system

Abstract

We provide the system. [Solution] Means for collecting user behavior data and feedback data, A means of identifying user interests and needs by analyzing collected data using natural language processing technology, A means for automatically generating service suggestions suitable for the user based on analysis results, A means of presenting the generated suggestions to the user interface, A system that includes this.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor and includes steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a character of the chatbot, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance 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] For operators providing communication-related services, it is a difficult task to quickly and appropriately respond to various needs and interests of users. In particular, it is important to accurately identify the services and promotions individually required by each user in order to enhance market competitiveness, but the conventional methods have not achieved sufficient efficiency in this regard. Therefore, there is a need for a mechanism that can effectively extract specific needs and interests from user behavior data and feedback, and propose appropriate services based on them.

Means for Solving the Problems

[0005] This invention provides a system that identifies user interests and needs by collecting user behavior data and feedback data and analyzing them using natural language processing technology. Based on the analysis results, this system generates service suggestions tailored to the user and presents them to the user through an interface. Furthermore, by collecting user feedback and continuously optimizing the content of the suggestions based on that feedback, it is possible to increase user satisfaction. Specifically, it includes means for performing topic modeling and sentiment analysis, with the aim of automating personalized suggestions for individual users and improving the efficiency of providing communication-related services.

[0006] "User behavior data" refers to information such as access history, operation logs, and communication history generated when a user uses a service.

[0007] "Feedback data" refers to information that includes opinions, evaluations, questions, and requests regarding the service provided by users.

[0008] "Natural language processing technology" is a technology that allows computers to analyze, understand, and generate human language, and it uses algorithms to process language data.

[0009] "Analysis" is the process of examining collected data in detail to find specific patterns or relationships.

[0010] "User interests and needs" refer to the characteristics related to information, products, or services that users are interested in and need.

[0011] "Service recommendation" refers to the act of recommending products, services, or promotions to users that are suitable for their interests and needs.

[0012] An "interface" is a means or method for a user to interact with a computer system or digital device.

[0013] Topic modeling is a technique that automatically extracts topics and themes from large amounts of text data.

[0014] "Sentiment analysis" is a technique that identifies emotions within text data and determines whether those emotions are positive, negative, or neutral.

[0015] A "generative AI model" is an algorithm or system used to generate suggestions or text based on user insights using artificial intelligence. [Brief explanation of the drawing]

[0016] [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] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This 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] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This 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] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This 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] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11]It is a sequence diagram showing the processing flow of the data processing system in Embodiment 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 Embodiment 2 when the 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 the emotion engine is combined.

Mode for Carrying Out the Invention

[0017] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described according to the accompanying drawings.

[0018] First, the terms used in the following description will be described.

[0019] In the following embodiments, the numbered processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of a plurality of arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of a plurality of 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.

[0020] In the following embodiments, the numbered RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0021] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0022] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0023] 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."

[0024] [First Embodiment]

[0025] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0026] 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.

[0027] 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).

[0028] 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.

[0029] 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.

[0030] 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.

[0031] 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.

[0032] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0033] 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.

[0034] 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.

[0035] 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.

[0036] 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".

[0037] The system of this invention utilizes user behavior data and feedback data, and analyzes them using natural language processing technology to identify user interests and needs, and then proposes services. This system is implemented in the following form.

[0038] Data collection and management

[0039] The server automatically collects each user's behavioral data, such as web browsing history, purchase history, and inquiry history. Furthermore, it manages customer support feedback from users. This data is stored in a database.

[0040] Data preprocessing

[0041] The server prepares the collected data into a format that can be analyzed. This process includes cleaning and normalizing text data and converting it into a format suitable for input to machine learning algorithms.

[0042] Data analysis and insight extraction

[0043] The server uses natural language processing techniques to analyze user behavior data and feedback. Specifically, it extracts relevant topics using topic modeling and performs sentiment analysis to understand the sentiment behind the feedback. This provides insights into identifying user interests and needs.

[0044] Proposal generation

[0045] The server generates service and promotional suggestions tailored to the user based on analysis. This generation process utilizes a generative AI model to create customized suggestions that match the user's areas of interest.

[0046] Presentation to the user

[0047] The device displays suggested content to the user. The system includes a feature that allows the user to review the suggestions on a dashboard the next time they access the system. The suggestions include specific service information and related promotions.

[0048] Utilizing and optimizing feedback

[0049] The server collects user feedback and responses to suggestions, and uses this to improve the accuracy of suggestions. The algorithm is fine-tuned based on feedback, aiming to provide a service that satisfies users.

[0050] For example, if a user has repeatedly searched for smart home appliance-related products in the past, the system can suggest new product announcements and special sale information tailored to their needs. It can also provide additional support information based on support inquiries. In this way, the system enables the provision of detailed services that meet the individual needs of each user.

[0051] The following describes the processing flow.

[0052] Step 1:

[0053] The server periodically collects user behavior data and feedback data. Behavioral data includes web browsing history, purchase history, and search queries, while feedback data includes customer support inquiries and ratings. This data is stored in a database with security measures in place.

[0054] Step 2:

[0055] The server preprocesses the collected data. This includes data cleansing and standardization of data formats. For example, it removes unnecessary characters from text data and converts it into a format suitable for analysis. It also removes noisy data and improves data quality.

[0056] Step 3:

[0057] The server analyzes the data using natural language processing technology. Topic modeling extracts themes of user interest, and sentiment analysis determines whether user feedback is positive or negative. From these analysis results, insights into user interests and needs are obtained.

[0058] Step 4:

[0059] The server generates service suggestions tailored to the user based on the analysis results. The generation model dynamically constructs personalized promotional plans and related service information for each user, creating customized suggestions.

[0060] Step 5:

[0061] The device presents generated suggestions to the user. Once the user logs into the system, personalized promotions and service suggestions are displayed on the dashboard. If the user is interested, links to access further information and purchase buttons are provided.

[0062] Step 6:

[0063] Users react to the suggestions presented. These reactions are indicated by actions such as clicking, viewing, or skipping. This reaction data is then sent back to the server for further analysis.

[0064] Step 7:

[0065] The server reanalyzes user responses and feedback data to optimize suggestions. Machine learning algorithms are used to analyze the data and adjust the system to provide more appropriate suggestions to each user. This process improves the accuracy of suggestions and enhances the user experience.

[0066] (Example 1)

[0067] 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."

[0068] In today's information society, users expect to be provided with information that is quick and accurate, tailored to their interests and needs. However, conventional systems fail to effectively utilize user behavior and feedback, making it difficult to provide sufficiently personalized suggestions. To solve this problem, there is a need for technology that optimally collects and analyzes user behavior data and feedback, and uses the results to generate highly customized suggestions.

[0069] 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.

[0070] In this invention, the server includes means for collecting behavioral data and feedback information from users and storing it in a data holder; means for normalizing and formatting the collected data using text processing technology and analyzing the data to recognize the user's interests and requests; and means for creating user-optimized service provision information using a generative model based on the analyzed information. This makes it possible to provide information tailored to individual needs based on each user's behavioral patterns and feedback.

[0071] A "user" is an individual or organization that uses an information system or service.

[0072] "Behavioral data" refers to information that shows a user's activities in the digital environment, such as web browsing and purchasing behavior.

[0073] "Feedback information" refers to information that includes opinions and evaluations that users give to a service or product.

[0074] A "data holder" refers to a storage device or database used to store and manage collected data.

[0075] "Text processing technology" refers to techniques for formatting collected text data into a parseable format, including cleaning and normalization.

[0076] "Normalization and formalization" is the process of converting data into a unified format and structuring it.

[0077] A "generative model" is a mathematical model that uses artificial intelligence to generate new data and suggestions.

[0078] "Service provision information" refers to information about services and promotions offered to users.

[0079] "Personalized recommendations" refer to recommendations of information and services that are customized based on the user's characteristics and past behavior.

[0080] The system of the present invention is designed to effectively collect, store, and analyze user behavior data and feedback information, and to provide users with information optimized for their needs. This system is implemented using the hardware and software configuration shown below.

[0081] The server is primarily responsible for data collection, processing, analysis, and suggestion generation. Data and feedback information regarding user behavior are automatically acquired by the server and stored in data holders. At this stage, the server uses a combination of web servers and database software (e.g., MySQL®, PostgreSQL). The information stored in the data holders is cleansed and normalized using text processing techniques.

[0082] In data analysis, the server applies natural language processing techniques, specifically performing topic clustering and sentiment evaluation. This process utilizes NLP libraries (e.g., NLTK, spaCy) and machine learning libraries (e.g., TENSORFLOW®, PyTorch). Based on the insights gained from this analysis, the server leverages generative models to generate personalized service information for the user. Generative AI models (e.g., GPT series) are used in this generation process. An example of a prompt is given to the AI ​​model: "Based on the user's recent smart home appliance search history, suggest suitable new products and promotions."

[0083] The generated suggestions are sent to the user's device and displayed on the user interface. These devices primarily refer to user interface devices such as PCs, tablets, or smartphones, and the suggested information is presented to the user visually in a dashboard format.

[0084] For example, if a user frequently searches for disaster preparedness supplies, the server analyzes their behavior and suggests the user information on the latest disaster preparedness kits and related campaigns. In this way, users can easily receive personalized information tailored to their interests and needs.

[0085] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0086] Step 1:

[0087] The server collects user behavior data and feedback information and stores it in a data holder. Specifically, it automatically logs pages the user accessed on the website, advertisements clicked, purchase history, and other similar information. Furthermore, feedback that the user sent to customer support is also collected. The input to this process is user behavior and feedback, and the output is a structured dataset.

[0088] Step 2:

[0089] The server preprocesses the collected data using text processing techniques. It removes unnecessary characters, normalizes the data, and standardizes the format from the raw input data. At this stage, morphological analysis is performed to segment the text, and the results are converted into a parseable format. The output is clean data suitable for analysis.

[0090] Step 3:

[0091] The server performs data analysis using natural language processing technology. Specifically, it extracts topics related to user interests using topic clustering and determines the sentiment of feedback through sentiment evaluation. The input for this process is pre-formatted data, and the analysis method outputs insights into the user's interests and needs.

[0092] Step 4:

[0093] Based on the insights gained from the analysis, the server uses a generative AI model to generate service information optimized for the user. In this process, prompt sentences are input to the generative AI model to generate new suggestions. The input consists of the analysis results and prompt sentences, and the output is personalized suggestion information.

[0094] Step 5:

[0095] The terminal visually presents the suggested information sent from the server to the user. Specifically, it displays new service and promotional information in a dashboard format within the user interface. The input to this process is the generated suggested information, and the output is a visual display that the user can view.

[0096] Step 6:

[0097] The server collects user feedback again and analyzes the effectiveness of the generated suggestions based on it. Based on the feedback, it refines the running algorithm and generative model, improving the accuracy of suggestions in future sessions. The input for this step is user feedback, and the output is the improved algorithm and model.

[0098] (Application Example 1)

[0099] 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."

[0100] Effectively presenting products or information tailored to the individual needs of entities on online platforms is a challenging task. In today's world, where diverse data is generated, traditional methods struggle to quickly and accurately capture entities' interests and provide appropriate suggestions. Furthermore, continuous optimization of suggestions through the use of entity feedback is also required.

[0101] 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.

[0102] In this invention, the server includes means for collecting entity behavior data and feedback data, means for analyzing the collected data using analytical techniques to identify the entity's interests and needs, and means for automatically generating product or information suggestions suitable for the entity based on the analysis results. This makes it possible to quickly grasp the individual needs of entities and make highly accurate suggestions.

[0103] An "entity" refers to the object that generates behavioral data and feedback data within the system, and in this invention, it mainly refers to individual users.

[0104] "Behavioral data" refers to data that includes information such as the history of actions performed by an entity on an online platform, browsing history, and choice behaviors.

[0105] "Feedback data" refers to data that includes evaluations, comments, and reactions that entities give to a proposed product or information.

[0106] "Analysis techniques" refer to technical methods used to process collected data and identify the interests and needs of entities, and in this invention, this includes natural language processing techniques.

[0107] "Presenting products or information" refers to the process of selecting products or related information suitable for an entity based on the analysis results and displaying them to the entity.

[0108] "Automatic generation" refers to the process by which a system uses analysis results to create suitable suggestions for entities without human intervention.

[0109] A "display device" refers to a terminal or screen device that has an interface function that allows an entity to visually confirm the proposed content.

[0110] "Evaluations and comments" refer to the act of providing feedback to an entity on a proposal it has received, and are used to improve the quality of the proposal.

[0111] The server uses a cloud infrastructure connected to a database management system to efficiently collect entity behavior data and feedback data. Amazon Web Services (AWS®) and Google Cloud Platform are often used for this purpose. The server processes the acquired data via streaming, enabling real-time data collection.

[0112] Next, the server analyzes the collected data using natural language processing technology. This analysis utilizes the Python programming language and its related libraries, spaCy and Gensim. Specifically, Gensim's LDA (Latent Dirichlet Allocation) is used for topic modeling, and the NLTK library's VADER is used for sentiment analysis. Using these, insights tailored to the interests and needs of the entities are extracted.

[0113] Based on the analysis results, the server uses a generative AI model to automatically generate product or information suggestions suitable for the entity. This model is based on Hugging Face's Transformer model and generates customized suggestions based on diverse data. The generated suggestions are formatted in JSON format and sent directly to the terminal.

[0114] On the device, the entity displays product or information generated via the display device. This is implemented using a dynamic web interface built into the application. Frontend libraries such as React and Vue.js are often used for this.

[0115] Users provide feedback on the presented products or information. This feedback is collected by the server and used to improve the algorithm. Specifically, learning-based improvements are made, such as increasing the frequency of suggesting product categories that users have given many positive ratings to.

[0116] For example, if a user has frequently viewed sports equipment in the past, the generative AI model will use that information to suggest the latest sports equipment and sales information to the user via push notifications. An example of a prompt to the generative AI model would be: "Generate promotional information related to sports equipment that the user is interested in. Please include specific product names and discount information, taking into account past purchase history and feedback."

[0117] In this way, based on entity behavior data and feedback data, it becomes possible to perform advanced analysis and automatically generate and present personalized suggestions in real time.

[0118] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0119] Step 1:

[0120] The server uses a cloud infrastructure to collect entity behavior data and feedback data. This data includes browsing history, purchase history, and support inquiries. The collected data is stored in an AWS S3 bucket and then accumulated in a database. The input is user behavior information across the network, and the output is structured data stored in the database.

[0121] Step 2:

[0122] The server preprocesses the collected data. Using Python, it formats the data with the pandas library. Specifically, it cleans the text data and performs normalization to make it suitable for analysis. The input is raw structured data, and the output is a tokenized dataset.

[0123] Step 3:

[0124] The server analyzes data using natural language processing techniques. It uses Gensim's LDA for topic modeling and NLTK's VADER for sentiment analysis. This provides insights into user interests and needs. The input is a pre-processed dataset, and the output is insightful information.

[0125] Step 4:

[0126] The server automatically generates suggestions using a generative AI model. It leverages Hugging Face's transformer model to create customized service and product information tailored to the user's areas of interest. Instructions are given to the model using prompts. Input consists of insight information and prompts, while output is the generated suggestions.

[0127] Step 5:

[0128] The terminal presents the generated suggestions to the user. Specifically, it visually displays the information to the user using a web interface within the application. The dashboard is built using the React library. The input is suggestion data in JSON format, and the output is a visual display for the user.

[0129] Step 6:

[0130] Users provide feedback based on the presented content. This feedback is sent to the server through the application and used to generate future proposals. The feedback input improves the quality of the proposals and enhances the accuracy of the algorithm. The input is user feedback, and the output is the improved proposal model.

[0131] 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.

[0132] This invention is a system that utilizes user behavior data and feedback data, combining natural language processing technology and an emotion engine. This system recognizes user emotions and adjusts suggestions based on those emotions, thereby achieving more personalized service delivery. Its specific form is described below.

[0133] Data collection

[0134] The server collects user behavior data, feedback data, and sensor data (e.g., voice tone, facial recognition results). This allows it to capture the user's intentions and emotional changes in the digital environment.

[0135] Data preprocessing

[0136] The server formats the collected data and converts it into a format suitable for analysis by natural language processing and sentiment engines. This includes cleaning text data and transcribing audio data.

[0137] Natural language processing and sentiment analysis

[0138] The server uses natural language processing technology to extract interests and needs from the user's text data. It also utilizes an emotion engine to analyze the user's emotional state, including negative emotion detection and positive emotion evaluation.

[0139] Proposal generation

[0140] Based on the analysis results, the server generates service and promotional suggestions tailored to the user. The generation model creates different suggestions depending on the user's emotions and provides information at the optimal time for their emotional state.

[0141] Emotion-based adjustment

[0142] The server adjusts suggestions based on the user's sentiment analysis. For example, it might suggest relaxing services when the user is stressed, or offer special benefits if positive emotions are detected.

[0143] Presentation to the user

[0144] The device presents personalized suggestions to the user. A dashboard that reflects the user's emotional state effectively delivers content that resonates with them. This process reduces unnecessary user interaction, providing a comfortable experience.

[0145] Gathering and adjusting feedback

[0146] Users react to suggestions and provide feedback. This feedback is used as an ongoing data source for the system to improve the suggestions, and the accuracy of the suggestions is readjusted.

[0147] As a concrete example, if the emotion engine detects that a user is feeling stressed, it will suggest a trial of a music streaming service that helps with relaxation. Furthermore, if a user shows interest in a new technology product and their mood is analyzed as positive, it will provide information on purchase benefits related to the latest technology product. In this way, the present invention provides users with an advanced experience and realizes services tailored to their individual needs and emotions.

[0148] The following describes the processing flow.

[0149] Step 1:

[0150] The server collects user behavior data. Specifically, it records the web pages users visit, their online shopping purchase history, and search queries on digital platforms. It also collects feedback data received directly from users, such as inquiries to customer support and survey responses.

[0151] Step 2:

[0152] The server formats the collected data. This process involves cleaning noisy text data, transcribing audio data, and standardizing the data format. This facilitates natural language processing and sentiment analysis.

[0153] Step 3:

[0154] The server uses natural language processing technology to analyze user behavior data and feedback data. This analysis extracts user interests and needs. For example, it identifies frequently occurring keywords and topics to pinpoint content that users are interested in.

[0155] Step 4:

[0156] The server uses an emotion engine to analyze user feedback data and real-time voice and facial expression data to understand emotions. The emotion engine evaluates the user's emotional state as positive, negative, or neutral.

[0157] Step 5:

[0158] The server generates optimal service suggestions for the user based on analyzed interest and emotion data. In this process, a generative AI model creates suggestions that correspond to the user's current emotional state. For example, if it is determined that relaxation is needed, it will suggest relaxing music services.

[0159] Step 6:

[0160] The device presents suggestions to the user. The displayed information is customized to the user's emotional state and clearly organized on the dashboard. The device provides an interactive interface that allows the user to easily accept or skip suggestions.

[0161] Step 7:

[0162] Users react to the suggestions presented. They can click positively, ignore, or provide additional feedback on the suggestions. These reactions are important indicators for evaluating whether the user's needs and feelings are being accurately captured.

[0163] Step 8:

[0164] The server recollects user responses and optimizes the system based on the feedback. The newly collected feedback data is analyzed, and the generative AI model and sentiment engine are adjusted to further improve the accuracy of future suggestions. This continuous analysis and optimization process constantly improves the relevance and usefulness of suggestions to the user.

[0165] (Example 2)

[0166] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0167] There is a growing need to improve the user experience by more accurately understanding the user's state and providing personalized services based on that understanding. However, current systems do not adequately optimize services based on the user's state and emotions, which is hindering improvements in user satisfaction.

[0168] 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.

[0169] In this invention, the server includes means for collecting user status data, means for analyzing the collected data using processing technology to identify the user's status, and means for generating service content suitable for the user based on the analysis results. This makes it possible to provide more personalized services according to the user's status.

[0170] "User state data" refers to information obtained from a user's actions and reactions, and is used to analyze the user's current emotions and needs.

[0171] "Collecting" refers to the process of gathering information and data for a specific purpose, which then integrates the necessary data.

[0172] "Processing technology" refers to a set of techniques for analyzing and understanding data and generating appropriate output based on a specific purpose.

[0173] "Identifying the user's state" refers to revealing the user's emotions and interests based on collected data and evaluating that situation.

[0174] "Generating service content" means assembling appropriate services and information to provide to users based on the analysis results.

[0175] "Presenting on a display device" means visually showing the generated service content in a user-friendly format.

[0176] "Collecting feedback" is the process of gathering user responses and opinions, and obtaining information to be used for future improvements and service optimization.

[0177] "Adjusting information" is the process of updating service content and proposals based on feedback and data received, with the aim of achieving better results.

[0178] "Analytical techniques" is a general term for methods and tools used to extract and interpret useful information from data.

[0179] "State analysis" is a technique that analyzes the internal and external states of a user, enabling an understanding of their background and influence.

[0180] The system of this invention incorporates technology to accurately understand the user's state and make appropriate suggestions based on that understanding, in order to provide personalized services to the user. Specifically, the server, terminal, and user work together.

[0181] The server takes the lead in collecting user behavior data and feedback data. This includes data obtained from sensors, such as voice tone and facial expressions. The data is then processed using advanced techniques to identify the user's current state. Natural language processing and emotion engines are used in the analysis to specifically extract the user's emotional state and interests. This involves the use of voice recognition software and emotion analysis tools.

[0182] Based on the analysis results, the server generates optimal service content for the user through a generative AI model. Examples include "recommended activities for relaxation" or "purchase benefits for technology products." An example of a prompt used in this process is "make suggestions based on the user's situation." This sentence is used to instruct the generative AI model on specific conditions, resulting in appropriate output according to the user's state.

[0183] The terminal visually displays the generated service details to the user. This utilizes a sophisticated user interface, designed to allow users to easily receive information.

[0184] Finally, the user provides feedback on the information presented. This feedback is sent to the server and used to optimize future service suggestions. In this way, the system reflects the user's state and delivers a more personalized user experience.

[0185] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0186] Step 1:

[0187] The server collects user behavior data and feedback data. Specifically, it obtains voice tone, facial expression data, application usage history, etc., transmitted from the device, from sensors and log databases. The input is the user's digital activity and sensor data, and the output is integrated state data. This makes it possible to comprehensively understand the user's intentions and emotional changes.

[0188] Step 2:

[0189] The server preprocesses the collected data. Specifically, it uses speech recognition software to convert speech data into text and then uses natural language processing techniques to clean the text into a parseable format. The input is integrated state data, and the output includes preprocessed text data. This process makes the data effectively usable by the analysis engine.

[0190] Step 3:

[0191] The server applies natural language processing and sentiment analysis to analyze pre-processed data. The analysis includes processing techniques to extract user interests and needs, and an emotion engine to analyze emotional states. The input is pre-processed text data, and the output includes analysis results indicating the user's state. This allows for a concrete understanding of the user's emotions and interests.

[0192] Step 4:

[0193] The server uses a generative AI model based on the analysis results to generate service content tailored to the user. Specifically, prompts are used to instruct the generative AI model, obtaining individual suggestions based on the user's state. The input consists of the analysis results and prompts, and the output includes personalized service suggestions. This makes it possible to provide content appropriate to the user's state.

[0194] Step 5:

[0195] The server adjusts the generated suggestions based on the user's emotions, and the device displays the content. Specifically, it re-evaluates the user's state data and optimizes the suggestion content and presentation timing. The input consists of personalized suggestions and state data, while the output includes the adjusted suggestions. The device presents the information in a visually easy-to-understand format within the user interface.

[0196] Step 6:

[0197] The system provides users with feedback on the information presented. Specifically, it allows users to easily input their satisfaction level and suggestions for improvement through a user interface on their device, and this feedback is sent to the server. The input is the user's response, and the output includes feedback data. This makes it possible to accumulate data necessary for optimizing the service content in the future.

[0198] (Application Example 2)

[0199] 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."

[0200] Modern content delivery demands suggestions tailored to user interests and needs, but the challenge lies in providing personalized content that also takes into account the user's emotional state. In particular, when selecting visually and aurally engaging content, it is crucial to deliver the appropriate emotional experience at the time when the user will enjoy it the most.

[0201] 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.

[0202] In this invention, the server includes means for collecting user behavior data, feedback data, and emotional data; means for analyzing the collected data using natural language processing technology and emotion analysis technology to identify the user's interests, needs, and emotional state; and means for selecting and automatically generating content optimized for the user based on the analysis results. This makes it possible to provide users with content that is tailored to their emotional state, thereby realizing an enhanced user experience.

[0203] "User behavior data" refers to records of actions and choices that users make on digital platforms.

[0204] "Feedback data" refers to information such as evaluations and comments on suggestions and services provided by users.

[0205] "Emotional data" refers to information that indicates a user's emotional state, obtained from their facial expressions, voice, and other sources.

[0206] "Natural language processing technology" is a technology that uses computers to analyze human language and understand its meaning and intent.

[0207] "Emotion analysis technology" is a technology used to identify and evaluate a user's emotional state from data.

[0208] "Content" refers to the raw materials for information and experiences intended to be provided to users, and includes digital formats such as video, audio, images, and text.

[0209] Personalization is the process of tailoring and delivering experiences based on the individual user's interests, needs, and emotional state.

[0210] An "interface" is a means or method for a user to interact with a system or device.

[0211] The system implementing this invention is designed to collect user behavior data, feedback data, and sentiment data, and to personalize and deliver content based on this data.

[0212] The server collects various user data through smartphones and other devices. Specifically, it uses cameras and microphones to acquire emotional data such as the user's facial expressions and voice tone, and accumulates behavioral data and feedback as a digital operation history. This data is aggregated on a cloud service and processed on high-performance servers equipped with GPUs.

[0213] The server analyzes the collected data using natural language processing toolkits (e.g., SpaCy) and sentiment analysis APIs (e.g., Azure®'s Sentiment Analysis) to identify the user's interests, needs, and even emotional state. Based on this analysis, it selects the most suitable content for the user and generates recommendations. For example, if the server determines that the user is fatigued, it will select relaxing videos to provide emotional satisfaction.

[0214] The device presents recommendations for generated content to the user through its interface. Content that appeals to both visual and auditory senses is displayed at the most appropriate time for the user's environment and emotions, enabling effective entertainment and information delivery.

[0215] Users provide feedback on the presented content, which contributes to generating more precise recommendations in the future. This feedback is used as training data to continuously refine the recommendation strategy.

[0216] For example, if a high school student is tired from studying and wants to listen to music to refresh themselves, the server will perform sentiment analysis and suggest a relaxing music playlist. If the user sends feedback indicating satisfaction with the suggestion, the system will learn similar tendencies and be able to make even more accurate suggestions in the future.

[0217] Example of a prompt:

[0218] Based on the current user's emotional data, suggest content suitable for relaxation. If the user is tired, prioritize selecting videos that include nature sounds.

[0219] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0220] Step 1:

[0221] The server collects facial expression and audio data through the camera and microphone on the user's smartphone. This collects raw data that reflects the user's emotions. At this point, the input is raw video and audio recordings, and the output is the unprocessed data transferred to the server.

[0222] Step 2:

[0223] The server first preprocesses the received raw data. Here, it uses OpenCV to extract facial features from the video data and the Google Speech-to-Text API to convert the audio to text. The input is the raw data from step 1, and the output is the transcribed audio data and the identified facial feature data.

[0224] Step 3:

[0225] The server performs analysis using a natural language processing engine and an emotion analysis engine with pre-processed data. Specifically, it uses Hugging Face's Transformers to analyze the user's interests and emotions from the text and evaluates the emotional state using Azure's Sentiment Analysis API. The input is the text and facial feature data from step 2, and the output is information about the user's interests, emotional state, and emotional intensity.

[0226] Step 4:

[0227] The server runs a model to suggest content best suited to the user's state based on the analysis results. It uses TensorFlow to select content that matches the user's emotional state. The input is the analysis results obtained in step 3, and the output is a personalized content recommendation list for each user.

[0228] Step 5:

[0229] The terminal displays the generated content recommendation list on the user's device screen. The interface is adjusted so that the user can directly experience the content visually and aurally. The input is the recommendation list from step 4, and the output is the specific content presented via the user interface.

[0230] Step 6:

[0231] Users submit feedback on the provided content through an in-app interface. This feedback is data indicating how satisfied the user was with the provided content. The input is the actual content presented, and the output includes user ratings, comments, and satisfaction scores.

[0232] Step 7:

[0233] The server analyzes the feedback data and feeds it to an algorithm that optimizes the next content suggestion based on that analysis. Past feedback stored in MongoDB is used to improve future learning. The input is the feedback data from step 6, and the output is the improved next suggestion strategy.

[0234] 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.

[0235] 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.

[0236] 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.

[0237] [Second Embodiment]

[0238] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0239] 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.

[0240] 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).

[0241] 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.

[0242] 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.

[0243] 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).

[0244] 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.

[0245] 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.

[0246] 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.

[0247] 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.

[0248] 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.

[0249] 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".

[0250] The system of this invention utilizes user behavior data and feedback data, and analyzes them using natural language processing technology to identify user interests and needs, and then proposes services. This system is implemented in the following form.

[0251] Data collection and management

[0252] The server automatically collects each user's behavioral data, such as web browsing history, purchase history, and inquiry history. Furthermore, it manages customer support feedback from users. This data is stored in a database.

[0253] Data preprocessing

[0254] The server prepares the collected data into a format that can be analyzed. This process includes cleaning and normalizing text data and converting it into a format suitable for input to machine learning algorithms.

[0255] Data analysis and insight extraction

[0256] The server uses natural language processing techniques to analyze user behavior data and feedback. Specifically, it extracts relevant topics using topic modeling and performs sentiment analysis to understand the sentiment behind the feedback. This provides insights into identifying user interests and needs.

[0257] Proposal generation

[0258] The server generates service and promotional suggestions tailored to the user based on analysis. This generation process utilizes a generative AI model to create customized suggestions that match the user's areas of interest.

[0259] Presentation to the user

[0260] The device displays suggested content to the user. The system includes a feature that allows the user to review the suggestions on a dashboard the next time they access the system. The suggestions include specific service information and related promotions.

[0261] Utilizing and optimizing feedback

[0262] The server collects user feedback and responses to suggestions, and uses this to improve the accuracy of suggestions. The algorithm is fine-tuned based on feedback, aiming to provide a service that satisfies users.

[0263] For example, if a user has repeatedly searched for smart home appliance-related products in the past, the system can suggest new product announcements and special sale information tailored to their needs. It can also provide additional support information based on support inquiries. In this way, the system enables the provision of detailed services that meet the individual needs of each user.

[0264] The following describes the processing flow.

[0265] Step 1:

[0266] The server periodically collects user behavior data and feedback data. Behavioral data includes web browsing history, purchase history, and search queries, while feedback data includes customer support inquiries and ratings. This data is stored in a database with security measures in place.

[0267] Step 2:

[0268] The server preprocesses the collected data. This includes data cleansing and standardization of data formats. For example, it removes unnecessary characters from text data and converts it into a format suitable for analysis. It also removes noisy data and improves data quality.

[0269] Step 3:

[0270] The server analyzes the data using natural language processing technology. Topic modeling extracts themes of user interest, and sentiment analysis determines whether user feedback is positive or negative. From these analysis results, insights into user interests and needs are obtained.

[0271] Step 4:

[0272] The server generates service suggestions tailored to the user based on the analysis results. The generation model dynamically constructs promotional proposals and related service information for each user, creating customized suggestions.

[0273] Step 5:

[0274] The device presents generated suggestions to the user. Once the user logs into the system, personalized promotions and service suggestions are displayed on the dashboard. If the user is interested, links to access further information and purchase buttons are provided.

[0275] Step 6:

[0276] Users react to the suggestions presented. These reactions are indicated by actions such as clicking, viewing, or skipping. This reaction data is then sent back to the server for further analysis.

[0277] Step 7:

[0278] The server reanalyzes user responses and feedback data to optimize suggestions. Machine learning algorithms are used to analyze the data and adjust the system to provide more appropriate suggestions to each user. This process improves the accuracy of suggestions and enhances the user experience.

[0279] (Example 1)

[0280] 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."

[0281] In modern information society, users expect to be provided with information quickly and accurately according to their interests and needs. However, in conventional systems, it is difficult to effectively utilize users' behaviors and feedback, and it is challenging to make highly individualized proposals. To solve this problem, there is a need for a technology that optimally collects and analyzes users' behavioral data and feedback, and uses the results to generate highly customized proposals.

[0282] The specific processing by the specific processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0283] In this invention, the server includes means for collecting data related to actions and feedback information from users and storing them in a data holder, means for normalizing and formatting the collected data using text processing technology and recognizing users' interests and requirements by analyzing the data, and means for creating service provision information optimized for users by utilizing a generation model based on the analyzed information. Thereby, it becomes possible to provide information tailored to individual needs based on the behavioral patterns and feedback of each user.

[0284] A "user" refers to an individual or organization that uses an information system or service.

[0285] "Data related to actions" refers to information indicating the content of activities in a digital environment, such as web browsing and purchasing actions performed by users.

[0286] "Feedback information" refers to information including opinions and evaluations shown by users for services or products.

[0287] A "data holder" refers to a storage device or database used to store and manage the collected data.

[0288] "Text processing technology" refers to techniques for formatting collected text data into a parseable format, including cleaning and normalization.

[0289] "Normalization and formalization" is the process of converting data into a unified format and structuring it.

[0290] A "generative model" is a mathematical model that uses artificial intelligence to generate new data and suggestions.

[0291] "Service provision information" refers to information about services and promotions offered to users.

[0292] "Personalized recommendations" refer to recommendations of information and services that are customized based on the user's characteristics and past behavior.

[0293] The system of the present invention is designed to effectively collect, store, and analyze user behavior data and feedback information, and to provide users with information optimized for their needs. This system is implemented using the hardware and software configuration shown below.

[0294] The server is primarily responsible for data collection, processing, analysis, and suggestion generation. Data and feedback information regarding user behavior are automatically acquired by the server and stored in data holders. At this stage, the server uses a combination of web servers and database software (e.g., MySQL, PostgreSQL). The information stored in the data holders is cleansed and normalized using text processing techniques.

[0295] In data analysis, the server applies natural language processing techniques, specifically performing topic clustering and sentiment evaluation. This process utilizes NLP libraries (e.g., NLTK, spaCy) and machine learning libraries (e.g., TensorFlow, PyTorch). Based on the insights gained from this analysis, the server leverages generative models to generate personalized service offerings for the user. Generative AI models (e.g., GPT series) are used in this generation process. An example of a prompt is given to the AI ​​model: "Based on the user's recent smart home appliance search history, suggest suitable new products and promotions."

[0296] The generated suggestions are sent to the user's device and displayed on the user interface. These devices primarily refer to user interface devices such as PCs, tablets, or smartphones, and the suggested information is presented to the user visually in a dashboard format.

[0297] For example, if a user frequently searches for disaster preparedness supplies, the server analyzes their behavior and suggests the user information on the latest disaster preparedness kits and related campaigns. In this way, users can easily receive personalized information tailored to their interests and needs.

[0298] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0299] Step 1:

[0300] The server collects user behavior data and feedback information and stores it in a data holder. Specifically, it automatically logs pages the user accessed on the website, advertisements clicked, purchase history, and other similar information. Furthermore, feedback that the user sent to customer support is also collected. The input to this process is user behavior and feedback, and the output is a structured dataset.

[0301] Step 2:

[0302] The server preprocesses the collected data using text processing technology. From the input raw data, it deletes unnecessary character strings, normalizes the data, and unifies the formats. At this stage, morphological analysis is performed to segment the text into words and convert the results into an analyzable format. The output is clean data suitable for analysis.

[0303] Step 3:

[0304] The server performs data analysis using natural language processing technology. Specifically, it extracts topics related to the user's interests using topic clustering and performs sentiment evaluation to determine the sentiment of the feedback. The input in this process is the formatted data, and insights regarding the user's interests and needs are output by the analysis method.

[0305] Step 4:

[0306] Based on the insights obtained from the analysis, the server uses a generative AI model to generate service offering information optimized for the user. In this process, a prompt sentence is input into the generative AI model to generate new proposals. The input is the analysis result and the prompt sentence, and the output is individualized proposal information.

[0307] Step 5:

[0308] The terminal visually presents the proposal information sent from the server to the user. Specifically, in the user interface, new service and promotion information is displayed in a dashboard format. The input of this process is the generated proposal information, and the output is a visual display that the user can view.

[0309] Step 6:

[0310] The server collects user feedback again and analyzes the effectiveness of the generated suggestions based on it. Based on the feedback, it refines the running algorithm and generative model, improving the accuracy of suggestions in future sessions. The input for this step is user feedback, and the output is the improved algorithm and model.

[0311] (Application Example 1)

[0312] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0313] Effectively presenting products or information tailored to the individual needs of entities on online platforms is a challenging task. In today's world, where diverse data is generated, traditional methods struggle to quickly and accurately capture entities' interests and provide appropriate suggestions. Furthermore, continuous optimization of suggestions through the use of entity feedback is also required.

[0314] 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.

[0315] In this invention, the server includes means for collecting entity behavior data and feedback data, means for analyzing the collected data using analytical techniques to identify the entity's interests and needs, and means for automatically generating product or information suggestions suitable for the entity based on the analysis results. This makes it possible to quickly grasp the individual needs of entities and make highly accurate suggestions.

[0316] An "entity" refers to the object that generates behavioral data and feedback data within the system, and in this invention, it mainly refers to individual users.

[0317] "Behavioral data" refers to data that includes information such as the history of actions performed by an entity on an online platform, browsing history, and choice behaviors.

[0318] "Feedback data" refers to data that includes evaluations, comments, and reactions that entities give to a proposed product or information.

[0319] "Analysis techniques" refer to technical methods used to process collected data and identify the interests and needs of entities, and in this invention, this includes natural language processing techniques.

[0320] "Presenting products or information" refers to the process of selecting products or related information suitable for an entity based on the analysis results and displaying them to the entity.

[0321] "Automatic generation" refers to the process by which a system uses analysis results to create suitable suggestions for entities without human intervention.

[0322] A "display device" refers to a terminal or screen device that has an interface function that allows an entity to visually confirm the proposed content.

[0323] "Evaluations and comments" refer to the act of providing feedback to an entity on a proposal it has received, and are used to improve the quality of the proposal.

[0324] The server uses a cloud infrastructure connected to a database management system to efficiently collect entity behavior data and feedback data. Amazon Web Services (AWS) and Google Cloud Platform are often used for this purpose. The server processes the acquired data via streaming, enabling real-time data collection.

[0325] Next, the server analyzes the collected data using natural language processing technology. This analysis utilizes the Python programming language and its related libraries, spaCy and Gensim. Specifically, Gensim's LDA (Latent Dirichlet Allocation) is used for topic modeling, and the NLTK library's VADER is used for sentiment analysis. Using these, insights tailored to the interests and needs of the entities are extracted.

[0326] Based on the analysis results, the server uses a generative AI model to automatically generate product or information suggestions suitable for the entity. This model is based on Hugging Face's Transformer model and generates customized suggestions based on diverse data. The generated suggestions are formatted in JSON format and sent directly to the terminal.

[0327] On the device, the entity displays product or information generated via the display device. This is implemented using a dynamic web interface built into the application. Frontend libraries such as React and Vue.js are often used for this.

[0328] Users provide feedback on the presented products or information. This feedback is collected by the server and used to improve the algorithm. Specifically, learning-based improvements are made, such as increasing the frequency of suggesting product categories that users have given many positive ratings to.

[0329] For example, if a user has frequently viewed sports equipment in the past, the generative AI model will use that information to suggest the latest sports equipment and sales information to the user via push notifications. An example of a prompt to the generative AI model would be: "Generate promotional information related to sports equipment that the user is interested in. Please include specific product names and discount information, taking into account past purchase history and feedback."

[0330] In this way, based on entity behavior data and feedback data, it becomes possible to perform advanced analysis and automatically generate and present personalized suggestions in real time.

[0331] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0332] Step 1:

[0333] The server uses a cloud infrastructure to collect entity behavior data and feedback data. This data includes browsing history, purchase history, and support inquiries. The collected data is stored in an AWS S3 bucket and then accumulated in a database. The input is user behavior information across the network, and the output is structured data stored in the database.

[0334] Step 2:

[0335] The server preprocesses the collected data. Using Python, it formats the data with the pandas library. Specifically, it cleans the text data and performs normalization to make it suitable for analysis. The input is raw structured data, and the output is a tokenized dataset.

[0336] Step 3:

[0337] The server analyzes data using natural language processing techniques. It uses Gensim's LDA for topic modeling and NLTK's VADER for sentiment analysis. This provides insights into user interests and needs. The input is a pre-processed dataset, and the output is insightful information.

[0338] Step 4:

[0339] The server automatically generates suggestions using a generative AI model. It leverages Hugging Face's transformer model to create customized service and product information tailored to the user's areas of interest. Instructions are given to the model using prompts. Input consists of insight information and prompts, while output is the generated suggestions.

[0340] Step 5:

[0341] The terminal presents the generated suggestions to the user. Specifically, it visually displays the information to the user using a web interface within the application. The dashboard is built using the React library. The input is suggestion data in JSON format, and the output is a visual display for the user.

[0342] Step 6:

[0343] Users provide feedback based on the presented content. This feedback is sent to the server through the application and used to generate future proposals. The feedback input improves the quality of the proposals and enhances the accuracy of the algorithm. The input is user feedback, and the output is the improved proposal model.

[0344] 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.

[0345] This invention is a system that utilizes user behavior data and feedback data, combining natural language processing technology and an emotion engine. This system recognizes user emotions and adjusts suggestions based on those emotions, thereby achieving more personalized service delivery. Its specific form is described below.

[0346] Data collection

[0347] The server collects user behavior data, feedback data, and sensor data (e.g., voice tone, facial recognition results). This allows it to capture the user's intentions and emotional changes in the digital environment.

[0348] Data preprocessing

[0349] The server formats the collected data and converts it into a format suitable for analysis by natural language processing and sentiment engines. This includes cleaning text data and transcribing audio data.

[0350] Natural language processing and sentiment analysis

[0351] The server uses natural language processing technology to extract interests and needs from the user's text data. It also utilizes an emotion engine to analyze the user's emotional state, including negative emotion detection and positive emotion evaluation.

[0352] Proposal generation

[0353] Based on the analysis results, the server generates service and promotional suggestions tailored to the user. The generation model creates different suggestions depending on the user's emotions and provides information at the optimal time for their emotional state.

[0354] Emotion-based adjustment

[0355] The server adjusts suggestions based on the user's sentiment analysis. For example, it might suggest relaxing services when the user is stressed, or offer special benefits if positive emotions are detected.

[0356] Presentation to the user

[0357] The device presents personalized suggestions to the user. A dashboard that reflects the user's emotional state effectively delivers content that resonates with them. This process reduces unnecessary user interaction, providing a comfortable experience.

[0358] Gathering and adjusting feedback

[0359] Users react to suggestions and provide feedback. This feedback is used as an ongoing data source for the system to improve the suggestions, and the accuracy of the suggestions is readjusted.

[0360] As a concrete example, if the emotion engine detects that a user is feeling stressed, it will suggest a trial of a music streaming service that helps with relaxation. Furthermore, if a user shows interest in a new technology product and their mood is analyzed as positive, it will provide information on purchase benefits related to the latest technology product. In this way, the present invention provides users with an advanced experience and realizes services tailored to their individual needs and emotions.

[0361] The following describes the processing flow.

[0362] Step 1:

[0363] The server collects user behavior data. Specifically, it records the web pages users visit, their online shopping purchase history, and search queries on digital platforms. It also collects feedback data received directly from users, such as inquiries to customer support and survey responses.

[0364] Step 2:

[0365] The server formats the collected data. This process involves cleaning noisy text data, transcribing audio data, and standardizing the data format. This facilitates natural language processing and sentiment analysis.

[0366] Step 3:

[0367] The server uses natural language processing technology to analyze user behavior data and feedback data. This analysis extracts user interests and needs. For example, it identifies frequently occurring keywords and topics to pinpoint content that users are interested in.

[0368] Step 4:

[0369] The server uses an emotion engine to analyze user feedback data and real-time voice and facial expression data to understand emotions. The emotion engine evaluates the user's emotional state as positive, negative, or neutral.

[0370] Step 5:

[0371] The server generates optimal service suggestions for the user based on analyzed interest and emotion data. In this process, a generative AI model creates suggestions that correspond to the user's current emotional state. For example, if it is determined that relaxation is needed, it will suggest relaxing music services.

[0372] Step 6:

[0373] The device presents suggestions to the user. The displayed information is customized to the user's emotional state and clearly organized on the dashboard. The device provides an interactive interface that allows the user to easily accept or skip suggestions.

[0374] Step 7:

[0375] Users react to the suggestions presented. They can click positively, ignore, or provide additional feedback on the suggestions. These reactions are important indicators for evaluating whether the user's needs and feelings are being accurately captured.

[0376] Step 8:

[0377] The server recollects user responses and optimizes the system based on the feedback. The newly collected feedback data is analyzed, and the generative AI model and sentiment engine are adjusted to further improve the accuracy of future suggestions. This continuous analysis and optimization process constantly improves the relevance and usefulness of suggestions to the user.

[0378] (Example 2)

[0379] 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".

[0380] There is a growing need to improve the user experience by more accurately understanding the user's state and providing personalized services based on that understanding. However, current systems do not adequately optimize services based on the user's state and emotions, which is hindering improvements in user satisfaction.

[0381] 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.

[0382] In this invention, the server includes means for collecting user status data, means for analyzing the collected data using processing technology to identify the user's status, and means for generating service content suitable for the user based on the analysis results. This makes it possible to provide more personalized services according to the user's status.

[0383] "User state data" refers to information obtained from a user's actions and reactions, and is used to analyze the user's current emotions and needs.

[0384] "Collecting" refers to the process of gathering information and data for a specific purpose, which then integrates the necessary data.

[0385] "Processing technology" refers to a set of techniques for analyzing and understanding data and generating appropriate output based on a specific purpose.

[0386] "Identifying the user's state" refers to revealing the user's emotions and interests based on collected data and evaluating that situation.

[0387] "Generating service content" means assembling appropriate services and information to provide to users based on the analysis results.

[0388] "Presenting on a display device" means visually showing the generated service content in a user-friendly format.

[0389] "Collecting feedback" is the process of gathering user responses and opinions, and obtaining information to be used for future improvements and service optimization.

[0390] "Adjusting information" is the process of updating service content and proposals based on feedback and data received, with the aim of achieving better results.

[0391] "Analytical techniques" is a general term for methods and tools used to extract and interpret useful information from data.

[0392] "State analysis" is a technique that analyzes the internal and external states of a user, enabling an understanding of their background and influence.

[0393] The system of this invention incorporates technology to accurately understand the user's state and make appropriate suggestions based on that understanding, in order to provide personalized services to the user. Specifically, the server, terminal, and user work together.

[0394] The server takes the lead in collecting user behavior data and feedback data. This includes data obtained from sensors, such as voice tone and facial expressions. The data is then processed using advanced techniques to identify the user's current state. Natural language processing and emotion engines are used in the analysis to specifically extract the user's emotional state and interests. This involves the use of voice recognition software and emotion analysis tools.

[0395] Based on the analysis results, the server generates optimal service content for the user through a generative AI model. Examples include "recommended activities for relaxation" or "purchase benefits for technology products." An example of a prompt used in this process is "make suggestions based on the user's situation." This sentence is used to instruct the generative AI model on specific conditions, resulting in appropriate output according to the user's state.

[0396] The terminal visually displays the generated service details to the user. This utilizes a sophisticated user interface, designed to allow users to easily receive information.

[0397] Finally, the user provides feedback on the information presented. This feedback is sent to the server and used to optimize future service suggestions. In this way, the system reflects the user's state and delivers a more personalized user experience.

[0398] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0399] Step 1:

[0400] The server collects user behavior data and feedback data. Specifically, it obtains voice tone, facial expression data, application usage history, etc., transmitted from the device, from sensors and log databases. The input is the user's digital activity and sensor data, and the output is integrated state data. This makes it possible to comprehensively understand the user's intentions and emotional changes.

[0401] Step 2:

[0402] The server preprocesses the collected data. Specifically, it uses speech recognition software to convert speech data into text and then uses natural language processing techniques to clean the text into a parseable format. The input is integrated state data, and the output includes preprocessed text data. This process makes the data effectively usable by the analysis engine.

[0403] Step 3:

[0404] The server applies natural language processing and sentiment analysis to analyze pre-processed data. The analysis includes processing techniques to extract user interests and needs, and an emotion engine to analyze emotional states. The input is pre-processed text data, and the output includes analysis results indicating the user's state. This allows for a concrete understanding of the user's emotions and interests.

[0405] Step 4:

[0406] The server uses a generative AI model based on the analysis results to generate service content tailored to the user. Specifically, prompts are used to instruct the generative AI model, obtaining individual suggestions based on the user's state. The input consists of the analysis results and prompts, and the output includes personalized service suggestions. This makes it possible to provide content appropriate to the user's state.

[0407] Step 5:

[0408] The server adjusts the generated suggestions based on the user's emotions, and the device displays the content. Specifically, it re-evaluates the user's state data and optimizes the suggestion content and presentation timing. The input consists of personalized suggestions and state data, while the output includes the adjusted suggestions. The device presents the information in a visually easy-to-understand format within the user interface.

[0409] Step 6:

[0410] The system provides users with feedback on the information presented. Specifically, it allows users to easily input their satisfaction level and suggestions for improvement through a user interface on their device, and this feedback is sent to the server. The input is the user's response, and the output includes feedback data. This makes it possible to accumulate data necessary for optimizing the service content in the future.

[0411] (Application Example 2)

[0412] 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."

[0413] Modern content delivery demands suggestions tailored to user interests and needs, but the challenge lies in providing personalized content that also takes into account the user's emotional state. In particular, when selecting visually and aurally engaging content, it is crucial to deliver the appropriate emotional experience at the time when the user will enjoy it the most.

[0414] 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.

[0415] In this invention, the server includes means for collecting user behavior data, feedback data, and emotional data; means for analyzing the collected data using natural language processing technology and emotion analysis technology to identify the user's interests, needs, and emotional state; and means for selecting and automatically generating content optimized for the user based on the analysis results. This makes it possible to provide users with content that is tailored to their emotional state, thereby realizing an enhanced user experience.

[0416] "User behavior data" refers to records of actions and choices that users make on digital platforms.

[0417] "Feedback data" refers to information such as evaluations and comments on suggestions and services provided by users.

[0418] "Emotional data" refers to information that indicates a user's emotional state, obtained from their facial expressions, voice, and other sources.

[0419] "Natural language processing technology" is a technology that uses computers to analyze human language and understand its meaning and intent.

[0420] "Emotion analysis technology" is a technology used to identify and evaluate a user's emotional state from data.

[0421] "Content" refers to the raw materials for information and experiences intended to be provided to users, and includes digital formats such as video, audio, images, and text.

[0422] Personalization is the process of tailoring and delivering experiences based on the individual user's interests, needs, and emotional state.

[0423] An "interface" is a means or method for a user to interact with a system or device.

[0424] The system implementing this invention is designed to collect user behavior data, feedback data, and sentiment data, and to personalize and deliver content based on this data.

[0425] The server collects various user data through smartphones and other devices. Specifically, it uses cameras and microphones to acquire emotional data such as the user's facial expressions and voice tone, and accumulates behavioral data and feedback as a digital operation history. This data is aggregated on a cloud service and processed on high-performance servers equipped with GPUs.

[0426] The server analyzes the collected data using natural language processing toolkits (such as SpaCy) and sentiment analysis APIs (such as Azure's Sentiment Analysis) to identify the user's interests, needs, and even emotional state. Based on this analysis, it selects the most suitable content for the user and generates recommendations. For example, if the server determines that the user is fatigued, it will select relaxing videos to provide emotional satisfaction.

[0427] The device presents recommendations for generated content to the user through its interface. Content that appeals to both visual and auditory senses is displayed at the most appropriate time for the user's environment and emotions, enabling effective entertainment and information delivery.

[0428] Users provide feedback on the presented content, which contributes to generating more precise recommendations in the future. This feedback is used as training data to continuously refine the recommendation strategy.

[0429] For example, if a high school student is tired from studying and wants to listen to music to refresh themselves, the server will perform sentiment analysis and suggest a relaxing music playlist. If the user sends feedback indicating satisfaction with the suggestion, the system will learn similar tendencies and be able to make even more accurate suggestions in the future.

[0430] Example of a prompt:

[0431] Based on the current user's emotional data, suggest content suitable for relaxation. If the user is tired, prioritize selecting videos that include nature sounds.

[0432] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0433] Step 1:

[0434] The server collects facial expression and audio data through the camera and microphone on the user's smartphone. This collects raw data that reflects the user's emotions. At this point, the input is raw video and audio recordings, and the output is the unprocessed data transferred to the server.

[0435] Step 2:

[0436] The server first preprocesses the received raw data. Here, it uses OpenCV to extract facial features from the video data and the Google Speech-to-Text API to convert the audio to text. The input is the raw data from step 1, and the output is the transcribed audio data and the identified facial feature data.

[0437] Step 3:

[0438] The server performs analysis using a natural language processing engine and an emotion analysis engine with pre-processed data. Specifically, it uses Hugging Face's Transformers to analyze the user's interests and emotions from the text and evaluates the emotional state using Azure's Sentiment Analysis API. The input is the text and facial feature data from step 2, and the output is information about the user's interests, emotional state, and emotional intensity.

[0439] Step 4:

[0440] The server runs a model to suggest content best suited to the user's state based on the analysis results. It uses TensorFlow to select content that matches the user's emotional state. The input is the analysis results obtained in step 3, and the output is a personalized content recommendation list for each user.

[0441] Step 5:

[0442] The terminal displays the generated content recommendation list on the user's device screen. The interface is adjusted so that the user can directly experience the content visually and aurally. The input is the recommendation list from step 4, and the output is the specific content presented via the user interface.

[0443] Step 6:

[0444] Users submit feedback on the provided content through an in-app interface. This feedback is data indicating how satisfied the user was with the provided content. The input is the actual content presented, and the output includes user ratings, comments, and satisfaction scores.

[0445] Step 7:

[0446] The server analyzes the feedback data and feeds it to an algorithm that optimizes the next content suggestion based on that analysis. Past feedback stored in MongoDB is used to improve future learning. The input is the feedback data from step 6, and the output is the improved next suggestion strategy.

[0447] 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.

[0448] 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.

[0449] 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.

[0450] [Third Embodiment]

[0451] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0452] 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.

[0453] 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).

[0454] 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.

[0455] 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.

[0456] 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).

[0457] 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.

[0458] 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.

[0459] 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.

[0460] 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.

[0461] 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.

[0462] 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".

[0463] The system of this invention utilizes user behavior data and feedback data, and analyzes them using natural language processing technology to identify user interests and needs, and then proposes services. This system is implemented in the following form.

[0464] Data collection and management

[0465] The server automatically collects each user's behavioral data, such as web browsing history, purchase history, and inquiry history. Furthermore, it manages customer support feedback from users. This data is stored in a database.

[0466] Data preprocessing

[0467] The server prepares the collected data into a format that can be analyzed. This process includes cleaning and normalizing text data and converting it into a format suitable for input to machine learning algorithms.

[0468] Data analysis and insight extraction

[0469] The server uses natural language processing techniques to analyze user behavior data and feedback. Specifically, it extracts relevant topics using topic modeling and performs sentiment analysis to understand the sentiment behind the feedback. This provides insights into identifying user interests and needs.

[0470] Proposal generation

[0471] The server generates service and promotional suggestions tailored to the user based on analysis. This generation process utilizes a generative AI model to create customized suggestions that match the user's areas of interest.

[0472] Presentation to the user

[0473] The device displays suggested content to the user. The system includes a feature that allows the user to review the suggestions on a dashboard the next time they access the system. The suggestions include specific service information and related promotions.

[0474] Utilizing and optimizing feedback

[0475] The server collects user feedback and responses to suggestions, and uses this to improve the accuracy of suggestions. The algorithm is fine-tuned based on feedback, aiming to provide a service that satisfies users.

[0476] For example, if a user has repeatedly searched for smart home appliance-related products in the past, the system can suggest new product announcements and special sale information tailored to their needs. It can also provide additional support information based on support inquiries. In this way, the system enables the provision of detailed services that meet the individual needs of each user.

[0477] The following describes the processing flow.

[0478] Step 1:

[0479] The server periodically collects user behavior data and feedback data. Behavioral data includes web browsing history, purchase history, and search queries, while feedback data includes customer support inquiries and ratings. This data is stored in a database with security measures in place.

[0480] Step 2:

[0481] The server preprocesses the collected data. This includes data cleansing and standardization of data formats. For example, it removes unnecessary characters from text data and converts it into a format suitable for analysis. It also removes noisy data and improves data quality.

[0482] Step 3:

[0483] The server analyzes the data using natural language processing technology. Topic modeling extracts themes of user interest, and sentiment analysis determines whether user feedback is positive or negative. From these analysis results, insights into user interests and needs are obtained.

[0484] Step 4:

[0485] The server generates service suggestions tailored to the user based on the analysis results. The generation model dynamically constructs promotional proposals and related service information for each user, creating customized suggestions.

[0486] Step 5:

[0487] The device presents generated suggestions to the user. Once the user logs into the system, personalized promotions and service suggestions are displayed on the dashboard. If the user is interested, links to access further information and purchase buttons are provided.

[0488] Step 6:

[0489] Users react to the suggestions presented. These reactions are indicated by actions such as clicking, viewing, or skipping. This reaction data is then sent back to the server for further analysis.

[0490] Step 7:

[0491] The server reanalyzes user responses and feedback data to optimize suggestions. Machine learning algorithms are used to analyze the data and adjust the system to provide more appropriate suggestions to each user. This process improves the accuracy of suggestions and enhances the user experience.

[0492] (Example 1)

[0493] 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."

[0494] In today's information society, users expect to be provided with information that is quick and accurate, tailored to their interests and needs. However, conventional systems fail to effectively utilize user behavior and feedback, making it difficult to provide sufficiently personalized suggestions. To solve this problem, there is a need for technology that optimally collects and analyzes user behavior data and feedback, and uses the results to generate highly customized suggestions.

[0495] 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.

[0496] In this invention, the server includes means for collecting behavioral data and feedback information from users and storing it in a data holder; means for normalizing and formatting the collected data using text processing technology and analyzing the data to recognize the user's interests and requests; and means for creating user-optimized service provision information using a generative model based on the analyzed information. This makes it possible to provide information tailored to individual needs based on each user's behavioral patterns and feedback.

[0497] A "user" is an individual or organization that uses an information system or service.

[0498] "Behavioral data" refers to information that shows a user's activities in the digital environment, such as web browsing and purchasing behavior.

[0499] "Feedback information" refers to information that includes opinions and evaluations that users give to a service or product.

[0500] A "data holder" refers to a storage device or database used to store and manage collected data.

[0501] "Text processing technology" refers to techniques for formatting collected text data into a parseable format, including cleaning and normalization.

[0502] "Normalization and formalization" is the process of converting data into a unified format and structuring it.

[0503] A "generative model" is a mathematical model that uses artificial intelligence to generate new data and suggestions.

[0504] "Service provision information" refers to information about services and promotions offered to users.

[0505] "Personalized recommendations" refer to recommendations of information and services that are customized based on the user's characteristics and past behavior.

[0506] The system of the present invention is designed to effectively collect, store, and analyze user behavior data and feedback information, and to provide users with information optimized for their needs. This system is implemented using the hardware and software configuration shown below.

[0507] The server is primarily responsible for data collection, processing, analysis, and suggestion generation. Data and feedback information regarding user behavior are automatically acquired by the server and stored in data holders. At this stage, the server uses a combination of web servers and database software (e.g., MySQL, PostgreSQL). The information stored in the data holders is cleansed and normalized using text processing techniques.

[0508] In data analysis, the server applies natural language processing techniques, specifically performing topic clustering and sentiment evaluation. This process utilizes NLP libraries (e.g., NLTK, spaCy) and machine learning libraries (e.g., TensorFlow, PyTorch). Based on the insights gained from this analysis, the server leverages generative models to generate personalized service offerings for the user. Generative AI models (e.g., GPT series) are used in this generation process. An example of a prompt is given to the AI ​​model: "Based on the user's recent smart home appliance search history, suggest suitable new products and promotions."

[0509] The generated suggestions are sent to the user's device and displayed on the user interface. These devices primarily refer to user interface devices such as PCs, tablets, or smartphones, and the suggested information is presented to the user visually in a dashboard format.

[0510] For example, if a user frequently searches for disaster preparedness supplies, the server analyzes their behavior and suggests the user information on the latest disaster preparedness kits and related campaigns. In this way, users can easily receive personalized information tailored to their interests and needs.

[0511] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0512] Step 1:

[0513] The server collects user behavior data and feedback information and stores it in a data holder. Specifically, it automatically logs pages the user accessed on the website, advertisements clicked, purchase history, and other similar information. Furthermore, feedback that the user sent to customer support is also collected. The input to this process is user behavior and feedback, and the output is a structured dataset.

[0514] Step 2:

[0515] The server preprocesses the collected data using text processing techniques. It removes unnecessary characters, normalizes the data, and standardizes the format from the raw input data. At this stage, morphological analysis is performed to segment the text, and the results are converted into a parseable format. The output is clean data suitable for analysis.

[0516] Step 3:

[0517] The server performs data analysis using natural language processing technology. Specifically, it extracts topics related to user interests using topic clustering and determines the sentiment of feedback through sentiment evaluation. The input for this process is pre-formatted data, and the analysis method outputs insights into the user's interests and needs.

[0518] Step 4:

[0519] Based on the insights gained from the analysis, the server uses a generative AI model to generate service information optimized for the user. In this process, prompt sentences are input to the generative AI model to generate new suggestions. The input consists of the analysis results and prompt sentences, and the output is personalized suggestion information.

[0520] Step 5:

[0521] The terminal visually presents the suggested information sent from the server to the user. Specifically, it displays new service and promotional information in a dashboard format within the user interface. The input to this process is the generated suggested information, and the output is a visual display that the user can view.

[0522] Step 6:

[0523] The server collects user feedback again and analyzes the effectiveness of the generated suggestions based on it. Based on the feedback, it refines the running algorithm and generative model, improving the accuracy of suggestions in future sessions. The input for this step is user feedback, and the output is the improved algorithm and model.

[0524] (Application Example 1)

[0525] 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."

[0526] Effectively presenting products or information tailored to the individual needs of entities on online platforms is a challenging task. In today's world, where diverse data is generated, traditional methods struggle to quickly and accurately capture entities' interests and provide appropriate suggestions. Furthermore, continuous optimization of suggestions through the use of entity feedback is also required.

[0527] 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.

[0528] In this invention, the server includes means for collecting entity behavior data and feedback data, means for analyzing the collected data using analytical techniques to identify the entity's interests and needs, and means for automatically generating product or information suggestions suitable for the entity based on the analysis results. This makes it possible to quickly grasp the individual needs of entities and make highly accurate suggestions.

[0529] An "entity" refers to the object that generates behavioral data and feedback data within the system, and in this invention, it mainly refers to individual users.

[0530] "Behavioral data" refers to data that includes information such as the history of actions performed by an entity on an online platform, browsing history, and choice behaviors.

[0531] "Feedback data" refers to data that includes evaluations, comments, and reactions that entities give to a proposed product or information.

[0532] "Analysis techniques" refer to technical methods used to process collected data and identify the interests and needs of entities, and in this invention, this includes natural language processing techniques.

[0533] "Presenting products or information" refers to the process of selecting products or related information suitable for an entity based on the analysis results and displaying them to the entity.

[0534] "Automatic generation" refers to the process by which a system uses analysis results to create suitable suggestions for entities without human intervention.

[0535] A "display device" refers to a terminal or screen device that has an interface function that allows an entity to visually confirm the proposed content.

[0536] "Evaluations and comments" refer to the act of providing feedback to an entity on a proposal it has received, and are used to improve the quality of the proposal.

[0537] The server uses a cloud infrastructure connected to a database management system to efficiently collect entity behavior data and feedback data. Amazon Web Services (AWS) and Google Cloud Platform are often used for this purpose. The server processes the acquired data via streaming, enabling real-time data collection.

[0538] Next, the server analyzes the collected data using natural language processing technology. This analysis utilizes the Python programming language and its related libraries, spaCy and Gensim. Specifically, Gensim's LDA (Latent Dirichlet Allocation) is used for topic modeling, and the NLTK library's VADER is used for sentiment analysis. Using these, insights tailored to the interests and needs of the entities are extracted.

[0539] Based on the analysis results, the server uses a generative AI model to automatically generate product or information suggestions suitable for the entity. This model is based on Hugging Face's Transformer model and generates customized suggestions based on diverse data. The generated suggestions are formatted in JSON format and sent directly to the terminal.

[0540] On the device, the entity displays product or information generated via the display device. This is implemented using a dynamic web interface built into the application. Frontend libraries such as React and Vue.js are often used for this.

[0541] Users provide feedback on the presented products or information. This feedback is collected by the server and used to improve the algorithm. Specifically, learning-based improvements are made, such as increasing the frequency of suggesting product categories that users have given many positive ratings to.

[0542] For example, if a user has frequently viewed sports equipment in the past, the generative AI model will use that information to suggest the latest sports equipment and sales information to the user via push notifications. An example of a prompt to the generative AI model would be: "Generate promotional information related to sports equipment that the user is interested in. Please include specific product names and discount information, taking into account past purchase history and feedback."

[0543] In this way, based on entity behavior data and feedback data, it becomes possible to perform advanced analysis and automatically generate and present personalized suggestions in real time.

[0544] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0545] Step 1:

[0546] The server uses a cloud infrastructure to collect entity behavior data and feedback data. This data includes browsing history, purchase history, and support inquiries. The collected data is stored in an AWS S3 bucket and then accumulated in a database. The input is user behavior information across the network, and the output is structured data stored in the database.

[0547] Step 2:

[0548] The server preprocesses the collected data. Using Python, it formats the data with the pandas library. Specifically, it cleans the text data and performs normalization to make it suitable for analysis. The input is raw structured data, and the output is a tokenized dataset.

[0549] Step 3:

[0550] The server analyzes data using natural language processing techniques. It uses Gensim's LDA for topic modeling and NLTK's VADER for sentiment analysis. This provides insights into user interests and needs. The input is a pre-processed dataset, and the output is insightful information.

[0551] Step 4:

[0552] The server automatically generates suggestions using a generative AI model. It leverages Hugging Face's transformer model to create customized service and product information tailored to the user's areas of interest. Instructions are given to the model using prompts. Input consists of insight information and prompts, while output is the generated suggestions.

[0553] Step 5:

[0554] The terminal presents the generated suggestions to the user. Specifically, it visually displays the information to the user using a web interface within the application. The dashboard is built using the React library. The input is suggestion data in JSON format, and the output is a visual display for the user.

[0555] Step 6:

[0556] Users provide feedback based on the presented content. This feedback is sent to the server through the application and used to generate future proposals. The feedback input improves the quality of the proposals and enhances the accuracy of the algorithm. The input is user feedback, and the output is the improved proposal model.

[0557] 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.

[0558] This invention is a system that utilizes user behavior data and feedback data, combining natural language processing technology and an emotion engine. This system recognizes user emotions and adjusts suggestions based on those emotions, thereby achieving more personalized service delivery. Its specific form is described below.

[0559] Data collection

[0560] The server collects user behavior data, feedback data, and sensor data (e.g., voice tone, facial recognition results). This allows it to capture the user's intentions and emotional changes in the digital environment.

[0561] Data preprocessing

[0562] The server formats the collected data and converts it into a format suitable for analysis by natural language processing and sentiment engines. This includes cleaning text data and transcribing audio data.

[0563] Natural language processing and sentiment analysis

[0564] The server uses natural language processing technology to extract interests and needs from the user's text data. It also utilizes an emotion engine to analyze the user's emotional state, including negative emotion detection and positive emotion evaluation.

[0565] Proposal generation

[0566] Based on the analysis results, the server generates service and promotional suggestions tailored to the user. The generation model creates different suggestions depending on the user's emotions and provides information at the optimal time for their emotional state.

[0567] Emotion-based adjustment

[0568] The server adjusts suggestions based on the user's sentiment analysis. For example, it might suggest relaxing services when the user is stressed, or offer special benefits if positive emotions are detected.

[0569] Presentation to the user

[0570] The device presents personalized suggestions to the user. A dashboard that reflects the user's emotional state effectively delivers content that resonates with them. This process reduces unnecessary user interaction, providing a comfortable experience.

[0571] Gathering and adjusting feedback

[0572] Users react to suggestions and provide feedback. This feedback is used as an ongoing data source for the system to improve the suggestions, and the accuracy of the suggestions is readjusted.

[0573] As a concrete example, if the emotion engine detects that a user is feeling stressed, it will suggest a trial of a music streaming service that helps with relaxation. Furthermore, if a user shows interest in a new technology product and their mood is analyzed as positive, it will provide information on purchase benefits related to the latest technology product. In this way, the present invention provides users with an advanced experience and realizes services tailored to their individual needs and emotions.

[0574] The following describes the processing flow.

[0575] Step 1:

[0576] The server collects user behavior data. Specifically, it records the web pages users visit, their online shopping purchase history, and search queries on digital platforms. It also collects feedback data received directly from users, such as inquiries to customer support and survey responses.

[0577] Step 2:

[0578] The server formats the collected data. This process involves cleaning noisy text data, transcribing audio data, and standardizing the data format. This facilitates natural language processing and sentiment analysis.

[0579] Step 3:

[0580] The server uses natural language processing technology to analyze user behavior data and feedback data. This analysis extracts user interests and needs. For example, it identifies frequently occurring keywords and topics to pinpoint content that users are interested in.

[0581] Step 4:

[0582] The server uses an emotion engine to analyze user feedback data and real-time voice and facial expression data to understand emotions. The emotion engine evaluates the user's emotional state as positive, negative, or neutral.

[0583] Step 5:

[0584] The server generates optimal service suggestions for the user based on analyzed interest and emotion data. In this process, a generative AI model creates suggestions that correspond to the user's current emotional state. For example, if it is determined that relaxation is needed, it will suggest relaxing music services.

[0585] Step 6:

[0586] The device presents suggestions to the user. The displayed information is customized to the user's emotional state and clearly organized on the dashboard. The device provides an interactive interface that allows the user to easily accept or skip suggestions.

[0587] Step 7:

[0588] Users react to the suggestions presented. They can click positively, ignore, or provide additional feedback on the suggestions. These reactions are important indicators for evaluating whether the user's needs and feelings are being accurately captured.

[0589] Step 8:

[0590] The server recollects user responses and optimizes the system based on the feedback. The newly collected feedback data is analyzed, and the generative AI model and sentiment engine are adjusted to further improve the accuracy of future suggestions. This continuous analysis and optimization process constantly improves the relevance and usefulness of suggestions to the user.

[0591] (Example 2)

[0592] 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."

[0593] There is a growing need to improve the user experience by more accurately understanding the user's state and providing personalized services based on that understanding. However, current systems do not adequately optimize services based on the user's state and emotions, which is hindering improvements in user satisfaction.

[0594] 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.

[0595] In this invention, the server includes means for collecting user status data, means for analyzing the collected data using processing technology to identify the user's status, and means for generating service content suitable for the user based on the analysis results. This makes it possible to provide more personalized services according to the user's status.

[0596] "User state data" refers to information obtained from a user's actions and reactions, and is used to analyze the user's current emotions and needs.

[0597] "Collecting" refers to the process of gathering information and data for a specific purpose, which then integrates the necessary data.

[0598] "Processing technology" refers to a set of techniques for analyzing and understanding data and generating appropriate output based on a specific purpose.

[0599] "Identifying the user's state" refers to revealing the user's emotions and interests based on collected data and evaluating that situation.

[0600] "Generating service content" means assembling appropriate services and information to provide to users based on the analysis results.

[0601] "Presenting on a display device" means visually showing the generated service content in a user-friendly format.

[0602] "Collecting feedback" is the process of gathering user responses and opinions, and obtaining information to be used for future improvements and service optimization.

[0603] "Adjusting information" is the process of updating service content and proposals based on feedback and data received, with the aim of achieving better results.

[0604] "Analytical techniques" is a general term for methods and tools used to extract and interpret useful information from data.

[0605] "State analysis" is a technique that analyzes the internal and external states of a user, enabling an understanding of their background and influence.

[0606] The system of this invention incorporates technology to accurately understand the user's state and make appropriate suggestions based on that understanding, in order to provide personalized services to the user. Specifically, the server, terminal, and user work together.

[0607] The server takes the lead in collecting user behavior data and feedback data. This includes data obtained from sensors, such as voice tone and facial expressions. The data is then processed using advanced techniques to identify the user's current state. Natural language processing and emotion engines are used in the analysis to specifically extract the user's emotional state and interests. This involves the use of voice recognition software and emotion analysis tools.

[0608] Based on the analysis results, the server generates optimal service content for the user through a generative AI model. Examples include "recommended activities for relaxation" or "purchase benefits for technology products." An example of a prompt used in this process is "make suggestions based on the user's situation." This sentence is used to instruct the generative AI model on specific conditions, resulting in appropriate output according to the user's state.

[0609] The terminal visually displays the generated service details to the user. This utilizes a sophisticated user interface, designed to allow users to easily receive information.

[0610] Finally, the user provides feedback on the information presented. This feedback is sent to the server and used to optimize future service suggestions. In this way, the system reflects the user's state and delivers a more personalized user experience.

[0611] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0612] Step 1:

[0613] The server collects user behavior data and feedback data. Specifically, it obtains voice tone, facial expression data, application usage history, etc., transmitted from the device, from sensors and log databases. The input is the user's digital activity and sensor data, and the output is integrated state data. This makes it possible to comprehensively understand the user's intentions and emotional changes.

[0614] Step 2:

[0615] The server preprocesses the collected data. Specifically, it uses speech recognition software to convert speech data into text and then uses natural language processing techniques to clean the text into a parseable format. The input is integrated state data, and the output includes preprocessed text data. This process makes the data effectively usable by the analysis engine.

[0616] Step 3:

[0617] The server applies natural language processing and sentiment analysis to analyze pre-processed data. The analysis includes processing techniques to extract user interests and needs, and an emotion engine to analyze emotional states. The input is pre-processed text data, and the output includes analysis results indicating the user's state. This allows for a concrete understanding of the user's emotions and interests.

[0618] Step 4:

[0619] The server uses a generative AI model based on the analysis results to generate service content tailored to the user. Specifically, prompts are used to instruct the generative AI model, obtaining individual suggestions based on the user's state. The input consists of the analysis results and prompts, and the output includes personalized service suggestions. This makes it possible to provide content appropriate to the user's state.

[0620] Step 5:

[0621] The server adjusts the generated suggestions based on the user's emotions, and the device displays the content. Specifically, it re-evaluates the user's state data and optimizes the suggestion content and presentation timing. The input consists of personalized suggestions and state data, while the output includes the adjusted suggestions. The device presents the information in a visually easy-to-understand format within the user interface.

[0622] Step 6:

[0623] The system provides users with feedback on the information presented. Specifically, it allows users to easily input their satisfaction level and suggestions for improvement through a user interface on their device, and this feedback is sent to the server. The input is the user's response, and the output includes feedback data. This makes it possible to accumulate data necessary for optimizing the service content in the future.

[0624] (Application Example 2)

[0625] 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."

[0626] Modern content delivery demands suggestions tailored to user interests and needs, but the challenge lies in providing personalized content that also takes into account the user's emotional state. In particular, when selecting visually and aurally engaging content, it is crucial to deliver the appropriate emotional experience at the time when the user will enjoy it the most.

[0627] 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.

[0628] In this invention, the server includes means for collecting user behavior data, feedback data, and emotional data; means for analyzing the collected data using natural language processing technology and emotion analysis technology to identify the user's interests, needs, and emotional state; and means for selecting and automatically generating content optimized for the user based on the analysis results. This makes it possible to provide users with content that is tailored to their emotional state, thereby realizing an enhanced user experience.

[0629] "User behavior data" refers to records of actions and choices that users make on digital platforms.

[0630] "Feedback data" refers to information such as evaluations and comments on suggestions and services provided by users.

[0631] "Emotional data" refers to information that indicates a user's emotional state, obtained from their facial expressions, voice, and other sources.

[0632] "Natural language processing technology" is a technology that uses computers to analyze human language and understand its meaning and intent.

[0633] "Emotion analysis technology" is a technology used to identify and evaluate a user's emotional state from data.

[0634] "Content" refers to the raw materials for information and experiences intended to be provided to users, and includes digital formats such as video, audio, images, and text.

[0635] Personalization is the process of tailoring and delivering experiences based on the individual user's interests, needs, and emotional state.

[0636] An "interface" is a means or method for a user to interact with a system or device.

[0637] The system implementing this invention is designed to collect user behavior data, feedback data, and sentiment data, and to personalize and deliver content based on this data.

[0638] The server collects various user data through smartphones and other devices. Specifically, it uses cameras and microphones to acquire emotional data such as the user's facial expressions and voice tone, and accumulates behavioral data and feedback as a digital operation history. This data is aggregated on a cloud service and processed on high-performance servers equipped with GPUs.

[0639] The server analyzes the collected data using natural language processing toolkits (such as SpaCy) and sentiment analysis APIs (such as Azure's Sentiment Analysis) to identify the user's interests, needs, and even emotional state. Based on this analysis, it selects the most suitable content for the user and generates recommendations. For example, if the server determines that the user is fatigued, it will select relaxing videos to provide emotional satisfaction.

[0640] The device presents recommendations for generated content to the user through its interface. Content that appeals to both visual and auditory senses is displayed at the most appropriate time for the user's environment and emotions, enabling effective entertainment and information delivery.

[0641] Users provide feedback on the presented content, which contributes to generating more precise recommendations in the future. This feedback is used as training data to continuously refine the recommendation strategy.

[0642] For example, if a high school student is tired from studying and wants to listen to music to refresh themselves, the server will perform sentiment analysis and suggest a relaxing music playlist. If the user sends feedback indicating satisfaction with the suggestion, the system will learn similar tendencies and be able to make even more accurate suggestions in the future.

[0643] Example of a prompt:

[0644] Based on the current user's emotional data, suggest content suitable for relaxation. If the user is tired, prioritize selecting videos that include nature sounds.

[0645] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0646] Step 1:

[0647] The server collects facial expression and audio data through the camera and microphone on the user's smartphone. This collects raw data that reflects the user's emotions. At this point, the input is raw video and audio recordings, and the output is the unprocessed data transferred to the server.

[0648] Step 2:

[0649] The server first preprocesses the received raw data. Here, it uses OpenCV to extract facial features from the video data and the Google Speech-to-Text API to convert the audio to text. The input is the raw data from step 1, and the output is the transcribed audio data and the identified facial feature data.

[0650] Step 3:

[0651] The server performs analysis using a natural language processing engine and an emotion analysis engine with pre-processed data. Specifically, it uses Hugging Face's Transformers to analyze the user's interests and emotions from the text and evaluates the emotional state using Azure's Sentiment Analysis API. The input is the text and facial feature data from step 2, and the output is information about the user's interests, emotional state, and emotional intensity.

[0652] Step 4:

[0653] The server runs a model to suggest content best suited to the user's state based on the analysis results. It uses TensorFlow to select content that matches the user's emotional state. The input is the analysis results obtained in step 3, and the output is a personalized content recommendation list for each user.

[0654] Step 5:

[0655] The terminal displays the generated content recommendation list on the user's device screen. The interface is adjusted so that the user can directly experience the content visually and aurally. The input is the recommendation list from step 4, and the output is the specific content presented via the user interface.

[0656] Step 6:

[0657] Users submit feedback on the provided content through an in-app interface. This feedback is data indicating how satisfied the user was with the provided content. The input is the actual content presented, and the output includes user ratings, comments, and satisfaction scores.

[0658] Step 7:

[0659] The server analyzes the feedback data and feeds it to an algorithm that optimizes the next content suggestion based on that analysis. Past feedback stored in MongoDB is used to improve future learning. The input is the feedback data from step 6, and the output is the improved next suggestion strategy.

[0660] 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.

[0661] 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.

[0662] 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.

[0663] [Fourth Embodiment]

[0664] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0665] 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.

[0666] 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).

[0667] 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.

[0668] 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.

[0669] 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).

[0670] 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.

[0671] 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.

[0672] 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.

[0673] 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.

[0674] 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.

[0675] 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.

[0676] 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".

[0677] The system of this invention utilizes user behavior data and feedback data, and analyzes them using natural language processing technology to identify user interests and needs, and then proposes services. This system is implemented in the following form.

[0678] Data collection and management

[0679] The server automatically collects each user's behavioral data, such as web browsing history, purchase history, and inquiry history. Furthermore, it manages customer support feedback from users. This data is stored in a database.

[0680] Data preprocessing

[0681] The server prepares the collected data into a format that can be analyzed. This process includes cleaning and normalizing text data and converting it into a format suitable for input to machine learning algorithms.

[0682] Data analysis and insight extraction

[0683] The server uses natural language processing techniques to analyze user behavior data and feedback. Specifically, it extracts relevant topics using topic modeling and performs sentiment analysis to understand the sentiment behind the feedback. This provides insights into identifying user interests and needs.

[0684] Proposal generation

[0685] The server generates service and promotional suggestions tailored to the user based on analysis. This generation process utilizes a generative AI model to create customized suggestions that match the user's areas of interest.

[0686] Presentation to the user

[0687] The device displays suggested content to the user. The system includes a feature that allows the user to review the suggestions on a dashboard the next time they access the system. The suggestions include specific service information and related promotions.

[0688] Utilizing and optimizing feedback

[0689] The server collects user feedback and responses to suggestions, and uses this to improve the accuracy of suggestions. The algorithm is fine-tuned based on feedback, aiming to provide a service that satisfies users.

[0690] For example, if a user has repeatedly searched for smart home appliance-related products in the past, the system can suggest new product announcements and special sale information tailored to their needs. It can also provide additional support information based on support inquiries. In this way, the system enables the provision of detailed services that meet the individual needs of each user.

[0691] The following describes the processing flow.

[0692] Step 1:

[0693] The server periodically collects user behavior data and feedback data. Behavioral data includes web browsing history, purchase history, and search queries, while feedback data includes customer support inquiries and ratings. This data is stored in a database with security measures in place.

[0694] Step 2:

[0695] The server preprocesses the collected data. This includes data cleansing and standardization of data formats. For example, it removes unnecessary characters from text data and converts it into a format suitable for analysis. It also removes noisy data and improves data quality.

[0696] Step 3:

[0697] The server analyzes the data using natural language processing technology. Topic modeling extracts themes of user interest, and sentiment analysis determines whether user feedback is positive or negative. From these analysis results, insights into user interests and needs are obtained.

[0698] Step 4:

[0699] The server generates service suggestions tailored to the user based on the analysis results. The generation model dynamically constructs promotional proposals and related service information for each user, creating customized suggestions.

[0700] Step 5:

[0701] The device presents generated suggestions to the user. Once the user logs into the system, personalized promotions and service suggestions are displayed on the dashboard. If the user is interested, links to access further information and purchase buttons are provided.

[0702] Step 6:

[0703] Users react to the suggestions presented. These reactions are indicated by actions such as clicking, viewing, or skipping. This reaction data is then sent back to the server for further analysis.

[0704] Step 7:

[0705] The server reanalyzes user responses and feedback data to optimize suggestions. Machine learning algorithms are used to analyze the data and adjust the system to provide more appropriate suggestions to each user. This process improves the accuracy of suggestions and enhances the user experience.

[0706] (Example 1)

[0707] 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".

[0708] In today's information society, users expect to be provided with information that is quick and accurate, tailored to their interests and needs. However, conventional systems fail to effectively utilize user behavior and feedback, making it difficult to provide sufficiently personalized suggestions. To solve this problem, there is a need for technology that optimally collects and analyzes user behavior data and feedback, and uses the results to generate highly customized suggestions.

[0709] 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.

[0710] In this invention, the server includes means for collecting behavioral data and feedback information from users and storing it in a data holder; means for normalizing and formatting the collected data using text processing technology and analyzing the data to recognize the user's interests and requests; and means for creating user-optimized service provision information using a generative model based on the analyzed information. This makes it possible to provide information tailored to individual needs based on each user's behavioral patterns and feedback.

[0711] A "user" is an individual or organization that uses an information system or service.

[0712] "Behavioral data" refers to information that shows a user's activities in the digital environment, such as web browsing and purchasing behavior.

[0713] "Feedback information" refers to information that includes opinions and evaluations that users give to a service or product.

[0714] A "data holder" refers to a storage device or database used to store and manage collected data.

[0715] "Text processing technology" refers to techniques for formatting collected text data into a parseable format, including cleaning and normalization.

[0716] "Normalization and formalization" is the process of converting data into a unified format and structuring it.

[0717] A "generative model" is a mathematical model that uses artificial intelligence to generate new data and suggestions.

[0718] "Service provision information" refers to information about services and promotions offered to users.

[0719] "Personalized recommendations" refer to recommendations of information and services that are customized based on the user's characteristics and past behavior.

[0720] The system of the present invention is designed to effectively collect, store, and analyze user behavior data and feedback information, and to provide users with information optimized for their needs. This system is implemented using the hardware and software configuration shown below.

[0721] The server is primarily responsible for data collection, processing, analysis, and suggestion generation. Data and feedback information regarding user behavior are automatically acquired by the server and stored in data holders. At this stage, the server uses a combination of web servers and database software (e.g., MySQL, PostgreSQL). The information stored in the data holders is cleansed and normalized using text processing techniques.

[0722] In data analysis, the server applies natural language processing techniques, specifically performing topic clustering and sentiment evaluation. This process utilizes NLP libraries (e.g., NLTK, spaCy) and machine learning libraries (e.g., TensorFlow, PyTorch). Based on the insights gained from this analysis, the server leverages generative models to generate personalized service offerings for the user. Generative AI models (e.g., GPT series) are used in this generation process. An example of a prompt is given to the AI ​​model: "Based on the user's recent smart home appliance search history, suggest suitable new products and promotions."

[0723] The generated suggestions are sent to the user's device and displayed on the user interface. These devices primarily refer to user interface devices such as PCs, tablets, or smartphones, and the suggested information is presented to the user visually in a dashboard format.

[0724] For example, if a user frequently searches for disaster preparedness supplies, the server analyzes their behavior and suggests the user information on the latest disaster preparedness kits and related campaigns. In this way, users can easily receive personalized information tailored to their interests and needs.

[0725] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0726] Step 1:

[0727] The server collects user behavior data and feedback information and stores it in a data holder. Specifically, it automatically logs pages the user accessed on the website, advertisements clicked, purchase history, and other similar information. Furthermore, feedback that the user sent to customer support is also collected. The input to this process is user behavior and feedback, and the output is a structured dataset.

[0728] Step 2:

[0729] The server preprocesses the collected data using text processing techniques. It removes unnecessary characters, normalizes the data, and standardizes the format from the raw input data. At this stage, morphological analysis is performed to segment the text, and the results are converted into a parseable format. The output is clean data suitable for analysis.

[0730] Step 3:

[0731] The server performs data analysis using natural language processing technology. Specifically, it extracts topics related to user interests using topic clustering and determines the sentiment of feedback through sentiment evaluation. The input for this process is pre-formatted data, and the analysis method outputs insights into the user's interests and needs.

[0732] Step 4:

[0733] Based on the insights gained from the analysis, the server uses a generative AI model to generate service information optimized for the user. In this process, prompt sentences are input to the generative AI model to generate new suggestions. The input consists of the analysis results and prompt sentences, and the output is personalized suggestion information.

[0734] Step 5:

[0735] The terminal visually presents the suggested information sent from the server to the user. Specifically, it displays new service and promotional information in a dashboard format within the user interface. The input to this process is the generated suggested information, and the output is a visual display that the user can view.

[0736] Step 6:

[0737] The server collects user feedback again and analyzes the effectiveness of the generated suggestions based on it. Based on the feedback, it refines the running algorithm and generative model, improving the accuracy of suggestions in future sessions. The input for this step is user feedback, and the output is the improved algorithm and model.

[0738] (Application Example 1)

[0739] 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".

[0740] Effectively presenting products or information tailored to the individual needs of entities on online platforms is a challenging task. In today's world, where diverse data is generated, traditional methods struggle to quickly and accurately capture entities' interests and provide appropriate suggestions. Furthermore, continuous optimization of suggestions through the use of entity feedback is also required.

[0741] 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.

[0742] In this invention, the server includes means for collecting entity behavior data and feedback data, means for analyzing the collected data using analytical techniques to identify the entity's interests and needs, and means for automatically generating product or information suggestions suitable for the entity based on the analysis results. This makes it possible to quickly grasp the individual needs of entities and make highly accurate suggestions.

[0743] An "entity" refers to the object that generates behavioral data and feedback data within the system, and in this invention, it mainly refers to individual users.

[0744] "Behavioral data" refers to data that includes information such as the history of actions performed by an entity on an online platform, browsing history, and choice behaviors.

[0745] "Feedback data" refers to data that includes evaluations, comments, and reactions that entities give to a proposed product or information.

[0746] "Analysis techniques" refer to technical methods used to process collected data and identify the interests and needs of entities, and in this invention, this includes natural language processing techniques.

[0747] "Presenting products or information" refers to the process of selecting products or related information suitable for an entity based on the analysis results and displaying them to the entity.

[0748] "Automatic generation" refers to the process by which a system uses analysis results to create suitable suggestions for entities without human intervention.

[0749] A "display device" refers to a terminal or screen device that has an interface function that allows an entity to visually confirm the proposed content.

[0750] "Evaluations and comments" refer to the act of providing feedback to an entity on a proposal it has received, and are used to improve the quality of the proposal.

[0751] The server uses a cloud infrastructure connected to a database management system to efficiently collect entity behavior data and feedback data. Amazon Web Services (AWS) and Google Cloud Platform are often used for this purpose. The server processes the acquired data via streaming, enabling real-time data collection.

[0752] Next, the server analyzes the collected data using natural language processing technology. This analysis utilizes the Python programming language and its related libraries, spaCy and Gensim. Specifically, Gensim's LDA (Latent Dirichlet Allocation) is used for topic modeling, and the NLTK library's VADER is used for sentiment analysis. Using these, insights tailored to the interests and needs of the entities are extracted.

[0753] Based on the analysis results, the server uses a generative AI model to automatically generate product or information suggestions suitable for the entity. This model is based on Hugging Face's Transformer model and generates customized suggestions based on diverse data. The generated suggestions are formatted in JSON format and sent directly to the terminal.

[0754] On the device, the entity displays product or information generated via the display device. This is implemented using a dynamic web interface built into the application. Frontend libraries such as React and Vue.js are often used for this.

[0755] Users provide feedback on the presented products or information. This feedback is collected by the server and used to improve the algorithm. Specifically, learning-based improvements are made, such as increasing the frequency of suggesting product categories that users have given many positive ratings to.

[0756] For example, if a user has frequently viewed sports equipment in the past, the generative AI model will use that information to suggest the latest sports equipment and sales information to the user via push notifications. An example of a prompt to the generative AI model would be: "Generate promotional information related to sports equipment that the user is interested in. Please include specific product names and discount information, taking into account past purchase history and feedback."

[0757] In this way, based on entity behavior data and feedback data, it becomes possible to perform advanced analysis and automatically generate and present personalized suggestions in real time.

[0758] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0759] Step 1:

[0760] The server uses a cloud infrastructure to collect entity behavior data and feedback data. This data includes browsing history, purchase history, and support inquiries. The collected data is stored in an AWS S3 bucket and then accumulated in a database. The input is user behavior information across the network, and the output is structured data stored in the database.

[0761] Step 2:

[0762] The server preprocesses the collected data. Using Python, it formats the data with the pandas library. Specifically, it cleans the text data and performs normalization to make it suitable for analysis. The input is raw structured data, and the output is a tokenized dataset.

[0763] Step 3:

[0764] The server analyzes data using natural language processing techniques. It uses Gensim's LDA for topic modeling and NLTK's VADER for sentiment analysis. This provides insights into user interests and needs. The input is a pre-processed dataset, and the output is insightful information.

[0765] Step 4:

[0766] The server automatically generates suggestions using a generative AI model. It leverages Hugging Face's transformer model to create customized service and product information tailored to the user's areas of interest. Instructions are given to the model using prompts. Input consists of insight information and prompts, while output is the generated suggestions.

[0767] Step 5:

[0768] The terminal presents the generated suggestions to the user. Specifically, it visually displays the information to the user using a web interface within the application. The dashboard is built using the React library. The input is suggestion data in JSON format, and the output is a visual display for the user.

[0769] Step 6:

[0770] Users provide feedback based on the presented content. This feedback is sent to the server through the application and used to generate future proposals. The feedback input improves the quality of the proposals and enhances the accuracy of the algorithm. The input is user feedback, and the output is the improved proposal model.

[0771] 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.

[0772] This invention is a system that utilizes user behavior data and feedback data, combining natural language processing technology and an emotion engine. This system recognizes user emotions and adjusts suggestions based on those emotions, thereby achieving more personalized service delivery. Its specific form is described below.

[0773] Data collection

[0774] The server collects user behavior data, feedback data, and sensor data (e.g., voice tone, facial recognition results). This allows it to capture the user's intentions and emotional changes in the digital environment.

[0775] Data preprocessing

[0776] The server formats the collected data and converts it into a format suitable for analysis by natural language processing and sentiment engines. This includes cleaning text data and transcribing audio data.

[0777] Natural language processing and sentiment analysis

[0778] The server uses natural language processing technology to extract interests and needs from the user's text data. It also utilizes an emotion engine to analyze the user's emotional state, including negative emotion detection and positive emotion evaluation.

[0779] Proposal generation

[0780] Based on the analysis results, the server generates service and promotional suggestions tailored to the user. The generation model creates different suggestions depending on the user's emotions and provides information at the optimal time for their emotional state.

[0781] Emotion-based adjustment

[0782] The server adjusts suggestions based on the user's sentiment analysis. For example, it might suggest relaxing services when the user is stressed, or offer special benefits if positive emotions are detected.

[0783] Presentation to the user

[0784] The device presents personalized suggestions to the user. A dashboard that reflects the user's emotional state effectively delivers content that resonates with them. This process reduces unnecessary user interaction, providing a comfortable experience.

[0785] Gathering and adjusting feedback

[0786] Users react to suggestions and provide feedback. This feedback is used as an ongoing data source for the system to improve the suggestions, and the accuracy of the suggestions is readjusted.

[0787] As a concrete example, if the emotion engine detects that a user is feeling stressed, it will suggest a trial of a music streaming service that helps with relaxation. Furthermore, if a user shows interest in a new technology product and their mood is analyzed as positive, it will provide information on purchase benefits related to the latest technology product. In this way, the present invention provides users with an advanced experience and realizes services tailored to their individual needs and emotions.

[0788] The following describes the processing flow.

[0789] Step 1:

[0790] The server collects user behavior data. Specifically, it records the web pages users visit, their online shopping purchase history, and search queries on digital platforms. It also collects feedback data received directly from users, such as inquiries to customer support and survey responses.

[0791] Step 2:

[0792] The server formats the collected data. This process involves cleaning noisy text data, transcribing audio data, and standardizing the data format. This facilitates natural language processing and sentiment analysis.

[0793] Step 3:

[0794] The server uses natural language processing technology to analyze user behavior data and feedback data. This analysis extracts user interests and needs. For example, it identifies frequently occurring keywords and topics to pinpoint content that users are interested in.

[0795] Step 4:

[0796] The server uses an emotion engine to analyze user feedback data and real-time voice and facial expression data to understand emotions. The emotion engine evaluates the user's emotional state as positive, negative, or neutral.

[0797] Step 5:

[0798] The server generates optimal service suggestions for the user based on analyzed interest and emotion data. In this process, a generative AI model creates suggestions that correspond to the user's current emotional state. For example, if it is determined that relaxation is needed, it will suggest relaxing music services.

[0799] Step 6:

[0800] The device presents suggestions to the user. The displayed information is customized to the user's emotional state and clearly organized on the dashboard. The device provides an interactive interface that allows the user to easily accept or skip suggestions.

[0801] Step 7:

[0802] Users react to the suggestions presented. They can click positively, ignore, or provide additional feedback on the suggestions. These reactions are important indicators for evaluating whether the user's needs and feelings are being accurately captured.

[0803] Step 8:

[0804] The server recollects user responses and optimizes the system based on the feedback. The newly collected feedback data is analyzed, and the generative AI model and sentiment engine are adjusted to further improve the accuracy of future suggestions. This continuous analysis and optimization process constantly improves the relevance and usefulness of suggestions to the user.

[0805] (Example 2)

[0806] 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".

[0807] There is a growing need to improve the user experience by more accurately understanding the user's state and providing personalized services based on that understanding. However, current systems do not adequately optimize services based on the user's state and emotions, which is hindering improvements in user satisfaction.

[0808] 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.

[0809] In this invention, the server includes means for collecting user status data, means for analyzing the collected data using processing technology to identify the user's status, and means for generating service content suitable for the user based on the analysis results. This makes it possible to provide more personalized services according to the user's status.

[0810] "User state data" refers to information obtained from a user's actions and reactions, and is used to analyze the user's current emotions and needs.

[0811] "Collecting" refers to the process of gathering information and data for a specific purpose, which then integrates the necessary data.

[0812] "Processing technology" refers to a set of techniques for analyzing and understanding data and generating appropriate output based on a specific purpose.

[0813] "Identifying the user's state" refers to revealing the user's emotions and interests based on collected data and evaluating that situation.

[0814] "Generating service content" means assembling appropriate services and information to provide to users based on the analysis results.

[0815] "Presenting on a display device" means visually showing the generated service content in a user-friendly format.

[0816] "Collecting feedback" is the process of gathering user responses and opinions, and obtaining information to be used for future improvements and service optimization.

[0817] "Adjusting information" is the process of updating service content and proposals based on feedback and data received, with the aim of achieving better results.

[0818] "Analytical techniques" is a general term for methods and tools used to extract and interpret useful information from data.

[0819] "State analysis" is a technique that analyzes the internal and external states of a user, enabling an understanding of their background and influence.

[0820] The system of this invention incorporates technology to accurately understand the user's state and make appropriate suggestions based on that understanding, in order to provide personalized services to the user. Specifically, the server, terminal, and user work together.

[0821] The server takes the lead in collecting user behavior data and feedback data. This includes data obtained from sensors, such as voice tone and facial expressions. The data is then processed using advanced techniques to identify the user's current state. Natural language processing and emotion engines are used in the analysis to specifically extract the user's emotional state and interests. This involves the use of voice recognition software and emotion analysis tools.

[0822] Based on the analysis results, the server generates optimal service content for the user through a generative AI model. Examples include "recommended activities for relaxation" or "purchase benefits for technology products." An example of a prompt used in this process is "make suggestions based on the user's situation." This sentence is used to instruct the generative AI model on specific conditions, resulting in appropriate output according to the user's state.

[0823] The terminal visually displays the generated service details to the user. This utilizes a sophisticated user interface, designed to allow users to easily receive information.

[0824] Finally, the user provides feedback on the information presented. This feedback is sent to the server and used to optimize future service suggestions. In this way, the system reflects the user's state and delivers a more personalized user experience.

[0825] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0826] Step 1:

[0827] The server collects user behavior data and feedback data. Specifically, it obtains voice tone, facial expression data, application usage history, etc., transmitted from the device, from sensors and log databases. The input is the user's digital activity and sensor data, and the output is integrated state data. This makes it possible to comprehensively understand the user's intentions and emotional changes.

[0828] Step 2:

[0829] The server preprocesses the collected data. Specifically, it uses speech recognition software to convert speech data into text and then uses natural language processing techniques to clean the text into a parseable format. The input is integrated state data, and the output includes preprocessed text data. This process makes the data effectively usable by the analysis engine.

[0830] Step 3:

[0831] The server applies natural language processing and sentiment analysis to analyze pre-processed data. The analysis includes processing techniques to extract user interests and needs, and an emotion engine to analyze emotional states. The input is pre-processed text data, and the output includes analysis results indicating the user's state. This allows for a concrete understanding of the user's emotions and interests.

[0832] Step 4:

[0833] The server uses a generative AI model based on the analysis results to generate service content tailored to the user. Specifically, prompts are used to instruct the generative AI model, obtaining individual suggestions based on the user's state. The input consists of the analysis results and prompts, and the output includes personalized service suggestions. This makes it possible to provide content appropriate to the user's state.

[0834] Step 5:

[0835] The server adjusts the generated suggestions based on the user's emotions, and the device displays the content. Specifically, it re-evaluates the user's state data and optimizes the suggestion content and presentation timing. The input consists of personalized suggestions and state data, while the output includes the adjusted suggestions. The device presents the information in a visually easy-to-understand format within the user interface.

[0836] Step 6:

[0837] The system provides users with feedback on the information presented. Specifically, it allows users to easily input their satisfaction level and suggestions for improvement through a user interface on their device, and this feedback is sent to the server. The input is the user's response, and the output includes feedback data. This makes it possible to accumulate data necessary for optimizing the service content in the future.

[0838] (Application Example 2)

[0839] 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".

[0840] Modern content delivery demands suggestions tailored to user interests and needs, but the challenge lies in providing personalized content that also takes into account the user's emotional state. In particular, when selecting visually and aurally engaging content, it is crucial to deliver the appropriate emotional experience at the time when the user will enjoy it the most.

[0841] 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.

[0842] In this invention, the server includes means for collecting user behavior data, feedback data, and emotional data; means for analyzing the collected data using natural language processing technology and emotion analysis technology to identify the user's interests, needs, and emotional state; and means for selecting and automatically generating content optimized for the user based on the analysis results. This makes it possible to provide users with content that is tailored to their emotional state, thereby realizing an enhanced user experience.

[0843] "User behavior data" refers to records of actions and choices that users make on digital platforms.

[0844] "Feedback data" refers to information such as evaluations and comments on suggestions and services provided by users.

[0845] "Emotional data" refers to information that indicates a user's emotional state, obtained from their facial expressions, voice, and other sources.

[0846] "Natural language processing technology" is a technology that uses computers to analyze human language and understand its meaning and intent.

[0847] "Emotion analysis technology" is a technology used to identify and evaluate a user's emotional state from data.

[0848] "Content" refers to the raw materials for information and experiences intended to be provided to users, and includes digital formats such as video, audio, images, and text.

[0849] Personalization is the process of tailoring and delivering experiences based on the individual user's interests, needs, and emotional state.

[0850] An "interface" is a means or method for a user to interact with a system or device.

[0851] The system implementing this invention is designed to collect user behavior data, feedback data, and sentiment data, and to personalize and deliver content based on this data.

[0852] The server collects various user data through smartphones and other devices. Specifically, it uses cameras and microphones to acquire emotional data such as the user's facial expressions and voice tone, and accumulates behavioral data and feedback as a digital operation history. This data is aggregated on a cloud service and processed on high-performance servers equipped with GPUs.

[0853] The server analyzes the collected data using natural language processing toolkits (such as SpaCy) and sentiment analysis APIs (such as Azure's Sentiment Analysis) to identify the user's interests, needs, and even emotional state. Based on this analysis, it selects the most suitable content for the user and generates recommendations. For example, if the server determines that the user is fatigued, it will select relaxing videos to provide emotional satisfaction.

[0854] The device presents recommendations for generated content to the user through its interface. Content that appeals to both visual and auditory senses is displayed at the most appropriate time for the user's environment and emotions, enabling effective entertainment and information delivery.

[0855] Users provide feedback on the presented content, which contributes to generating more precise recommendations in the future. This feedback is used as training data to continuously refine the recommendation strategy.

[0856] For example, if a high school student is tired from studying and wants to listen to music to refresh themselves, the server will perform sentiment analysis and suggest a relaxing music playlist. If the user sends feedback indicating satisfaction with the suggestion, the system will learn similar tendencies and be able to make even more accurate suggestions in the future.

[0857] Example of a prompt:

[0858] Based on the current user's emotional data, suggest content suitable for relaxation. If the user is tired, prioritize selecting videos that include nature sounds.

[0859] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0860] Step 1:

[0861] The server collects facial expression and audio data through the camera and microphone on the user's smartphone. This collects raw data that reflects the user's emotions. At this point, the input is raw video and audio recordings, and the output is the unprocessed data transferred to the server.

[0862] Step 2:

[0863] The server first preprocesses the received raw data. Here, it uses OpenCV to extract facial features from the video data and the Google Speech-to-Text API to convert the audio to text. The input is the raw data from step 1, and the output is the transcribed audio data and the identified facial feature data.

[0864] Step 3:

[0865] The server performs analysis using a natural language processing engine and an emotion analysis engine with pre-processed data. Specifically, it uses Hugging Face's Transformers to analyze the user's interests and emotions from the text and evaluates the emotional state using Azure's Sentiment Analysis API. The input is the text and facial feature data from step 2, and the output is information about the user's interests, emotional state, and emotional intensity.

[0866] Step 4:

[0867] The server runs a model to suggest content best suited to the user's state based on the analysis results. It uses TensorFlow to select content that matches the user's emotional state. The input is the analysis results obtained in step 3, and the output is a personalized content recommendation list for each user.

[0868] Step 5:

[0869] The terminal displays the generated content recommendation list on the user's device screen. The interface is adjusted so that the user can directly experience the content visually and aurally. The input is the recommendation list from step 4, and the output is the specific content presented via the user interface.

[0870] Step 6:

[0871] Users submit feedback on the provided content through an in-app interface. This feedback is data indicating how satisfied the user was with the provided content. The input is the actual content presented, and the output includes user ratings, comments, and satisfaction scores.

[0872] Step 7:

[0873] The server analyzes the feedback data and feeds it to an algorithm that optimizes the next content suggestion based on that analysis. Past feedback stored in MongoDB is used to improve future learning. The input is the feedback data from step 6, and the output is the improved next suggestion strategy.

[0874] 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.

[0875] 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.

[0876] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0877] 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.

[0878] 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.

[0879] 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.

[0880] 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.

[0881] 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.

[0882] 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."

[0883] 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.

[0884] 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.

[0885] 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.

[0886] 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.

[0887] 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.

[0888] 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.

[0889] 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.

[0890] 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.

[0891] 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.

[0892] 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.

[0893] 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.

[0894] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted as being incorporated by reference.

[0895] The following is further disclosed regarding the embodiments described above.

[0896] (Claim 1)

[0897] Means for collecting user behavior data and feedback data,

[0898] A means of identifying user interests and needs by analyzing collected data using natural language processing technology,

[0899] A means for automatically generating service suggestions suitable for the user based on analysis results,

[0900] A means of presenting the generated suggestions to the user interface,

[0901] A system that includes this.

[0902] (Claim 2)

[0903] The system according to claim 1, further comprising means for collecting user feedback on generated suggestions and optimizing the content of the suggestions based on that feedback.

[0904] (Claim 3)

[0905] The system according to claim 1, comprising means for applying topic modeling and sentiment analysis as natural language processing techniques.

[0906] "Example 1"

[0907] (Claim 1)

[0908] A means of collecting behavioral data and feedback information from users and storing it in a data holder,

[0909] By normalizing and formatting the collected data using text processing technology, and then analyzing that data, a means of recognizing user interests and needs is provided.

[0910] A means of creating user-optimized service provision information using a generative model based on the analyzed information,

[0911] A means of displaying the created information on the user's device screen,

[0912] A system that includes this.

[0913] (Claim 2)

[0914] The system according to claim 1, which collects user feedback information and uses that information to improve service provision information.

[0915] (Claim 3)

[0916] The system according to claim 1, which uses topic clustering and sentiment assessment as text processing techniques.

[0917] "Application Example 1"

[0918] (Claim 1)

[0919] Means for collecting entity behavior data and feedback data,

[0920] A means of analyzing collected data using analytical techniques to identify the interests and needs of entities,

[0921] A means for automatically generating product or information presentations suitable for an entity based on analysis results,

[0922] A means for presenting the generated presentation on the entity's display device,

[0923] A means to collect evaluations and comments on the displayed content of entities and reflect them in future automated generation,

[0924] A means for dynamically generating a customized interface when presenting generated products or information to an entity by display means,

[0925] A system that includes this.

[0926] (Claim 2)

[0927] The system according to claim 1, which collects an entity's evaluation of a generated product or information and optimizes the content of the product or information based on that evaluation.

[0928] (Claim 3)

[0929] The system according to claim 1, which applies subject modeling or sentiment analysis as analytical techniques.

[0930] "Example 2 of combining an emotion engine"

[0931] (Claim 1)

[0932] Means for collecting user status data,

[0933] A means of analyzing the collected data using processing technology to identify the user's state,

[0934] A means for generating service content suitable for the user based on the analysis results,

[0935] A means for presenting the generated content on the user's display device,

[0936] A system that includes this.

[0937] (Claim 2)

[0938] The system according to claim 1, which collects user reactions to generated content and adjusts the content based on that information.

[0939] (Claim 3)

[0940] The system according to claim 1, wherein analysis techniques and state analysis are applied as processing techniques.

[0941] "Application example 2 when combining with an emotional engine"

[0942] (Claim 1)

[0943] Means for collecting user behavior data, feedback data, and sentiment data,

[0944] The collected data is analyzed using natural language processing and sentiment analysis technologies to identify the user's interests, needs, and emotional state.

[0945] A means of selecting and automatically generating user-optimized content based on analysis results,

[0946] Means for providing the generated content to the user through visual and auditory interfaces,

[0947] A system that includes this.

[0948] (Claim 2)

[0949] The system according to claim 1, which collects user feedback on generated content and dynamically adjusts the content suggestion strategy based on that feedback.

[0950] (Claim 3)

[0951] The system according to claim 1, which applies topic modeling, sentiment analysis, and content suggestion based on user state as natural language processing and sentiment analysis techniques. [Explanation of Symbols]

[0952] 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. Means for collecting user behavior data and feedback data, A means of identifying user interests and needs by analyzing collected data using natural language processing technology, A means for automatically generating service suggestions suitable for the user based on analysis results, A means of presenting the generated suggestions to the user interface, A system that includes this.

2. The system according to claim 1, further comprising means for collecting user feedback on generated proposals and optimizing the content of the proposals based on that feedback.

3. The system according to claim 1, comprising means for applying topic modeling and sentiment analysis as natural language processing techniques.

Images