users' purchasing intentions and provide highly accurate product recommendations. However, conventional systems have been unable to reflect the individual emotional state of users in their recommendations, resulting in a decrease in the usefulness of the suggestions. To address this situation, there is a need to provide a system that considers past response information and current emotional state to provide personalized recommendations.

[0774] 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.

[0775] In this invention, the server includes means for collecting past response information and emotional data of the user, means for performing natural language processing on the response information and emotional data to preprocess the data, and means for training a machine learning model using the preprocessed data. This enables personalized suggestions based on the user's emotional tendencies.

[0776] A "user" is an individual or legal entity that uses the system or service.

[0777] "Response information" refers to the content of responses that users have given to past surveys or questions.

[0778] "Emotional data" refers to data that indicates the user's emotional state, and is extracted from voice and text input.

[0779] "Natural language processing" is a technology that enables computers to understand and process human language.

[0780] "Data preprocessing" is the process of preparing data to be suitable for machine learning algorithms.

[0781] A "machine learning model" is a statistical model that learns patterns from data to perform predictions and classifications.

[0782] "Training methods" refer to the process of providing data to a machine learning model and allowing it to learn patterns and relationships.

[0783] "Emotional state" refers to information that indicates the user's current psychological or emotional condition.

[0784] "Personalized recommendations" refer to recommendations for products and services tailored to the specific needs and emotions of the user.

[0785] A "connection device" is a device used by users to access the system.

[0786] The system for implementing the present invention includes a program that collects the user's past response information and emotional data, and utilizes natural language processing and machine learning techniques. The server runs on a platform such as Python or TensorFlow and extracts emotional data from the user's text and voice data. The extracted data is stored in a database and then used to train a machine learning model after preprocessing.

[0787] The server preprocesses data using a natural language processing toolkit (e.g., NLTK) and optimizes machine learning models. Based on the user's response patterns and emotional tendencies, it can generate personalized suggestions. These suggestions, which take into account the user's current emotional state, are provided on the user's device.

[0788] When a user accesses the system using a connected device such as a smartphone, the device presents the user with predicted responses and suggestions. The user's feedback is then sent back to the server and used to improve the model's performance. This feedback process allows the system to continuously improve the accuracy of the suggestions it provides.

[0789] For example, if a user enters "I'm feeling good today," the server will recommend appropriate products from those that the user has considered purchasing in similar emotional states in the past. An example of a prompt for the generative AI model used in this process could be in the form of, "What products would you recommend if the user's emotional data indicates they are feeling good?"

[0790] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0791] Step 1:

[0792] The server receives the user's past survey responses and emotional data collected from voice, text, etc. The input data includes the user's response history and information indicating their emotional state. The server stores this data in a database for later analysis. During this process, data cleansing is performed to remove inaccurate or invalid data.

[0793] Step 2:

[0794] The server preprocesses the collected response information and emotional data using a natural language processing toolkit (such as NLTK). This step converts the input data into a format that can be easily handled by machine learning models. Specifically, it performs tasks such as text tokenization, removal of special characters, and extraction of sentiment words to extract meaningful features.

[0795] Step 3:

[0796] The server trains a machine learning model using pre-processed data. The input data consists of past responses and emotional patterns extracted as features. The output is a model that generates responses and product suggestions based on the user's emotional state. This process utilizes libraries such as TensorFlow to train a neural network.

[0797] Step 4:

[0798] The server receives the user's current questions and emotional state as input and uses a trained model to predict the best response and product suggestion. The input data includes real-time user emotional data. As output, it generates personalized suggestions and responses and sends them to the terminal.

[0799] Step 5:

[0800] The terminal presents the user with predicted responses and suggestions received from the server. In this step, the suggestions are visually displayed through the user interface for the user to review. The user can then make decisions, select, or change based on the suggestions.

[0801] Step 6:

[0802] The device receives user feedback and sends it to the server. The input data consists of user opinions regarding the usefulness and satisfaction level of the suggestions. The server uses this feedback to retrain the machine learning model to improve its performance. This makes it possible to further improve the accuracy of future suggestions.

[0803] 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.

[0804] 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.

[0805] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0806] 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.

[0807] 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.

[0808] 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.

[0809] 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.

[0810] 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.

[0811] 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."

[0812] 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.

[0813] 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.

[0814] 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.

[0815] 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.

[0816] 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.

[0817] 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.

[0818] 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.

[0819] 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.

[0820] 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.

[0821] 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.

[0822] 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.

[0823] 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.

[0824] The following is further disclosed regarding the embodiments described above.

[0825] (Claim 1)

[0826] A means of collecting users' past response information,

[0827] A means for performing natural language processing on the aforementioned response information to preprocess the data,

[0828] A means for training a machine learning model using the aforementioned preprocessed data,

[0829] A means of using the trained model to predict the optimal answer to the current question,

[0830] A system including means for automatically inputting the predicted response into the user interface.

[0831] (Claim 2)

[0832] The system according to claim 1, wherein the machine learning model is configured to identify past question-and-answer patterns and predict an answer to the current question.

[0833] (Claim 3)

[0834] The system according to claim 1, further comprising a function to receive feedback from users and improve the performance of the machine learning model.

[0835] "Example 1"

[0836] (Claim 1)

[0837] A means of accumulating users' past response data,

[0838] A means of performing language processing on the aforementioned response data to organize the information,

[0839] A means for training a learning model using the aforementioned prepared information,

[0840] Means for utilizing the trained model to predict the optimal solution to the current question,

[0841] A means for automatically reflecting the aforementioned predicted solution in the user's operating device,

[0842] A system including means for receiving evaluations from users and improving the performance of the learning model.

[0843] (Claim 2)

[0844] The system according to claim 1, wherein the learning model is designed to recognize past question-and-answer formats and infer a solution to the current question.

[0845] (Claim 3)

[0846] The system according to claim 1, which constructs input sentences for a generating AI model based on user input and provides criteria for obtaining the optimal answer.

[0847] "Application Example 1"

[0848] (Claim 1)

[0849] A means of collecting information on the user's past instructions,

[0850] A means for performing natural language processing based on the aforementioned instruction information to preprocess the data,

[0851] A means for automatically generating optimal suggestions for home appliances based on user preferences,

[0852] A means for automatically inputting the generated proposal into the interface of a home appliance,

[0853] A system that includes means for retraining to improve the accuracy of suggestions based on feedback.

[0854] (Claim 2)

[0855] The system according to claim 1, configured to identify past instruction and response patterns and generate suggestions for the current situation.

[0856] (Claim 3)

[0857] The system according to claim 1, comprising a function to analyze usage patterns within the home and streamline device settings and operating procedures.

[0858] "Example 2 of combining an emotion engine"

[0859] (Claim 1)

[0860] A means of collecting users' past response information and sentiment information,

[0861] A means for performing natural language processing on the aforementioned response information and sentiment information to preprocess the data,

[0862] A means for training a machine learning model using the aforementioned preprocessed data and learning emotional patterns,

[0863] A means for predicting the optimal answer to the current question using the trained model, taking into account the user's emotional state,

[0864] A means for automatically inputting the predicted answer into the user interface and allowing the user to confirm it,

[0865] A means of receiving feedback from users and improving the accuracy of a machine learning model by taking the aforementioned sentiment information into consideration,

[0866] A system that includes this.

[0867] (Claim 2)

[0868] The system according to claim 1, wherein the machine learning model is configured to identify past question-and-answer patterns and sentiment patterns to predict an individualized answer to the current question.

[0869] (Claim 3)

[0870] The system according to claim 1, comprising a function for extracting emotional information from a user's voice or text using an emotion engine.

[0871] "Application example 2 when combining with an emotional engine"

[0872] (Claim 1)

[0873] A means of collecting users' past response information and emotional data,

[0874] A means for performing natural language processing on the response information and emotional data to preprocess the data,

[0875] A means for training a machine learning model using the aforementioned preprocessed data,

[0876] Means for using the trained model to predict the optimal response to the current question,

[0877] Means for automatically inputting the predicted response to the user's connection device,

[0878] A system that includes means for generating personalized suggestions based on the user's current emotional state.

[0879] (Claim 2)

[0880] The system according to claim 1, wherein the machine learning model is configured to identify past question-and-answer patterns to predict a response to a current question and to suggest products according to the user's emotional tendencies.

[0881] (Claim 3)

[0882] The system according to claim 1, further comprising a function to receive feedback from users, improve the performance of the machine learning model, and enhance the accuracy of the suggestions. [Explanation of Symbols]

[0883] 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 collecting information on the user's past instructions, A means for performing natural language processing based on the aforementioned instruction information to preprocess the data, A means for automatically generating optimal suggestions for home appliances based on user preferences, A means for automatically inputting the generated proposal into the interface of a home appliance, A system that includes means for retraining to improve the accuracy of suggestions based on feedback.

2. The system according to claim 1, configured to identify past instruction and response patterns and generate suggestions for the current situation.

3. The system according to claim 1, comprising a function to analyze usage patterns within the home and streamline device settings and operating procedures.

Images