system

Abstract

We provide the system. [Solution] A system including means for customizing content based on user attributes and preferences, means for generating content using generative AI, and means for providing the generated content to the user.

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, including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the conventional content providing method, since the same content is provided to all users, it is difficult to customize according to the individual preferences and attributes of users. As a result, there is a problem that user feedback such as "not suitable" may be received, and it is difficult to maintain high customer satisfaction.

Means for Solving the Problems

[0005] The present invention provides means for customizing content based on user attributes and preferences, means for generating content using generative AI, and means for providing the generated content to users. Thereby, it becomes possible to provide content optimized for individual users and achieve high customer satisfaction. [Brief explanation of the drawing]

[0006] [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] This is a sequence diagram showing the processing flow of the data processing system in Embodiment 1 of Example 1. [Figure 12] This is a sequence diagram showing the processing flow of the data processing system in Application Example 1 of Form Example 1. [Figure 13] This is a sequence diagram showing the processing flow of the data processing system in Example 2 of Embodiment 2. [Figure 14] This is a sequence diagram showing the processing flow of the data processing system in Application Example 2 of Form Example 2. [Figure 15] This is a sequence diagram showing the processing flow of the data processing system in Embodiment 3 of Example 3. [Figure 16] It is a sequence diagram showing the processing flow of the data processing system in Application Example 3 of Embodiment Example 3. [Figure 17] It is a sequence diagram showing the processing flow of the data processing system in Example 1 of Embodiment Example 1 when combined with an emotion engine. [Figure 18] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1 of Embodiment Example 1 when combined with an emotion engine. [Figure 19] It is a sequence diagram showing the processing flow of the data processing system in Example 2 of Embodiment Example 2 when combined with an emotion engine. [Figure 20] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 of Embodiment Example 2 when combined with an emotion engine. [Figure 21] It is a sequence diagram showing the processing flow of the data processing system in Example 3 of Embodiment Example 3 when combined with an emotion engine. [Figure 22] It is a sequence diagram showing the processing flow of the data processing system in Application Example 3 of Embodiment Example 3 when combined with an emotion engine. [Figure 23] It is a sequence diagram showing the processing flow of the data processing system in other embodiments.

Embodiments for Carrying out the Invention

[0007] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0008] First, the terms used in the following description will be explained.

[0009] In the following embodiments, the numbered processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), or a TPU (TENSOR PROCESSING UNIT (registered trademark)), etc.

[0010] 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.

[0011] In the following embodiments, the numbered storage is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, etc.

[0012] In the following embodiments, the numbered communication I / F (Interface) is an interface including a communication processor and an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark), etc.

[0013] 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."

[0014] [First Embodiment]

[0015] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0016] 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.

[0017] 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).

[0018] 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.

[0019] 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.

[0020] 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.

[0021] 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.

[0022] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0023] 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.

[0024] 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.

[0025] 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.

[0026] Next, the identification process performed by the identification processing unit 290 of the data processing device 12 will be described.

[0027] "Example of form 1"

[0028] One embodiment of the present invention is a system comprising means for customizing content based on user attributes and preferences, means for generating content using generative AI, and means for providing the generated content to the user. Specifically, user attributes and preferences are learned from user behavior data and feedback, and the generative AI generates content based on the results. The generated content is provided in different formats such as text, audio, video, and design.

[0029] "Example of form 2"

[0030] As a concrete example, consider the case where a user purchases a book from an online bookstore. The system learns the user's preferences and attributes from behavioral data such as purchase history, browsing history, and reviews. Based on this learning, the generative AI generates summaries and recommendations of books that the user might be interested in. The generated content is then displayed and provided to the user the next time they visit the online bookstore.

[0031] "Example of form 3"

[0032] Another scenario is when users utilize music streaming services. The system learns the user's musical preferences and attributes from behavioral data such as their playback history, playlists, and ratings. Based on this learning, the generative AI generates playlists that the user is likely to enjoy. These generated playlists are then displayed and provided to the user the next time they use the music streaming service.

[0033] The following describes the processing flow for each example of the form.

[0034] "Example of form 1"

[0035] Step 1: Collect user behavior data and feedback.

[0036] Step 2: Learn user attributes and preferences from the collected data.

[0037] Step 3: Based on the learning results, the generative AI generates content.

[0038] Step 4: Provide the generated content to the user.

[0039] "Example of form 2"

[0040] Step 1: When a user visits an online bookstore, collect behavioral data such as their purchase history, browsing history, and reviews.

[0041] Step 2: Learn user preferences and attributes from the collected data.

[0042] Step 3: Based on the learning results, the generative AI generates summaries and recommendations of books that the user might be interested in.

[0043] Step 4: Display and provide the generated content to the user the next time they visit the online bookstore.

[0044] "Example of form 3"

[0045] Step 1: When a user uses a music streaming service, collect behavioral data such as the user's playback history, playlists, and ratings.

[0046] Step 2: Learn the user's music preferences and attributes from the collected data.

[0047] Step 3: Based on the learning results, the generative AI generates a playlist that the user is likely to like.

[0048] Step 4: Display and provide the generated playlist to the user the next time they use the music streaming service.

[0049] (Example 1)

[0050] Next, we will describe Example 1 of Form 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."

[0051] In today's information-saturated society, users face the challenge of efficiently acquiring information that matches their characteristics and preferences. Furthermore, conventional systems are unable to fully utilize user behavior data and opinions, making personalized information delivery difficult.

[0052] 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.

[0053] In this invention, the server includes means for personalizing information based on the user's characteristics and preferences, means for generating information using generative artificial intelligence, and means for providing the generated information to the user. This enables the user to efficiently obtain information that suits their characteristics and preferences.

[0054] "User characteristics" refer to information that indicates a user's behavioral patterns, preferences, interests, and other related information.

[0055] "Preferences" refer to information that indicates the tendencies and preferences that a user particularly likes.

[0056] "Means of personalizing information" refer to methods and technologies for customizing information based on user characteristics and preferences, and providing it in a format suitable for each individual user.

[0057] "Generative artificial intelligence" is an artificial intelligence technology that has the ability to automatically generate new information and content based on given instructions and data.

[0058] "Means of generating information" refers to methods and technologies that use generative artificial intelligence to create information tailored to the user's characteristics and preferences.

[0059] "Means of providing information to users" refers to the methods and technologies used to deliver generated information to users.

[0060] "Behavioral information" refers to data such as the actions a user performs on the system and their browsing history.

[0061] "Opinions" refer to information such as feedback, ratings, and comments provided by users.

[0062] An "instruction statement" is a text used to give instructions to a generative artificial intelligence to generate information.

[0063] To implement this invention, the server needs to generate a program for personalizing information based on the user's characteristics and preferences. This program includes means for collecting user behavior information and opinions, and generating information using generative artificial intelligence based on that information.

[0064] The server uses a database management system (e.g., MySQL®) to collect user behavior information and opinions. The collected data is analyzed using data analysis tools (e.g., Python's Pandas library) to identify user characteristics and preferences. Next, a generative artificial intelligence model (e.g., a large-scale language model) is used to generate information based on user characteristics and preferences. This generated information is provided in various formats, such as text, audio, video, and design.

[0065] As a concrete example, consider a scenario where a user requests information about movies. Based on the user's past viewing history and ratings, the server prompts a generative artificial intelligence model with the message, "Create a list of movies that the user might like." Based on this prompt, the generative AI generates a list of movies that match the user's preferences and provides it to the user through the terminal.

[0066] This system allows users to efficiently obtain information that matches their characteristics and preferences.

[0067] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0068] Step 1:

[0069] The server collects user behavior information and opinions. Specifically, it collects data such as user actions on websites and applications, browsing history, ratings, and comments. This data is entered into and stored in a database management system (e.g., MySQL).

[0070] Step 2:

[0071] The server analyzes the collected data. Specifically, it uses the Python Pandas library to organize the data and identify user behavior patterns and preferences. The input is the data collected in step 1, and the output is information about user characteristics and preferences. This analysis reveals trends in categories that users frequently view and content that they rate highly.

[0072] Step 3:

[0073] The server creates prompt statements to be input to the generative artificial intelligence model. Specifically, based on the user's characteristics and preferences obtained in step 2, it generates prompts such as, "Suggest content that the user might be interested in." The input is information about the user's characteristics and preferences, and the output is a prompt statement.

[0074] Step 4:

[0075] The server generates information using a generative artificial intelligence model. Specifically, it inputs the prompt text created in step 3 into a generative artificial intelligence model (e.g., a large-scale language model) to generate information that matches the user's characteristics and preferences. The input is a prompt text, and the output is information in the form of text, audio, video, design, etc.

[0076] Step 5:

[0077] The server sends the generated information to the terminal and provides it to the user. Specifically, it sends notifications to the user's device or displays the information within the application. The input is the information generated in step 4, and the output is the information provided in a format that the user can view. The user can view and enjoy the provided information.

[0078] (Application Example 1)

[0079] Next, we will describe Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server," and the smart device 14 will be referred to as a "terminal."

[0080] In modern information distribution services, providing content tailored to diverse user preferences and attributes in real time is challenging. In particular, there is a need for technology that leverages user behavior history and feedback to generate and instantly deliver video and audio information optimized for individual users.

[0081] 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.

[0082] In this invention, the server includes means for customizing information based on user attributes and preferences, means for generating information using generative artificial intelligence, and means for providing the generated information to the user. This makes it possible to generate and deliver video and audio information optimized for individual users in real time, based on the user's behavior history and feedback.

[0083] "User attributes" refer to the unique characteristics and traits of each individual user, including age, gender, interests, and preferences.

[0084] "Preferences" refer to the things that users are particularly interested in or tend to prefer, and are inferred from their viewing history and feedback.

[0085] "Means of customizing information" refers to methods and technologies for individually adjusting the information provided based on the user's attributes and preferences.

[0086] "Generative artificial intelligence" refers to artificial intelligence technology that has the ability to automatically generate new information and content based on given data and instructions.

[0087] "Means of generating information" refers to methods and technologies that use generative artificial intelligence to create information tailored to the user's attributes and preferences.

[0088] "Means of providing information" refers to methods and technologies for delivering generated information to users, and includes distribution technologies and display technologies.

[0089] "Behavioral history" refers to a record of a user's past actions, including viewing history and search history.

[0090] "Feedback" refers to the evaluations and opinions provided by users, indicating their reaction to and satisfaction with the content.

[0091] "Means of real-time delivery" refers to methods and technologies for delivering generated information to users immediately, with the aim of providing information without delay.

[0092] The system for carrying out this invention customizes information based on user attributes and preferences, generates information using generative artificial intelligence, and provides the generated information to the user. The system includes a server, a user terminal, and a generative artificial intelligence model.

[0093] The server collects user behavior history and feedback, and analyzes user attributes and preferences based on this data. Based on the analysis results, it sends prompts to a generative artificial intelligence (AI) system to generate information optimized for the user. Examples of generative AI systems used include OpenAI's GPT-4®.

[0094] User devices, such as smartphones and smart glasses, receive generated information in real time and provide it to the user. The information is delivered in video and audio formats, and content tailored to the user's preferences can be viewed instantly.

[0095] As a concrete example, based on the genres and ratings of movies a user has watched in the past, a generative artificial intelligence can generate and provide a new movie trailer to the user. An example of a prompt to input into the generative AI model would be, "Generate a new movie trailer that includes elements of action movies that the user likes."

[0096] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0097] Step 1:

[0098] The server collects user behavior history and feedback. It receives user viewing history and rating data as input and stores this in a database. As output, it generates analytical data showing user attributes and preferences. Specifically, the server executes database queries to aggregate users' past behavior.

[0099] Step 2:

[0100] The server analyzes user attributes and preferences based on the collected data. It uses the analysis data obtained in Step 1 as input and applies a machine learning algorithm. As output, it generates a profile that reflects the user's preferences. Specifically, the server uses a clustering algorithm to classify user interests.

[0101] Step 3:

[0102] The server sends a prompt to the generative artificial intelligence. It uses the user profile generated in step 2 as input to create the prompt text. It receives information generated by the generative artificial intelligence as output. Specifically, the server generates the prompt text in text format and sends it to the generative AI model.

[0103] Step 4:

[0104] Generative artificial intelligence generates information based on received prompts. It receives prompt text from a server as input and generates information using a generative AI model. It returns the generated video or audio information to the server as output. Specifically, the generative AI model generates content using natural language processing techniques.

[0105] Step 5:

[0106] The server delivers the generated information to the user's terminal. It receives the information generated in step 4 as input and sends it to the user's terminal. As output, it provides the information in a format viewable by the user. Specifically, the server delivers the information in real time using streaming technology.

[0107] Step 6:

[0108] The user terminal provides the user with the information it receives. As input, it receives information distributed from the server and displays or plays it for the user. As output, it provides content for the user to view. Specifically, the terminal launches a video or audio player and plays the content.

[0109] (Example 2)

[0110] Next, we will describe Example 2 of Form 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".

[0111] Traditional online content delivery systems have the challenge of not adequately providing personalized information based on individual user preferences. Furthermore, there is the difficulty in effectively utilizing user behavior data to generate information that is optimal for each user.

[0112] 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.

[0113] In this invention, the server includes means for collecting user behavior information, means for preprocessing the collected behavior information, and means for constructing a model that learns user preferences using the preprocessed behavior information. This makes it possible to provide personalized information based on the individual preferences of the user.

[0114] "User behavior information" refers to data about users' online activities, such as their purchase history, browsing history, and reviews.

[0115] "Preprocessing" refers to processes performed to convert collected data into an analyzable format, such as imputing missing values, removing outliers, and extracting features.

[0116] A "preference-learning model" is a machine learning model built to predict a user's preferences and interests based on their behavioral data.

[0117] A "generative AI model" is an artificial intelligence model that uses natural language processing technology to generate information tailored to the user.

[0118] A "prompt statement" is an instruction given to a generative AI model, used to specify the content and format of the information to be generated.

[0119] "Personalized information provision" refers to providing information that is customized based on the individual user's preferences and attributes.

[0120] This invention is a system that provides personalized information by utilizing user behavior data. The server collects behavioral information such as purchase history, browsing history, and reviews from users on online platforms. This data is stored in a database and used for later analysis.

[0121] The server uses programming languages ​​such as Python and R to preprocess the collected behavioral data. Specifically, it performs tasks such as imputing missing values ​​and removing outliers, and extracts user behavior information as features. This preprocessed data serves as the foundation for applying machine learning algorithms.

[0122] Next, the server uses the pre-processed data to build a model that learns user preferences. This model is built using libraries such as Scikit-learn and TENSORFLOW®. The model analyzes user behavior patterns and predicts information that users are likely to be interested in.

[0123] The generative AI model generates information tailored to the user based on a pre-trained model. The server inputs prompts to the generative AI model, which then generates the information to provide to the user. For example, a prompt such as "Generate a summary of a new mystery novel that the user might be interested in" might be used.

[0124] The device provides the user with generated information. The next time the user visits the online platform, personalized information will be displayed on the screen. This allows users to easily obtain information tailored to their preferences.

[0125] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0126] Step 1:

[0127] The server collects user behavior information. Specifically, it collects data such as purchase history, browsing history, and reviews from users on online platforms and stores it in a database. The input for this step is user behavior information, and the output is the raw data stored in the database.

[0128] Step 2:

[0129] The server preprocesses the collected behavioral information. The input is raw data stored in a database. The server uses Python or R to impute missing values ​​and remove outliers, converting the data into an analyzable format. The output is preprocessed, clean data.

[0130] Step 3:

[0131] The server builds a model that learns user preferences using preprocessed data. The input is clean, preprocessed data. The server applies machine learning algorithms using Scikit-learn and TensorFlow to generate a model that predicts user behavior patterns. The output is the trained preference prediction model.

[0132] Step 4:

[0133] The server inputs prompt text into a generative AI model, which then generates information tailored to the user. The input consists of a trained preference prediction model and prompt text. The server uses the generative AI model to generate information that the user is likely to find interesting. The output is the generated, personalized information.

[0134] Step 5:

[0135] The device provides the user with generated information. The input is the generated, personalized information. The device displays this information on the screen the next time the user visits the online platform. The output is the personalized information displayed to the user.

[0136] (Application Example 2)

[0137] Next, we will describe application example 2 of form example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0138] In today's information-saturated world, it is difficult for users to efficiently find products and information that suit their preferences. Furthermore, traditional recommendation systems struggle to provide accurate recommendations because they do not fully utilize diverse user behavior data.

[0139] 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.

[0140] In this invention, the server includes means for customizing information based on user attributes and preferences, means for generating information using generative artificial intelligence, and means for providing the generated information to the user. This makes it possible to provide individually customized product recommendations by utilizing user behavior data.

[0141] "User attributes" refer to information related to an individual, such as the user's age, gender, interests, and purchase history.

[0142] "Preference" refers to a user's taste or preference for a particular genre or style.

[0143] "Means of customizing information" refers to methods of individually adjusting and optimizing information based on the user's attributes and preferences.

[0144] "Generative artificial intelligence" refers to artificial intelligence technology that has the ability to generate new information and content based on data.

[0145] "Means of generating information" refers to methods of creating new information and content using generative artificial intelligence.

[0146] "Means of providing information" refers to the method of presenting and making available the generated information to the user.

[0147] "Behavioral data" refers to data that records a user's online activity history and operation history.

[0148] "Product recommendation" refers to suggesting products that a user might be interested in, based on their attributes and behavioral data.

[0149] A "communication terminal" is an electronic device, such as a smartphone or tablet, that can receive and display information.

[0150] The system for implementing this invention uses generative artificial intelligence to provide product recommendations based on user behavior data. The server customizes information based on user attributes and preferences and generates information using generative artificial intelligence. Specifically, it collects user behavior data and inputs it into a generative artificial intelligence model to generate optimal product recommendations for the user. The generated product recommendations are provided to the user via a communication terminal.

[0151] This system utilizes communication devices such as smartphones and tablets. The server is programmed using Python, and the Pandas library is used for data processing. OpenAI's GPT is employed as the generative artificial intelligence model. User behavior data (purchase history, browsing history, reviews, etc.) is processed using Pandas and input into the GPT model. The model generates product recommendations tailored to the user's preferences and displays them on the communication device.

[0152] As a concrete example, it's possible to recommend similar new books based on the genres and authors of books a user has previously purchased. An example of a prompt to the generative AI model might be, "Based on the user's past purchase history, please recommend new books that might interest them." In this way, users can efficiently find products that suit their preferences.

[0153] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0154] Step 1:

[0155] The server collects user behavior data. Specifically, it retrieves data such as user purchase history, browsing history, and reviews from a database. This data serves as input for understanding user attributes and preferences.

[0156] Step 2:

[0157] The server preprocesses the collected behavioral data using the Pandas library. Specifically, it cleans and formats the data, converting it into a format suitable for the generative AI model. This process prepares the input data for the generative AI model.

[0158] Step 3:

[0159] The server inputs pre-processed data into a generative AI model (OpenAI's GPT). The prompt used is, "Recommend new books that the user might be interested in, based on their past purchase history." The generative AI model uses this prompt and the input data to generate product recommendations that are best suited to the user.

[0160] Step 4:

[0161] The server receives product recommendations output from the generated AI model and sends them to the communication terminal. Specifically, it converts the generated recommendation results into a data format that can be displayed on the user's smartphone or tablet. This output data becomes the product recommendation information provided to the user.

[0162] Step 5:

[0163] The terminal displays product recommendations received from the server to the user. Specifically, it visually presents recommended product information on the terminal's screen, allowing the user to browse and select items. This step enables the user to efficiently find products that suit their preferences.

[0164] (Example 3)

[0165] Next, we will describe Embodiment 3 of Embodiment Example 3. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0166] Traditional content delivery systems fail to adequately provide content tailored to individual user preferences, highlighting the need for improved user experience. Furthermore, effectively utilizing user behavior data to generate highly accurate content remains a challenge.

[0167] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 3 is realized by the following means.

[0168] In this invention, the server includes means for collecting user behavior information, means for preprocessing the collected behavior information, means for learning user preferences from the preprocessed behavior information, means for generating content based on user preferences using a generative AI model, and means for providing the generated content to the user. This makes it possible to provide highly accurate content based on the individual preferences of the user.

[0169] "User behavior information" refers to data such as playback history, ratings, and playlists that are generated when users use the system.

[0170] "Preprocessing" refers to processes such as imputing missing values ​​and normalizing data, which are performed to convert collected data into a format that is easy to analyze.

[0171] "User preferences" refers to information that indicates the trends and patterns of content that users like.

[0172] A "generative AI model" refers to an artificial intelligence model used to generate content based on user preferences.

[0173] "Content" refers to information such as music, videos, and text provided to users.

[0174] The following systems are conceivable as embodiments for carrying out this invention.

[0175] The server collects user behavior information when users use music streaming services. This behavioral information includes playback history, ratings, and playlists. The server performs preprocessing, such as imputing missing values ​​and normalizing data, to convert the collected data into a format that is easy to analyze. Database management systems and data processing software can be used for this preprocessing.

[0176] Next, the server learns user preferences using pre-processed data. This learning process uses software that implements machine learning algorithms (e.g., TensorFlow or PyTorch). Based on the learning results, the server utilizes a generative AI model to generate content based on user preferences. This generative AI model uses natural language processing techniques to select music that matches the user's taste.

[0177] The generated content is delivered to the user through their device. The next time the user uses a music streaming service, the generated playlist will appear on their home screen. By playing the suggested playlist, the user can enjoy a new musical experience.

[0178] For example, if the server learns that the user prefers "rock" music, it will generate a playlist containing the latest rock songs. An example of a prompt might be, "Create a playlist based on the user's preferences." This enables the provision of highly accurate content based on the user's individual preferences. The specific processing flow in Example 3 will be explained using Figure 15.

[0179] Step 1:

[0180] The server collects behavioral information such as playback history, ratings, and playlists when users utilize music streaming services. It receives user operation logs and rating data as input and stores this information in a database. The output is a record of each user's behavioral information. Specifically, the server retrieves data in real time via an API and stores it in the database.

[0181] Step 2:

[0182] The server preprocesses the collected behavioral information. It receives the raw data collected in step 1 as input, and performs data imputation and normalization. The output is data converted into a format suitable for analysis. Specifically, the server uses data cleaning tools to impute missing values ​​with the mean and standardize numerical data.

[0183] Step 3:

[0184] The server learns user preferences using pre-processed data. It receives the pre-processed data from step 2 as input and applies a machine learning algorithm. The output is a model that represents user preferences. Specifically, the server uses a machine learning framework to perform collaborative filtering and content-based filtering.

[0185] Step 4:

[0186] The server uses a generative AI model based on the learning results to generate content based on the user's preferences. It receives the user preference model obtained in step 3 as input and inputs a prompt message into the generative AI model. The output is a playlist optimized for the user. Specifically, the server inputs the prompt message "Create a playlist based on the user's preferences" into the generative AI model and retrieves the generated playlist.

[0187] Step 5:

[0188] The terminal provides the generated playlist to the user. It receives the playlist generated in step 4 as input and displays it on the user's device. The output is the display of the playlist on the user's device. Specifically, the terminal displays the playlist through the user interface, allowing the user to start playback.

[0189] (Application Example 3)

[0190] Next, we will describe application example 3 of form example 3. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0191] In modern content delivery services, providing content tailored to diverse user preferences and attributes in real time is challenging. In particular, there is a need for technology that leverages user behavior history to quickly generate and deliver personalized, recommended content.

[0192] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means.

[0193] In this invention, the server includes means for customizing content based on user attributes and preferences, means for generating content using generative AI, means for providing the generated content to the user, means for analyzing the user's behavior history and generating recommended content in real time, and means for displaying the generated recommended content on the user's terminal. This makes it possible to provide content tailored to the user's preferences in real time.

[0194] "User attributes" refer to personal information about the user, such as age, gender, region, and interests.

[0195] "Preference" refers to a user's taste or preference for specific content or genres.

[0196] "Means of customizing content" refer to methods and technologies for adjusting and personalizing content based on user attributes and preferences.

[0197] "Generative AI" refers to a system that automatically generates new content using artificial intelligence technology.

[0198] "Means of generating content" refers to methods and technologies for creating new content using generative AI.

[0199] "Means of delivering content" refers to the methods and technologies used to deliver generated content to users.

[0200] "Behavioral history" refers to records of actions, choices, and viewing history that a user has performed in the past.

[0201] "Methods for generating recommended content in real time" refer to methods and technologies for instantly analyzing a user's behavior history and generating the most suitable content on the spot.

[0202] "Means of displaying on a device" refers to methods and technologies for visually presenting generated content on a user's device.

[0203] The system for implementing this invention customizes content based on user attributes and preferences, generates content using generative AI, and provides the generated content to the user. The server analyzes the user's behavior history and generates recommended content in real time. The generated recommended content is displayed on the user's device.

[0204] The server collects the user's music playback history and rating data, and uses this to learn the user's musical preferences. The learning results are input into a generative AI model, which generates a playlist tailored to the user. For example, OpenAI's GPT-3 (registered trademark) is used as this generative AI model.

[0205] The user's device receives the generated playlist and displays it within the application. Each time the user opens the application, new recommended playlists are displayed in real time.

[0206] For example, if a user has recently been listening to a lot of jazz and classical music, the server will generate a new playlist focusing on these genres. An example of a prompt to the generation AI model would be: "Based on the user's recent listening history, we know they like jazz and classical music. Please generate a new playlist that the user will enjoy based on this."

[0207] In this way, users can easily find music that suits their preferences, resulting in a more fulfilling music experience.

[0208] The flow of the specific processing in Application Example 3 will be explained using Figure 16.

[0209] Step 1:

[0210] The server collects users' music playback history and rating data. The input is the user's past playback history data, and the output is a dataset for analysis. This dataset is used as foundational information to learn user preferences.

[0211] Step 2:

[0212] The server learns the user's music preferences based on the collected data. The input is the dataset obtained in step 1, and the output is a profile indicating the user's music preferences. As part of data processing, a machine learning algorithm is used to extract the user's preference patterns.

[0213] Step 3:

[0214] The server inputs the user's profile into the generative AI model and generates a playlist tailored to the user. The input is the user profile obtained in step 2, and the output is the recommended playlist. The generative AI model generates the playlist using prompts. An example of such a prompt is, "Based on the user's recent playback history, we know they like jazz and classical music. Based on this, please generate a new playlist that the user will enjoy."

[0215] Step 4:

[0216] The server sends the generated playlist to the user's device. The input is the playlist generated in step 3, and the output is the playlist displayed on the user's device. The device displays the received playlist in the application, allowing the user to play it immediately.

[0217] Step 5:

[0218] The user checks the playlist displayed on their device and starts playback. The input is the playlist displayed on the device, and the output is the user's music playback experience. Through the generated playlist, the user can discover and enjoy new music.

[0219] 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.

[0220] "Example of form 1"

[0221] One embodiment of the present invention involves a system that combines an emotion engine to recognize user emotions. This system learns user emotions from user behavior data and feedback, and a generative AI uses the results to generate content. Specifically, it grasps user emotions from behavior data such as when users browse books on online bookstores and their reactions to reviews, and based on those emotions, the generative AI generates book recommendations that the user is likely to enjoy.

[0222] "Example of form 2"

[0223] Another embodiment involves user emotion recognition in music streaming services. The system understands the user's emotions from behavioral data when they listen to a particular song and their evaluation of the song, and then a generative AI generates a playlist that the user is likely to like based on those emotions.

[0224] "Example of form 3"

[0225] Another embodiment involves user emotion recognition in a movie recommendation service. The system captures the user's emotions from behavioral data when they watch a movie and their ratings of the movie, and then a generative AI generates movie recommendations that the user is likely to enjoy based on those emotions.

[0226] The following describes the processing flow for each example of the form.

[0227] "Example of form 1"

[0228] Step 1: The user browses books at an online bookstore.

[0229] Step 2: The system collects user behavior data and responses to reviews.

[0230] Step 3: The emotion engine learns the user's emotions from the data it collects.

[0231] Step 4: The generative AI generates book recommendations that the user will likely enjoy, based on the emotions it has learned.

[0232] "Example of form 2"

[0233] Step 1: The user listens to a song on a music streaming service.

[0234] Step 2: The system collects user behavior data and song ratings.

[0235] Step 3: The emotion engine learns the user's emotions from the data it collects.

[0236] Step 4: The generative AI generates a playlist that the user is likely to like based on the emotions it has learned.

[0237] "Example of form 3"

[0238] Step 1: The user watches a movie using the movie recommendation service.

[0239] Step 2: The system collects user behavior data and movie ratings.

[0240] Step 3: The emotion engine learns the user's emotions from the data it collects.

[0241] Step 4: The generative AI generates movie recommendations that the user would likely enjoy, based on the emotions it has learned.

[0242] (Example 1)

[0243] Next, we will describe Example 1 of Form 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."

[0244] Conventional information provision systems did not adequately customize information according to user attributes and preferences, making it difficult to provide users with the most relevant information. Furthermore, there was a lack of effective means to utilize user behavior data and feedback, resulting in challenges in generating information that met user needs.

[0245] 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.

[0246] In this invention, the server includes means for customizing information based on user attributes and preferences, means for generating information using generative artificial intelligence, and means for providing the generated information to the user. This makes it possible to provide information optimized for the user by utilizing user behavior data and feedback.

[0247] "User attributes" refer to personal characteristics of users, such as their age, gender, interests, and preferences.

[0248] "Preferences" refer to the user's particular interests, such as genres, themes, and styles.

[0249] "Means of customizing information" refers to methods and technologies for adjusting the content and format of information provided based on user attributes and preferences.

[0250] "Generative artificial intelligence" refers to artificial intelligence technology that uses machine learning and natural language processing techniques to automatically generate new information and content.

[0251] "Means of generating information" refers to methods and technologies that use generative artificial intelligence to create information tailored to the user's attributes and preferences.

[0252] "Means of providing information" refers to the methods and technologies used to deliver generated information to users.

[0253] "Behavioral data" refers to data such as the actions, choices, and browsing history that users perform online.

[0254] "Feedback" refers to the reactions and evaluations that users give to the information provided.

[0255] An "instruction statement" refers to a sentence used to tell a generative artificial intelligence what kind of information it should generate.

[0256] A server plays a central role in implementing this invention. The server collects user behavior data and feedback and stores it in a database. General-purpose database software can be used as the database management system. The collected data is analyzed using data analysis tools such as Python's Pandas library. This analysis identifies user attributes and preferences.

[0257] Next, the server creates a prompt message to input to the generative artificial intelligence based on the analysis results. For example, a model using natural language processing techniques can be used as the generative AI. The prompt message includes instructions for generating information tailored to the user's interests and preferences.

[0258] The generated information is provided to the user through the device. The device displays the information in a format accessible to the user via a web browser or mobile application. This allows users to easily obtain information that matches their interests.

[0259] For example, if a user frequently browses mystery novels on an online bookstore, the server analyzes this behavioral data and identifies that the user is interested in mystery novels. It then creates a prompt for the generative artificial intelligence (AI) stating, "The user is interested in mystery novels. Based on recent behavioral data, please generate book recommendations that may interest them." Based on this prompt, the AI ​​generates recommendations tailored to the user and provides them to the user through the terminal.

[0260] In this way, the system can efficiently provide information tailored to the user's attributes and preferences.

[0261] The flow of the specific processing in Example 1 will be explained using Figure 17.

[0262] Step 1:

[0263] The server collects user behavior data and feedback. Inputs include user actions, selections, and browsing history on online platforms. This data is stored in a database. Specifically, the server records user clicks and browsing time in real time. Output is the accumulation of user behavior data in the database.

[0264] Step 2:

[0265] The server analyzes the collected data. The input is user behavior data stored in a database. The server uses the Python Pandas library to analyze the data and identify user attributes and preferences. Specifically, the server aggregates data from the past month and identifies the most frequently viewed genres. The output identifies user attributes and preferences.

[0266] Step 3:

[0267] The server creates prompt text to input to the generative artificial intelligence based on the analysis results. The input includes information about the user's attributes and preferences. Based on this, the server creates a prompt text such as, "The user is interested in mystery novels. Based on recent behavioral data, please generate book recommendations that will interest them." In its specific operation, the server generates instructions that reflect the user's interests. The output is a prompt text to input to the generative artificial intelligence.

[0268] Step 4:

[0269] The server generates information using generative artificial intelligence. The input is a prompt. The generative AI uses natural language processing techniques to generate information relevant to the user. Specifically, the generative AI analyzes the prompt and generates book recommendations that the user might be interested in. The output is the information provided to the user.

[0270] Step 5:

[0271] The device provides the user with generated information. The input is information generated by a generative artificial intelligence. The device displays the information in a format accessible to the user via a web browser or mobile application. Specifically, when the user logs in, the device displays personalized recommended content on the homepage. The output is information that the user is interested in.

[0272] (Application Example 1)

[0273] Next, we will describe Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server," and the smart device 14 will be referred to as a "terminal."

[0274] In modern information distribution services, it is difficult to appropriately provide content that caters to the diverse emotions and preferences of users. In particular, there is a demand to generate and provide content customized based on the emotions of users in real time, but conventional technologies have the problem that they cannot efficiently achieve this.

[0275] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following respective means.

[0276] In this invention, the server includes means for customizing information based on the attributes and preferences of users, means for generating information using generative artificial intelligence, means for providing the generated information to users, emotion analysis means for recognizing the emotions of users, and means for the generative artificial intelligence to generate information based on the emotions recognized by the emotion analysis means. Thereby, it becomes possible to generate and provide customized content in real time according to the emotions and preferences of users.

[0277] "User attributes" refer to information related to an individual such as the user's age, gender, interests, hobbies, etc.

[0278] "Preferences" refer to information indicating the objects that a user is particularly interested in or the tendency to prefer.

[0279] "Means for customizing information" refers to methods and technologies for adjusting and individualizing information based on the attributes and preferences of users.

[0280] "Generative artificial intelligence" refers to an artificial intelligence technology having the ability to automatically generate new information using machine learning and deep learning.

[0281] "Means for generating information" refers to methods and technologies for creating new information using generative artificial intelligence.

[0282] "Means for providing information" refers to methods and technologies for delivering the generated information to users.

[0283] "Emotional analysis methods" refer to methods and techniques for recognizing and analyzing emotions from users' behavior and feedback.

[0284] "Means of generating information based on emotions" refers to methods and technologies for generative artificial intelligence to create information based on recognized emotions.

[0285] The system for implementing this invention mainly consists of a server and a terminal. The server has a database for customizing information based on user attributes and preferences, and generates information using generative artificial intelligence. The generated information is provided to the user through the terminal.

[0286] The server executes generative artificial intelligence models using software such as Python and TensorFlow. User behavior data and feedback are stored in a database and analyzed by sentiment analysis tools. It is possible to recognize user emotions in real time using libraries such as OpenCV.

[0287] The terminal is a device such as a smartphone or tablet, which displays information generated through its user interface. Users provide information to the system through viewing history and feedback, and the system uses this information to generate even more accurate information.

[0288] For example, if a user feels the need to relax, the emotion analysis system recognizes that emotion, and the generative artificial intelligence generates relaxing music or videos, which are then provided to the device. An example of a prompt message would be, "Generate music suitable for when the user feels the need to relax."

[0289] In this way, the system can generate and deliver customized content in real time that is tailored to the user's emotions and preferences.

[0290] The flow of a specific process in Application Example 1 will be explained using Figure 18.

[0291] Step 1:

[0292] The server receives behavioral data and feedback sent from the user's device. This input data includes information about the user's viewing history and emotions. The server stores this data in a database in preparation for later processing.

[0293] Step 2:

[0294] The server uses stored behavioral data and feedback to analyze the user's emotions using emotion analysis tools. Specifically, it uses libraries such as OpenCV to analyze the user's facial expressions and behavioral patterns to identify the user's emotional state. The output of this process is data indicating the user's current emotional state.

[0295] Step 3:

[0296] The server uses the user's emotional state, obtained through emotion analysis, as input to execute a generative artificial intelligence model. Using frameworks such as TensorFlow, it generates content appropriate to the user's emotions. The output of this process is customized content that matches the user's emotions.

[0297] Step 4:

[0298] The server sends the generated content to the user's device. The device displays the received content through a user interface. The user views the provided content and contributes to improving the system's accuracy by sending feedback to the server as needed.

[0299] Step 5:

[0300] The terminal sends feedback on the content viewed by the user to the server. The server stores this feedback in a database and utilizes it for content generation from the next time onwards. As a result, the system can provide content that is more adapted to the user's preferences and emotions.

[0301] (Example 2)

[0302] Next, Example 2 of Form Example 2 will be described. In the following description, the data processing device 12 is referred to as the "server", and the smart device 14 is referred to as the "terminal".

[0303] In a conventional content providing system, there was a problem that content customization based on individual user preferences and attributes was not sufficiently performed, and it was difficult to provide content that attracted the user's interest. Also, it was not possible to effectively utilize the user's behavior information to perform content generation using a generative AI model.

[0304] The specific processing by the specific processing unit 290 of the data processing device 12 in Example 2 is realized by the following respective means.

[0305] In this invention, the server includes means for collecting the user's behavior information, means for analyzing the collected behavior information, and means for generating content using a generative AI model based on the analysis result. Thereby, it becomes possible to generate and provide content based on individual user preferences and attributes.

[0306] "The user's behavior information" refers to data such as purchase history, browsing history, and evaluation information that the user has performed on an online platform.

[0307] "The means for collecting" refers to technical means for storing the user's behavior information in a database.

[0308] "Means of analysis" refers to technical means used to analyze collected user behavior information and understand user preferences and trends.

[0309] A "generative AI model" refers to an artificial intelligence model used to generate content based on user behavior data.

[0310] "Means of generating content" refers to technical means of creating user-friendly content using generative AI models.

[0311] "Means of providing to the terminal" refers to the technical means of displaying the generated content on the user's device.

[0312] A "prompt sentence" is an instruction sentence input into a generative AI model, and it refers to a sentence that determines the direction of content generation.

[0313] This invention is a system that generates and provides content to users using a generative AI model based on user behavior information. The server uses a database management system to collect user behavior information. Specifically, it uses a database such as MySQL to store user purchase history, browsing history, and rating information. This allows for the systematic management of user behavior information.

[0314] The server uses the Python pandas library to analyze the collected behavioral information. This allows for data analysis to understand user preferences and trends. The analysis results are then input into a generative AI model.

[0315] For example, OpenAI's GPT-3 is used as the generative AI model. The server generates prompt sentences based on the analysis results and inputs these prompt sentences into the generative AI model to generate content suitable for the user. An example of a prompt sentence is, "Based on the user's past purchase history, please recommend a mystery novel to read next."

[0316] The generated content is delivered to the user's device via a server. Specifically, the generated summaries and recommendations are displayed in the user's browser through a web server (e.g., Apache®). This allows users to easily browse content that suits their preferences.

[0317] For example, if a user has previously purchased many mystery novels, the AI ​​model will generate a summary of a new mystery novel and display it on their next visit. In this way, content tailored to the user's individual preferences and attributes is provided.

[0318] The flow of the specific processing in Example 2 will be explained using Figure 19.

[0319] Step 1:

[0320] The server collects user behavior information. Specifically, it stores user purchase history, browsing history, and rating information on online platforms in a database. The input is user behavior data, and the output is this data stored in the database. A database management system (e.g., MySQL) is used to systematically manage the data.

[0321] Step 2:

[0322] The server analyzes the collected behavioral information. The input is user behavioral information stored in a database, and the output is analysis results showing user preferences and trends. The Python pandas library is used to organize and aggregate the data to understand user behavior patterns. Specifically, it analyzes purchase frequency and browsing trends.

[0323] Step 3:

[0324] The server generates content using a generative AI model based on the analysis results. The input is analysis results indicating the user's preferences and tendencies, and the output is content tailored to the user. A prompt is input to the generative AI model (e.g., OpenAI's GPT-3) to generate content that will interest the user. A specific example uses the prompt: "Based on the user's past purchase history, recommend a mystery novel to read next."

[0325] Step 4:

[0326] The server provides the generated content to the user's terminal. The input is the generated content, and the output is the content displayed on the user's terminal. The generated summaries and testimonials are displayed in the user's browser via a web server (e.g., Apache). Specifically, when a user visits an online platform, the content is displayed in the browser.

[0327] (Application Example 2)

[0328] Next, we will describe application example 2 of form example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0329] It is difficult for users to choose the most suitable content from a vast amount of information. Furthermore, there is a need to provide personalized content based on user preferences and attributes, but traditional methods are insufficient to address this. Additionally, there is a lack of effective means to utilize user viewing history and rating data to accurately recommend the next content they should watch.

[0330] 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.

[0331] In this invention, the server includes means for customizing content based on user attributes and preferences, means for generating content using generative AI, and means for providing the generated content to the user. This makes it possible to collect the user's viewing history and evaluation data, input it into a generative AI model, learn the user's preferences, and generate summaries and recommendations of content that the user should watch next.

[0332] "User attributes" refer to personal characteristics of the user, such as age, gender, interests, and preferences.

[0333] "Preference" refers to a user's personal preference for specific content or genres.

[0334] "Means of customizing content" refers to methods and technologies for personalizing the content provided based on user attributes and preferences.

[0335] "Generative AI" refers to a system that automatically generates new content using artificial intelligence technology.

[0336] "Generated content" refers to information such as text, audio, video, and design created by generative AI.

[0337] "Viewing history" refers to a record of content that a user has watched in the past.

[0338] "Rating data" refers to information about the ratings and feedback that users have given to the content they have watched.

[0339] "Generative AI models" refer to the algorithms and learning models that form the core of generative AI.

[0340] A "summary" refers to information that concisely summarizes the main points of a piece of content.

[0341] A "recommendation letter" refers to a piece of writing created to recommend specific content to a user.

[0342] The system for implementing this invention collects the user's viewing history and evaluation data, inputs it into a generative AI model to learn the user's preferences, and generates summaries and recommendations for content the user should watch next.

[0343] The server stores user viewing history and rating data in a database. Cloud-based data storage such as Firebase can be used as the database. The server preprocesses this data and converts it into a format suitable for generative AI models. OpenAI's GPT-3 can be used as a generative AI model.

[0344] Generative AI models learn user preferences based on their viewing history and rating data. Based on this learning, they generate summaries and recommendations for content the user should watch next. The generated content is then sent to the user's device and provided to them.

[0345] As a concrete example, a generative AI model generates a summary of the next movie a user should watch, based on the genre and rating of the movies they have recently watched. An example of a prompt to input to the generative AI model might be: "The user recently watched an action movie with a high rating. Please generate a summary of the next movie you would recommend for this user."

[0346] The flow of a specific process in Application Example 2 will be explained using Figure 20.

[0347] Step 1:

[0348] The server collects user viewing history and rating data. Information about the content viewed by the user and their ratings are sent from the user's device to the server. The input is the user's viewing history and rating data, and the output is the storage of this data in a database.

[0349] Step 2:

[0350] The server preprocesses the collected viewing history and evaluation data. Specifically, it cleans and converts the data to a format suitable for the generative AI model. The input is the viewing history and evaluation data stored in the database, and the output is the preprocessed data.

[0351] Step 3:

[0352] The server inputs pre-processed data into a generative AI model. The generative AI model learns the user's preferences and generates summaries and recommendations for content to watch next. The input is pre-processed data, and the output is the generated summaries and recommendations.

[0353] Step 4:

[0354] The server sends the generated summary and recommendations to the user's device. The user receives information about the next content to watch through their device. The input is the generated summary and recommendations, and the output is the content information displayed on the user's device.

[0355] Step 5:

[0356] Users select the next content to watch based on summaries and recommendations displayed on their device. This selection is sent back to the server to influence future recommendations. The input is the user's selection, and the output is the updated viewing history and evaluation data.

[0357] (Example 3)

[0358] Next, we will describe Embodiment 3 of Embodiment Example 3. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0359] Traditional content delivery systems have a challenge in that they do not adequately provide content based on individual user preferences. In particular, there is a need to effectively utilize user behavior history and evaluation data to generate content that is optimal for each user.

[0360] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 3 is realized by the following means.

[0361] In this invention, the server includes means for collecting user behavior history, means for analyzing the collected behavior history and learning user preferences, and means for generating content based on user preferences using a generative AI model. This makes it possible to provide optimal content tailored to the individual preferences of each user.

[0362] "User activity history" refers to the record of a series of operations and choices that a user makes when using a system.

[0363] "Preferences" refer to the likes and tendencies that users show towards specific content or services.

[0364] A "generative AI model" is an algorithm or system that uses artificial intelligence technology to generate new content based on input data.

[0365] "Content" is a general term for information and media provided to users, such as music, videos, text, and images.

[0366] "Means of collection" refers to methods and technologies for acquiring and storing user behavior history and evaluation data.

[0367] "Means of analysis" refers to methods and techniques for processing collected data and understanding user preferences and behavioral patterns.

[0368] "Means of delivery" refers to the methods and technologies used to present and make available the generated content to users.

[0369] This invention is a system that provides content tailored to individual preferences based on the user's behavior history. The server collects and analyzes the user's behavior history to learn their preferences. Specifically, the server uses a database to store the user's playback history and evaluation data, and analyzes this data using machine learning algorithms. For analysis, the Python library Scikit-learn is used to perform clustering and classification.

[0370] The generative AI model used will utilize natural language processing technology. For example, by using OpenAI's GPT-3, it is possible to generate content based on user preferences. The generated content will be provided in various formats, including music, videos, and text.

[0371] The device displays content provided by the server to the user. The user can view, play, or use the content provided through the device. For example, if the user prefers rock music, the server will generate a playlist centered around rock music and display it on the device. An example of a prompt message would be, "Please generate a recommended playlist based on the user's playback history."

[0372] This system allows users to efficiently enjoy content tailored to their preferences. The flow of specific processing in Example 3 will be explained using Figure 21.

[0373] Step 1:

[0374] The server collects user activity history. Inputs include user playback history, playlists, and rating data. This data is stored in a database. Specifically, the server monitors user activity in real time and periodically updates the data.

[0375] Step 2:

[0376] The server analyzes the collected behavioral history to learn user preferences. The input is the data collected in Step 1. The server processes the data using machine learning algorithms to generate a user preference model. Specifically, it uses Python's Scikit-learn to perform clustering and classification to identify user preferences. The output is a model that shows the user's preferences.

[0377] Step 3:

[0378] The server generates content based on user preferences using a generative AI model. The input is the user preference model obtained in step 2. The generative AI model utilizes natural language processing techniques and generates content in response to prompt text. Specifically, it uses OpenAI's GPT-3 to generate playlists and recommendations that the user is likely to enjoy. The output is the generated content.

[0379] Step 4:

[0380] The device provides the generated content to the user. The input is the content generated in step 3. The device displays the content through the user interface, making it available to the user. Specifically, the device uses a notification function to inform the user of new content and displays playlists and recommendations on the screen. The output is the content available to the user.

[0381] (Application Example 3)

[0382] Next, we will describe application example 3 of form example 3. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0383] In modern content distribution services, providing content that caters to diverse user preferences is crucial. However, traditional systems struggle to fully utilize user behavior data, making it difficult to deliver content optimized for individual users. In particular, music streaming services require the automatic generation of personalized playlists based on users' listening history and ratings.

[0384] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means.

[0385] In this invention, the server includes means for customizing content based on user attributes and preferences, means for generating content using generative AI, means for providing the generated content to the user, means for collecting user behavior data and preprocessing it for input into a generative AI model, and means for analyzing the user's music playback history and ratings and automatically generating playlists that match the user's preferences. This makes it possible to provide personalized content that meets the individual preferences of each user.

[0386] "User attributes" refer to personal information about the user, such as age, gender, region, and interests.

[0387] "Preference" refers to a user's taste or preference for specific content or genres.

[0388] "Means of customizing content" refer to methods and technologies for individually adjusting the content provided based on the user's attributes and preferences.

[0389] "Generative AI" refers to a system that automatically generates new content using artificial intelligence technology.

[0390] "Means of generating content" refers to methods and technologies for creating new content such as music, videos, and text using generative AI.

[0391] "Means of delivering content" refers to the methods and technologies used to deliver generated content to users.

[0392] "Behavioral data" refers to data such as playback history, ratings, and click history that is generated when users use a service.

[0393] "Means of preprocessing" refer to methods and techniques for converting collected behavioral data into a format suitable for generative AI models.

[0394] "Music playback history" is a record of the music a user has played in the past.

[0395] "Rating" refers to the positive or negative feedback that users give to content.

[0396] "Methods for automatically generating playlists" refer to methods and technologies for automatically creating music lists based on the user's preferences.

[0397] The system for implementing this invention automatically generates personalized playlists using a generative AI model based on the user's music playback history and ratings. The system mainly consists of a server and the user's terminal.

[0398] The server collects user behavior data and preprocesses it for input into a generating AI model. Specifically, it uses Python to retrieve user playback history and rating data from music streaming service APIs. This data is used to analyze user preferences.

[0399] OpenAI's GPT-3 is used as a generative AI model. This model generates music playlists tailored to the user's preferences based on user behavior data. The generated playlists are provided to the user's device and displayed the next time they use the music streaming service.

[0400] For example, if the user's recently listened-to music genres are pop and rock, the generative AI model will suggest a playlist containing new artists and songs based on this. An example of a prompt might be, "The user's recently listened-to music genres are pop and rock. Please generate a new playlist based on this."

[0401] This system makes it easy for users to discover new songs that suit their musical preferences, enriching their musical experience.

[0402] The flow of the specific processing in Application Example 3 will be explained using Figure 22.

[0403] Step 1:

[0404] The server retrieves user playback history and rating data through the music streaming service's API. The input is the user ID, and the output is the user's playback history and rating data. This data serves as foundational information for analyzing the user's musical preferences.

[0405] Step 2:

[0406] The server preprocesses the acquired playback history and evaluation data. The input is the playback history and evaluation data, and the output is data converted into a format suitable for the generative AI model. Specifically, it performs data normalization and filtering to remove noise.

[0407] Step 3:

[0408] The server inputs pre-processed data into a generative AI model. The input is pre-processed data, and the output is a playlist based on the user's preferences. The generative AI model uses prompts to select music that matches the user's preferences.

[0409] Step 4:

[0410] The server sends the generated playlist to the user's device. The input is the generated playlist, and the output is the playlist displayed on the user's device. Specifically, the server converts the playlist to JSON format and sends it to the device via the API.

[0411] Step 5:

[0412] The user views the playlist displayed on their device and plays the music. The input is the playlist displayed on the device, and the output is the user's music experience. The user can discover and enjoy new music.

[0413] (Other examples)

[0414] Next, other embodiments will be described. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0415] In recent years, there has been a growing demand for information tailored to the diverse needs of users. However, conventional systems struggle to provide information that adequately reflects user characteristics and preferences. Furthermore, the lack of mechanisms to effectively utilize user feedback to improve information provision makes enhancing the user experience a significant challenge.

[0416] The identification process performed by the identification processing unit 290 of the data processing device 12 in other embodiments is realized by the following means.

[0417] In this invention, the server includes means for collecting and storing user behavior history and feedback in a database, means for analyzing the collected data to identify user characteristics, and means for automatically generating prompt sentences to instruct a generation AI model to generate information based on the identified characteristics. This enables the provision of personalized information tailored to the user's characteristics.

[0418] "User characteristics" refer to the individual user traits identified based on their behavioral history, preferences, interests, feedback, and other factors.

[0419] A "generative AI model" is a model that uses artificial intelligence technology to generate information based on input data and prompts.

[0420] A "prompt message" is a sentence input to a generative AI model to instruct it to generate specific information, and it is automatically generated based on the user's characteristics.

[0421] A "database" is a collection of information that systematically stores user behavior history, feedback, and other data, and allows for searching and analysis as needed.

[0422] "Feedback" refers to evaluations and opinions that users provide regarding the information and services offered, and is used to improve the system.

[0423] This invention is a system that generates and provides information based on user characteristics. The system mainly consists of three elements: a server, a terminal, and a user.

[0424] The server collects user activity history and feedback and stores it in a MySQL database. Activity history includes web pages visited by the user, purchase history, and survey responses. This data is used as foundational information to identify user characteristics.

[0425] The server uses the Python pandas library to create a dataframe and the scikit-learn library to perform K-means clustering to identify user characteristics. This analysis allows for grouping users based on their interests and preferences.

[0426] Based on the identified user's characteristics, the server automatically generates prompt messages to instruct the AI ​​model to generate information. An example of a prompt message is: "The user is interested in outdoor activities and has purchased camping equipment in the past. Please generate information on new camping-related products."

[0427] The generative AI model used is OpenAI's GPT-3. The server inputs the generated prompt text into GPT-3 and generates information suitable for the user. The generated information includes articles and reviews introducing new camping equipment.

[0428] The generated information is sent from the server to the user's terminal. The user's terminal displays the information using a web browser. The user interface is built using HTML and CSS, and dynamic content display is achieved using JavaScript (registered trademark).

[0429] Users provide feedback on the information presented through their device. Using the feedback form, they can rate the usefulness and level of interest of the information. This feedback is sent to the server and stored in a database. Based on the collected feedback, the server updates user characteristic data, enabling it to provide more personalized content in future information generation.

[0430] In this way, the system can generate information tailored to the user's characteristics and continuously improve the user experience by utilizing feedback on the information provided.

[0431] The flow of specific processing in other embodiments will be explained using Figure 23.

[0432] Step 1:

[0433] The server collects user behavior history and feedback and stores it in a MySQL database. Inputs include page views from website visits, purchase history, and survey responses. This data is used as foundational information to identify user characteristics. The output is the user behavior history and feedback stored in the database.

[0434] Step 2:

[0435] The server creates a dataframe using the Python pandas library to analyze the collected data. The input is user behavior history and feedback obtained from the database. Using the dataframe, K-means clustering is performed using the scikit-learn library to identify user characteristics. The output is the clustering result showing the user characteristics.

[0436] Step 3:

[0437] The server automatically generates prompt statements to instruct the AI ​​model to generate information based on the identified user characteristics. The input is the user characteristics obtained as a result of clustering. An example of a prompt statement is generated: "The user is interested in outdoor activities and has purchased camping equipment in the past. Please generate information on new camping-related products." The output is the generated prompt statement.

[0438] Step 4:

[0439] The server inputs the generated prompt text into OpenAI's GPT-3 and generates information suitable for the user. The input is the prompt text. The generating AI model generates information based on the prompt text and outputs articles and reviews introducing new camping equipment, etc. The output is the generated information.

[0440] Step 5:

[0441] The server sends the generated information to the user's terminal. The input is the generated information. The user's terminal displays the information using a web browser. The user interface is built using HTML and CSS, and dynamic content display is achieved using JavaScript. The output is the information displayed on the user's terminal.

[0442] Step 6:

[0443] Users provide feedback on information provided through their device. This feedback consists of the user's evaluation of the information's usefulness and level of interest. The evaluation is performed using a feedback form. The output is the feedback sent to the server and stored in the database.

[0444] Step 7:

[0445] The server updates user characteristic data based on collected feedback. The input is the feedback stored in the database. By analyzing the feedback and re-evaluating user characteristics, it becomes possible to provide more personalized content in future information generation. The output is the updated user characteristic data.

[0446] 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.

[0447] Data generation model 58 is a form of 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> 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.

[0448] Other examples of generative AI include Gemini® (registered trademark) (Internet search). <url: https: gemini.google.com ?hl="ja">) are some examples.

[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 device 14.

[0450] [Second Embodiment]

[0451] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0452] 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.

[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 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.

[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 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.

[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 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.

[0462] Next, the identification process performed by the identification processing unit 290 of the data processing device 12 will be described.

[0463] "Example of form 1"

[0464] One embodiment of the present invention is a system comprising means for customizing content based on user attributes and preferences, means for generating content using generative AI, and means for providing the generated content to the user. Specifically, user attributes and preferences are learned from user behavior data and feedback, and the generative AI generates content based on the results. The generated content is provided in different formats such as text, audio, video, and design.

[0465] "Example of form 2"

[0466] As a concrete example, consider the case where a user purchases a book from an online bookstore. The system learns the user's preferences and attributes from behavioral data such as purchase history, browsing history, and reviews. Based on this learning, the generative AI generates summaries and recommendations of books that the user might be interested in. The generated content is then displayed and provided to the user the next time they visit the online bookstore.

[0467] "Example of form 3"

[0468] Another scenario is when users utilize music streaming services. The system learns the user's musical preferences and attributes from behavioral data such as their playback history, playlists, and ratings. Based on this learning, the generative AI generates playlists that the user is likely to enjoy. These generated playlists are then displayed and provided to the user the next time they use the music streaming service.

[0469] The following describes the processing flow for each example of the form.

[0470] "Example of form 1"

[0471] Step 1: Collect user behavior data and feedback.

[0472] Step 2: Learn user attributes and preferences from the collected data.

[0473] Step 3: Based on the learning results, the generative AI generates content.

[0474] Step 4: Provide the generated content to the user.

[0475] "Example of form 2"

[0476] Step 1: When a user visits an online bookstore, collect behavioral data such as their purchase history, browsing history, and reviews.

[0477] Step 2: Learn user preferences and attributes from the collected data.

[0478] Step 3: Based on the learning results, the generative AI generates summaries and recommendations of books that the user might be interested in.

[0479] Step 4: Display and provide the generated content to the user the next time they visit the online bookstore.

[0480] "Example of form 3"

[0481] Step 1: When a user uses a music streaming service, collect behavioral data such as the user's playback history, playlists, and ratings.

[0482] Step 2: Learn the user's music preferences and attributes from the collected data.

[0483] Step 3: Based on the learning results, the generative AI generates a playlist that the user is likely to like.

[0484] Step 4: Display and provide the generated playlist to the user the next time they use the music streaming service.

[0485] (Example 1)

[0486] Next, we will describe Example 1 of Form 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".

[0487] In today's information-saturated society, users face the challenge of efficiently acquiring information that matches their characteristics and preferences. Furthermore, conventional systems are unable to fully utilize user behavior data and opinions, making personalized information delivery difficult.

[0488] 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.

[0489] In this invention, the server includes means for personalizing information based on the user's characteristics and preferences, means for generating information using generative artificial intelligence, and means for providing the generated information to the user. This enables the user to efficiently obtain information that suits their characteristics and preferences.

[0490] "User characteristics" refer to information that indicates a user's behavioral patterns, preferences, interests, and other related information.

[0491] "Preferences" refer to information that indicates the tendencies and preferences that a user particularly likes.

[0492] "Means of personalizing information" refer to methods and technologies for customizing information based on user characteristics and preferences, and providing it in a format suitable for each individual user.

[0493] "Generative artificial intelligence" is an artificial intelligence technology that has the ability to automatically generate new information and content based on given instructions and data.

[0494] "Means of generating information" refers to methods and technologies that use generative artificial intelligence to create information tailored to the user's characteristics and preferences.

[0495] "Means of providing information to users" refers to the methods and technologies used to deliver generated information to users.

[0496] "Behavioral information" refers to data such as the actions a user performs on the system and their browsing history.

[0497] "Opinions" refer to information such as feedback, ratings, and comments provided by users.

[0498] An "instruction statement" is a text used to give instructions to a generative artificial intelligence to generate information.

[0499] To implement this invention, the server needs to generate a program for personalizing information based on the user's characteristics and preferences. This program includes means for collecting user behavior information and opinions, and generating information using generative artificial intelligence based on that information.

[0500] The server uses a database management system (e.g., MySQL) to collect user behavior information and opinions. The collected data is analyzed using data analysis tools (e.g., Python's Pandas library) to identify user characteristics and preferences. Next, a generative artificial intelligence model (e.g., a large-scale language model) is used to generate information based on user characteristics and preferences. This generated information is provided in various formats, such as text, audio, video, and design.

[0501] As a concrete example, consider a scenario where a user requests information about movies. Based on the user's past viewing history and ratings, the server prompts a generative artificial intelligence model with the message, "Create a list of movies that the user might like." Based on this prompt, the generative AI generates a list of movies that match the user's preferences and provides it to the user through the terminal.

[0502] This system allows users to efficiently obtain information that matches their characteristics and preferences.

[0503] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0504] Step 1:

[0505] The server collects user behavior information and opinions. Specifically, it collects data such as user actions on websites and applications, browsing history, ratings, and comments. This data is entered into and stored in a database management system (e.g., MySQL).

[0506] Step 2:

[0507] The server analyzes the collected data. Specifically, it uses the Python Pandas library to organize the data and identify user behavior patterns and preferences. The input is the data collected in step 1, and the output is information about user characteristics and preferences. This analysis reveals trends in categories that users frequently view and content that they rate highly.

[0508] Step 3:

[0509] The server creates prompt statements to be input to the generative artificial intelligence model. Specifically, based on the user's characteristics and preferences obtained in step 2, it generates prompts such as, "Suggest content that the user might be interested in." The input is information about the user's characteristics and preferences, and the output is a prompt statement.

[0510] Step 4:

[0511] The server generates information using a generative artificial intelligence model. Specifically, it inputs the prompt text created in step 3 into a generative artificial intelligence model (e.g., a large-scale language model) to generate information that matches the user's characteristics and preferences. The input is a prompt text, and the output is information in the form of text, audio, video, design, etc.

[0512] Step 5:

[0513] The server sends the generated information to the terminal and provides it to the user. Specifically, it sends notifications to the user's device or displays the information within the application. The input is the information generated in step 4, and the output is the information provided in a format that the user can view. The user can view and enjoy the provided information.

[0514] (Application Example 1)

[0515] Next, we will describe Application Example 1 of Form 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."

[0516] In modern information distribution services, providing content tailored to diverse user preferences and attributes in real time is challenging. In particular, there is a need for technology that leverages user behavior history and feedback to generate and instantly deliver video and audio information optimized for individual users.

[0517] 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.

[0518] In this invention, the server includes means for customizing information based on user attributes and preferences, means for generating information using generative artificial intelligence, and means for providing the generated information to the user. This makes it possible to generate and deliver video and audio information optimized for individual users in real time, based on the user's behavior history and feedback.

[0519] "User attributes" refer to the unique characteristics and traits of each individual user, including age, gender, interests, and preferences.

[0520] "Preferences" refer to the things that users are particularly interested in or tend to prefer, and are inferred from their viewing history and feedback.

[0521] "Means of customizing information" refers to methods and technologies for individually adjusting the information provided based on the user's attributes and preferences.

[0522] "Generative artificial intelligence" refers to artificial intelligence technology that has the ability to automatically generate new information and content based on given data and instructions.

[0523] "Means of generating information" refers to methods and technologies that use generative artificial intelligence to create information tailored to the user's attributes and preferences.

[0524] "Means of providing information" refers to methods and technologies for delivering generated information to users, and includes distribution technologies and display technologies.

[0525] "Behavioral history" refers to a record of a user's past actions, including viewing history and search history.

[0526] "Feedback" refers to the evaluations and opinions provided by users, indicating their reaction to and satisfaction with the content.

[0527] "Means of real-time delivery" refers to methods and technologies for delivering generated information to users immediately, with the aim of providing information without delay.

[0528] The system for carrying out this invention customizes information based on user attributes and preferences, generates information using generative artificial intelligence, and provides the generated information to the user. The system includes a server, a user terminal, and a generative artificial intelligence model.

[0529] The server collects user behavior history and feedback, and uses this to analyze user attributes and preferences. Based on the analysis results, it sends prompts to a generative artificial intelligence (AI) to generate information optimized for the user. For example, OpenAI's GPT-4 is used as the generative AI.

[0530] User devices, such as smartphones and smart glasses, receive generated information in real time and provide it to the user. The information is delivered in video and audio formats, and content tailored to the user's preferences can be viewed instantly.

[0531] As a concrete example, based on the genres and ratings of movies a user has watched in the past, a generative artificial intelligence can generate and provide a new movie trailer to the user. An example of a prompt to input into the generative AI model would be, "Generate a new movie trailer that includes elements of action movies that the user likes."

[0532] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0533] Step 1:

[0534] The server collects user behavior history and feedback. It receives user viewing history and rating data as input and stores this in a database. As output, it generates analytical data showing user attributes and preferences. Specifically, the server executes database queries to aggregate users' past behavior.

[0535] Step 2:

[0536] The server analyzes user attributes and preferences based on the collected data. It uses the analysis data obtained in Step 1 as input and applies a machine learning algorithm. As output, it generates a profile that reflects the user's preferences. Specifically, the server uses a clustering algorithm to classify user interests.

[0537] Step 3:

[0538] The server sends a prompt to the generative artificial intelligence. It uses the user profile generated in step 2 as input to create the prompt text. It receives information generated by the generative artificial intelligence as output. Specifically, the server generates the prompt text in text format and sends it to the generative AI model.

[0539] Step 4:

[0540] Generative artificial intelligence generates information based on received prompts. It receives prompt text from a server as input and generates information using a generative AI model. It returns the generated video or audio information to the server as output. Specifically, the generative AI model generates content using natural language processing techniques.

[0541] Step 5:

[0542] The server delivers the generated information to the user's terminal. It receives the information generated in step 4 as input and sends it to the user's terminal. As output, it provides the information in a format viewable by the user. Specifically, the server delivers the information in real time using streaming technology.

[0543] Step 6:

[0544] The user terminal provides the user with the information it receives. As input, it receives information distributed from the server and displays or plays it for the user. As output, it provides content for the user to view. Specifically, the terminal launches a video or audio player and plays the content.

[0545] (Example 2)

[0546] Next, we will describe Example 2 of Form 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".

[0547] Traditional online content delivery systems have the challenge of not adequately providing personalized information based on individual user preferences. Furthermore, there is the difficulty in effectively utilizing user behavior data to generate information that is optimal for each user.

[0548] 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.

[0549] In this invention, the server includes means for collecting user behavior information, means for preprocessing the collected behavior information, and means for constructing a model that learns user preferences using the preprocessed behavior information. This makes it possible to provide personalized information based on the individual preferences of the user.

[0550] "User behavior information" refers to data about users' online activities, such as their purchase history, browsing history, and reviews.

[0551] "Preprocessing" refers to processes performed to convert collected data into an analyzable format, such as imputing missing values, removing outliers, and extracting features.

[0552] A "preference-learning model" is a machine learning model built to predict a user's preferences and interests based on their behavioral data.

[0553] A "generative AI model" is an artificial intelligence model that uses natural language processing technology to generate information tailored to the user.

[0554] A "prompt statement" is an instruction given to a generative AI model, used to specify the content and format of the information to be generated.

[0555] "Personalized information provision" refers to providing information that is customized based on the individual user's preferences and attributes.

[0556] This invention is a system that provides personalized information by utilizing user behavior data. The server collects behavioral information such as purchase history, browsing history, and reviews from users on online platforms. This data is stored in a database and used for later analysis.

[0557] The server uses programming languages ​​such as Python and R to preprocess the collected behavioral data. Specifically, it performs tasks such as imputing missing values ​​and removing outliers, and extracts user behavior information as features. This preprocessed data serves as the foundation for applying machine learning algorithms.

[0558] Next, the server uses the pre-processed data to build a model that learns user preferences. This model is built using libraries such as Scikit-learn and TensorFlow. The model analyzes user behavior patterns and predicts information that users are likely to be interested in.

[0559] The generative AI model generates information tailored to the user based on a pre-trained model. The server inputs prompts to the generative AI model, which then generates the information to provide to the user. For example, a prompt such as "Generate a summary of a new mystery novel that the user might be interested in" might be used.

[0560] The device provides the user with generated information. The next time the user visits the online platform, personalized information will be displayed on the screen. This allows users to easily obtain information tailored to their preferences.

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

[0562] Step 1:

[0563] The server collects user behavior information. Specifically, it collects data such as purchase history, browsing history, and reviews from users on online platforms and stores it in a database. The input for this step is user behavior information, and the output is the raw data stored in the database.

[0564] Step 2:

[0565] The server preprocesses the collected behavioral information. The input is raw data stored in a database. The server uses Python or R to impute missing values ​​and remove outliers, converting the data into an analyzable format. The output is preprocessed, clean data.

[0566] Step 3:

[0567] The server builds a model that learns user preferences using preprocessed data. The input is clean, preprocessed data. The server applies machine learning algorithms using Scikit-learn and TensorFlow to generate a model that predicts user behavior patterns. The output is the trained preference prediction model.

[0568] Step 4:

[0569] The server inputs prompt text into a generative AI model, which then generates information tailored to the user. The input consists of a trained preference prediction model and prompt text. The server uses the generative AI model to generate information that the user is likely to find interesting. The output is the generated, personalized information.

[0570] Step 5:

[0571] The device provides the user with generated information. The input is the generated, personalized information. The device displays this information on the screen the next time the user visits the online platform. The output is the personalized information displayed to the user.

[0572] (Application Example 2)

[0573] Next, we will describe application example 2 of form example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 as the "terminal".

[0574] In today's information-saturated world, it is difficult for users to efficiently find products and information that suit their preferences. Furthermore, traditional recommendation systems struggle to provide accurate recommendations because they do not fully utilize diverse user behavior data.

[0575] 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.

[0576] In this invention, the server includes means for customizing information based on user attributes and preferences, means for generating information using generative artificial intelligence, and means for providing the generated information to the user. This makes it possible to provide individually customized product recommendations by utilizing user behavior data.

[0577] "User attributes" refer to information related to an individual, such as the user's age, gender, interests, and purchase history.

[0578] "Preference" refers to a user's taste or preference for a particular genre or style.

[0579] "Means of customizing information" refers to methods of individually adjusting and optimizing information based on the user's attributes and preferences.

[0580] "Generative artificial intelligence" refers to artificial intelligence technology that has the ability to generate new information and content based on data.

[0581] "Means of generating information" refers to methods of creating new information and content using generative artificial intelligence.

[0582] "Means of providing information" refers to the method of presenting and making available the generated information to the user.

[0583] "Behavioral data" refers to data that records a user's online activity history and operation history.

[0584] "Product recommendation" refers to suggesting products that a user might be interested in, based on their attributes and behavioral data.

[0585] A "communication terminal" is an electronic device, such as a smartphone or tablet, that can receive and display information.

[0586] The system for implementing this invention uses generative artificial intelligence to provide product recommendations based on user behavior data. The server customizes information based on user attributes and preferences and generates information using generative artificial intelligence. Specifically, it collects user behavior data and inputs it into a generative artificial intelligence model to generate optimal product recommendations for the user. The generated product recommendations are provided to the user via a communication terminal.

[0587] This system utilizes communication devices such as smartphones and tablets. The server is programmed using Python, and the Pandas library is used for data processing. OpenAI's GPT is employed as the generative artificial intelligence model. User behavior data (purchase history, browsing history, reviews, etc.) is processed using Pandas and input into the GPT model. The model generates product recommendations tailored to the user's preferences and displays them on the communication device.

[0588] As a concrete example, it's possible to recommend similar new books based on the genres and authors of books a user has previously purchased. An example of a prompt to the generative AI model might be, "Based on the user's past purchase history, please recommend new books that might interest them." In this way, users can efficiently find products that suit their preferences.

[0589] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0590] Step 1:

[0591] The server collects user behavior data. Specifically, it retrieves data such as user purchase history, browsing history, and reviews from a database. This data serves as input for understanding user attributes and preferences.

[0592] Step 2:

[0593] The server preprocesses the collected behavioral data using the Pandas library. Specifically, it cleans and formats the data, converting it into a format suitable for the generative AI model. This process prepares the input data for the generative AI model.

[0594] Step 3:

[0595] The server inputs pre-processed data into a generative AI model (OpenAI's GPT). The prompt used is, "Recommend new books that the user might be interested in, based on their past purchase history." The generative AI model uses this prompt and the input data to generate product recommendations that are best suited to the user.

[0596] Step 4:

[0597] The server receives product recommendations output from the generated AI model and sends them to the communication terminal. Specifically, it converts the generated recommendation results into a data format that can be displayed on the user's smartphone or tablet. This output data becomes the product recommendation information provided to the user.

[0598] Step 5:

[0599] The terminal displays product recommendations received from the server to the user. Specifically, it visually presents recommended product information on the terminal's screen, allowing the user to browse and select items. This step enables the user to efficiently find products that suit their preferences.

[0600] (Example 3)

[0601] Next, we will describe Embodiment 3 of Embodiment Example 3. 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".

[0602] Traditional content delivery systems fail to adequately provide content tailored to individual user preferences, highlighting the need for improved user experience. Furthermore, effectively utilizing user behavior data to generate highly accurate content remains a challenge.

[0603] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 3 is realized by the following means.

[0604] In this invention, the server includes means for collecting user behavior information, means for preprocessing the collected behavior information, means for learning user preferences from the preprocessed behavior information, means for generating content based on user preferences using a generative AI model, and means for providing the generated content to the user. This makes it possible to provide highly accurate content based on the individual preferences of the user.

[0605] "User behavior information" refers to data such as playback history, ratings, and playlists that are generated when users use the system.

[0606] "Preprocessing" refers to processes such as imputing missing values ​​and normalizing data, which are performed to convert collected data into a format that is easy to analyze.

[0607] "User preferences" refers to information that indicates the trends and patterns of content that users like.

[0608] A "generative AI model" refers to an artificial intelligence model used to generate content based on user preferences.

[0609] "Content" refers to information such as music, videos, and text provided to users.

[0610] The following systems are conceivable as embodiments for carrying out this invention.

[0611] The server collects user behavior information when users use music streaming services. This behavioral information includes playback history, ratings, and playlists. The server performs preprocessing, such as imputing missing values ​​and normalizing data, to convert the collected data into a format that is easy to analyze. Database management systems and data processing software can be used for this preprocessing.

[0612] Next, the server learns user preferences using pre-processed data. This learning process uses software that implements machine learning algorithms (e.g., TensorFlow or PyTorch). Based on the learning results, the server utilizes a generative AI model to generate content based on user preferences. This generative AI model uses natural language processing techniques to select music that matches the user's taste.

[0613] The generated content is delivered to the user through their device. The next time the user uses a music streaming service, the generated playlist will appear on their home screen. By playing the suggested playlist, the user can enjoy a new musical experience.

[0614] For example, if the server learns that the user prefers "rock" music, it will generate a playlist containing the latest rock songs. An example of a prompt might be, "Create a playlist based on the user's preferences." This enables the provision of highly accurate content based on the user's individual preferences. The specific processing flow in Example 3 will be explained using Figure 15.

[0615] Step 1:

[0616] The server collects behavioral information such as playback history, ratings, and playlists when users utilize music streaming services. It receives user operation logs and rating data as input and stores this information in a database. The output is a record of each user's behavioral information. Specifically, the server retrieves data in real time via an API and stores it in the database.

[0617] Step 2:

[0618] The server preprocesses the collected behavioral information. It receives the raw data collected in step 1 as input, and performs data imputation and normalization. The output is data converted into a format suitable for analysis. Specifically, the server uses data cleaning tools to impute missing values ​​with the mean and standardize numerical data.

[0619] Step 3:

[0620] The server learns user preferences using pre-processed data. It receives the pre-processed data from step 2 as input and applies a machine learning algorithm. The output is a model that represents user preferences. Specifically, the server uses a machine learning framework to perform collaborative filtering and content-based filtering.

[0621] Step 4:

[0622] The server uses a generative AI model based on the learning results to generate content based on the user's preferences. It receives the user preference model obtained in step 3 as input and inputs a prompt message into the generative AI model. The output is a playlist optimized for the user. Specifically, the server inputs the prompt message "Create a playlist based on the user's preferences" into the generative AI model and retrieves the generated playlist.

[0623] Step 5:

[0624] The terminal provides the generated playlist to the user. It receives the playlist generated in step 4 as input and displays it on the user's device. The output is the display of the playlist on the user's device. Specifically, the terminal displays the playlist through the user interface, allowing the user to start playback.

[0625] (Application Example 3)

[0626] Next, we will describe application example 3 of form example 3. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 as the "terminal".

[0627] In modern content delivery services, providing content tailored to diverse user preferences and attributes in real time is challenging. In particular, there is a need for technology that leverages user behavior history to quickly generate and deliver personalized, recommended content.

[0628] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means.

[0629] In this invention, the server includes means for customizing content based on user attributes and preferences, means for generating content using generative AI, means for providing the generated content to the user, means for analyzing the user's behavior history and generating recommended content in real time, and means for displaying the generated recommended content on the user's terminal. This makes it possible to provide content tailored to the user's preferences in real time.

[0630] "User attributes" refer to personal information about the user, such as age, gender, region, and interests.

[0631] "Preference" refers to a user's taste or preference for specific content or genres.

[0632] "Means of customizing content" refer to methods and technologies for adjusting and personalizing content based on user attributes and preferences.

[0633] "Generative AI" refers to a system that automatically generates new content using artificial intelligence technology.

[0634] "Means of generating content" refers to methods and technologies for creating new content using generative AI.

[0635] "Means of delivering content" refers to the methods and technologies used to deliver generated content to users.

[0636] "Behavioral history" refers to records of actions, choices, and viewing history that a user has performed in the past.

[0637] "Methods for generating recommended content in real time" refer to methods and technologies for instantly analyzing a user's behavior history and generating the most suitable content on the spot.

[0638] "Means of displaying on a device" refers to methods and technologies for visually presenting generated content on a user's device.

[0639] The system for implementing this invention customizes content based on user attributes and preferences, generates content using generative AI, and provides the generated content to the user. The server analyzes the user's behavior history and generates recommended content in real time. The generated recommended content is displayed on the user's device.

[0640] The server collects the user's music playback history and rating data, and uses this to learn the user's musical preferences. The learning results are input into a generative AI model, which generates a playlist tailored to the user. For example, OpenAI's GPT-3 is used as this generative AI model.

[0641] The user's device receives the generated playlist and displays it within the application. Each time the user opens the application, new recommended playlists are displayed in real time.

[0642] For example, if a user has recently been listening to a lot of jazz and classical music, the server will generate a new playlist focusing on these genres. An example of a prompt to the generation AI model would be: "Based on the user's recent listening history, we know they like jazz and classical music. Please generate a new playlist that the user will enjoy based on this."

[0643] In this way, users can easily find music that suits their preferences, resulting in a more fulfilling music experience.

[0644] The flow of the specific processing in Application Example 3 will be explained using Figure 16.

[0645] Step 1:

[0646] The server collects users' music playback history and rating data. The input is the user's past playback history data, and the output is a dataset for analysis. This dataset is used as foundational information to learn user preferences.

[0647] Step 2:

[0648] The server learns the user's music preferences based on the collected data. The input is the dataset obtained in step 1, and the output is a profile indicating the user's music preferences. As part of data processing, a machine learning algorithm is used to extract the user's preference patterns.

[0649] Step 3:

[0650] The server inputs the user's profile into the generative AI model and generates a playlist tailored to the user. The input is the user profile obtained in step 2, and the output is the recommended playlist. The generative AI model generates the playlist using prompts. An example of such a prompt is, "Based on the user's recent playback history, we know they like jazz and classical music. Based on this, please generate a new playlist that the user will enjoy."

[0651] Step 4:

[0652] The server sends the generated playlist to the user's device. The input is the playlist generated in step 3, and the output is the playlist displayed on the user's device. The device displays the received playlist in the application, allowing the user to play it immediately.

[0653] Step 5:

[0654] The user checks the playlist displayed on their device and starts playback. The input is the playlist displayed on the device, and the output is the user's music playback experience. Through the generated playlist, the user can discover and enjoy new music.

[0655] 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.

[0656] "Example of form 1"

[0657] One embodiment of the present invention involves a system that combines an emotion engine to recognize user emotions. This system learns user emotions from user behavior data and feedback, and a generative AI uses the results to generate content. Specifically, it grasps user emotions from behavior data such as when users browse books on online bookstores and their reactions to reviews, and based on those emotions, the generative AI generates book recommendations that the user is likely to enjoy.

[0658] "Example of form 2"

[0659] Another embodiment involves user emotion recognition in music streaming services. The system understands the user's emotions from behavioral data when they listen to a particular song and their evaluation of the song, and then a generative AI generates a playlist that the user is likely to like based on those emotions.

[0660] "Example of form 3"

[0661] Another embodiment involves user emotion recognition in a movie recommendation service. The system captures the user's emotions from behavioral data when they watch a movie and their ratings of the movie, and then a generative AI generates movie recommendations that the user is likely to enjoy based on those emotions.

[0662] The following describes the processing flow for each example of the form.

[0663] "Example of form 1"

[0664] Step 1: The user browses books at an online bookstore.

[0665] Step 2: The system collects user behavior data and responses to reviews.

[0666] Step 3: The emotion engine learns the user's emotions from the data it collects.

[0667] Step 4: The generative AI generates book recommendations that the user will likely enjoy, based on the emotions it has learned.

[0668] "Example of form 2"

[0669] Step 1: The user listens to a song on a music streaming service.

[0670] Step 2: The system collects user behavior data and song ratings.

[0671] Step 3: The emotion engine learns the user's emotions from the data it collects.

[0672] Step 4: The generative AI generates a playlist that the user is likely to like based on the emotions it has learned.

[0673] "Example of form 3"

[0674] Step 1: The user watches a movie using the movie recommendation service.

[0675] Step 2: The system collects user behavior data and movie ratings.

[0676] Step 3: The emotion engine learns the user's emotions from the data it collects.

[0677] Step 4: The generative AI generates movie recommendations that the user would likely enjoy, based on the emotions it has learned.

[0678] (Example 1)

[0679] Next, we will describe Example 1 of Form 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".

[0680] Conventional information provision systems did not adequately customize information according to user attributes and preferences, making it difficult to provide users with the most relevant information. Furthermore, there was a lack of effective means to utilize user behavior data and feedback, resulting in challenges in generating information that met user needs.

[0681] 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.

[0682] In this invention, the server includes means for customizing information based on user attributes and preferences, means for generating information using generative artificial intelligence, and means for providing the generated information to the user. This makes it possible to provide information optimized for the user by utilizing user behavior data and feedback.

[0683] "User attributes" refer to personal characteristics of users, such as their age, gender, interests, and preferences.

[0684] "Preferences" refer to the user's particular interests, such as genres, themes, and styles.

[0685] "Means of customizing information" refers to methods and technologies for adjusting the content and format of information provided based on user attributes and preferences.

[0686] "Generative artificial intelligence" refers to artificial intelligence technology that uses machine learning and natural language processing techniques to automatically generate new information and content.

[0687] "Means of generating information" refers to methods and technologies that use generative artificial intelligence to create information tailored to the user's attributes and preferences.

[0688] "Means of providing information" refers to the methods and technologies used to deliver generated information to users.

[0689] "Behavioral data" refers to data such as the actions, choices, and browsing history that users perform online.

[0690] "Feedback" refers to the reactions and evaluations that users give to the information provided.

[0691] An "instruction statement" refers to a sentence used to tell a generative artificial intelligence what kind of information it should generate.

[0692] A server plays a central role in implementing this invention. The server collects user behavior data and feedback and stores it in a database. General-purpose database software can be used as the database management system. The collected data is analyzed using data analysis tools such as Python's Pandas library. This analysis identifies user attributes and preferences.

[0693] Next, the server creates a prompt message to input to the generative artificial intelligence based on the analysis results. For example, a model using natural language processing techniques can be used as the generative AI. The prompt message includes instructions for generating information tailored to the user's interests and preferences.

[0694] The generated information is provided to the user through the device. The device displays the information in a format accessible to the user via a web browser or mobile application. This allows users to easily obtain information that matches their interests.

[0695] For example, if a user frequently browses mystery novels on an online bookstore, the server analyzes this behavioral data and identifies that the user is interested in mystery novels. It then creates a prompt for the generative artificial intelligence (AI) stating, "The user is interested in mystery novels. Based on recent behavioral data, please generate book recommendations that may interest them." Based on this prompt, the AI ​​generates recommendations tailored to the user and provides them to the user through the terminal.

[0696] In this way, the system can efficiently provide information tailored to the user's attributes and preferences.

[0697] The flow of the specific processing in Example 1 will be explained using Figure 17.

[0698] Step 1:

[0699] The server collects user behavior data and feedback. Inputs include user actions, selections, and browsing history on online platforms. This data is stored in a database. Specifically, the server records user clicks and browsing time in real time. Output is the accumulation of user behavior data in the database.

[0700] Step 2:

[0701] The server analyzes the collected data. The input is user behavior data stored in a database. The server uses the Python Pandas library to analyze the data and identify user attributes and preferences. Specifically, the server aggregates data from the past month and identifies the most frequently viewed genres. The output identifies user attributes and preferences.

[0702] Step 3:

[0703] The server creates prompt text to input to the generative artificial intelligence based on the analysis results. The input includes information about the user's attributes and preferences. Based on this, the server creates a prompt text such as, "The user is interested in mystery novels. Based on recent behavioral data, please generate book recommendations that will interest them." In its specific operation, the server generates instructions that reflect the user's interests. The output is a prompt text to input to the generative artificial intelligence.

[0704] Step 4:

[0705] The server generates information using generative artificial intelligence. The input is a prompt. The generative AI uses natural language processing techniques to generate information relevant to the user. Specifically, the generative AI analyzes the prompt and generates book recommendations that the user might be interested in. The output is the information provided to the user.

[0706] Step 5:

[0707] The device provides the user with generated information. The input is information generated by a generative artificial intelligence. The device displays the information in a format accessible to the user via a web browser or mobile application. Specifically, when the user logs in, the device displays personalized recommended content on the homepage. The output is information that the user is interested in.

[0708] (Application Example 1)

[0709] Next, we will describe Application Example 1 of Form 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."

[0710] In modern information distribution services, it is difficult to appropriately provide content that caters to the diverse emotions and preferences of users. In particular, there is a demand for generating and delivering customized content based on users' emotions in real time, but conventional technologies cannot efficiently achieve this.

[0711] 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.

[0712] In this invention, the server includes means for customizing information based on user attributes and preferences, means for generating information using generative artificial intelligence, means for providing the generated information to the user, means for emotion analysis to recognize the user's emotions, and means for the generative artificial intelligence to generate information based on the emotions recognized by the emotion analysis means. This enables the real-time generation and provision of customized content that corresponds to the user's emotions and preferences.

[0713] "User attributes" refer to information related to an individual, such as the user's age, gender, interests, and preferences.

[0714] "Preferences" refer to information that indicates what a user is particularly interested in or what their preferences tend to be.

[0715] "Means of customizing information" refer to methods and technologies for adjusting and personalizing information based on the user's attributes and preferences.

[0716] "Generative artificial intelligence" refers to artificial intelligence technology that has the ability to automatically generate new information using machine learning and deep learning.

[0717] "Means of generating information" refers to methods and technologies for creating new information using generative artificial intelligence.

[0718] "Means of providing information" refers to the methods and technologies used to deliver generated information to users.

[0719] "Emotional analysis methods" refer to methods and techniques for recognizing and analyzing emotions from users' behavior and feedback.

[0720] "Means of generating information based on emotions" refers to methods and technologies for generative artificial intelligence to create information based on recognized emotions.

[0721] The system for implementing this invention mainly consists of a server and a terminal. The server has a database for customizing information based on user attributes and preferences, and generates information using generative artificial intelligence. The generated information is provided to the user through the terminal.

[0722] The server executes generative artificial intelligence models using software such as Python and TensorFlow. User behavior data and feedback are stored in a database and analyzed by sentiment analysis tools. It is possible to recognize user emotions in real time using libraries such as OpenCV.

[0723] The terminal is a device such as a smartphone or tablet, which displays information generated through its user interface. Users provide information to the system through viewing history and feedback, and the system uses this information to generate even more accurate information.

[0724] For example, if a user feels the need to relax, the emotion analysis system recognizes that emotion, and the generative artificial intelligence generates relaxing music or videos, which are then provided to the device. An example of a prompt message would be, "Generate music suitable for when the user feels the need to relax."

[0725] In this way, the system can generate and deliver customized content in real time that is tailored to the user's emotions and preferences.

[0726] The flow of a specific process in Application Example 1 will be explained using Figure 18.

[0727] Step 1:

[0728] The server receives behavioral data and feedback sent from the user's device. This input data includes information about the user's viewing history and emotions. The server stores this data in a database in preparation for later processing.

[0729] Step 2:

[0730] The server uses stored behavioral data and feedback to analyze the user's emotions using emotion analysis tools. Specifically, it uses libraries such as OpenCV to analyze the user's facial expressions and behavioral patterns to identify the user's emotional state. The output of this process is data indicating the user's current emotional state.

[0731] Step 3:

[0732] The server uses the user's emotional state, obtained through emotion analysis, as input to execute a generative artificial intelligence model. Using frameworks such as TensorFlow, it generates content appropriate to the user's emotions. The output of this process is customized content that matches the user's emotions.

[0733] Step 4:

[0734] The server sends the generated content to the user's device. The device displays the received content through a user interface. The user views the provided content and contributes to improving the system's accuracy by sending feedback to the server as needed.

[0735] Step 5:

[0736] The user sends feedback on the content they have watched from their device to the server. The server stores this feedback in a database and uses it to generate future content. This allows the system to provide content that is more tailored to the user's preferences and emotions.

[0737] (Example 2)

[0738] Next, we will describe Example 2 of Form 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".

[0739] Traditional content delivery systems have faced challenges in providing content that appeals to users because they do not adequately customize content based on individual user preferences and attributes. Furthermore, they have not been able to effectively utilize user behavior data to generate content using AI-generated models.

[0740] 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.

[0741] In this invention, the server includes means for collecting user behavior information, means for analyzing the collected behavior information, and means for generating content using a generative AI model based on the analysis results. This makes it possible to generate and provide content based on the individual preferences and attributes of the user.

[0742] "User behavior information" refers to data such as purchase history, browsing history, and ratings information that users have on online platforms.

[0743] "Means of collection" refers to the technical means of storing user behavior information in a database.

[0744] "Means of analysis" refers to technical means used to analyze collected user behavior information and understand user preferences and trends.

[0745] A "generative AI model" refers to an artificial intelligence model used to generate content based on user behavior data.

[0746] "Means of generating content" refers to technical means of creating user-friendly content using generative AI models.

[0747] "Means of providing to the terminal" refers to the technical means of displaying the generated content on the user's device.

[0748] A "prompt sentence" is an instruction sentence input into a generative AI model, and it refers to a sentence that determines the direction of content generation.

[0749] This invention is a system that generates and provides content to users using a generative AI model based on user behavior information. The server uses a database management system to collect user behavior information. Specifically, it uses a database such as MySQL to store user purchase history, browsing history, and rating information. This allows for the systematic management of user behavior information.

[0750] The server uses the Python pandas library to analyze the collected behavioral information. This allows for data analysis to understand user preferences and trends. The analysis results are then input into a generative AI model.

[0751] For example, OpenAI's GPT-3 is used as the generative AI model. The server generates prompt sentences based on the analysis results and inputs these prompt sentences into the generative AI model to generate content suitable for the user. An example of a prompt sentence is, "Based on the user's past purchase history, please recommend a mystery novel to read next."

[0752] The generated content is delivered to the user's device via a server. Specifically, the generated summaries and recommendations are displayed in the user's browser through a web server (e.g., Apache). This allows users to easily browse content that suits their preferences.

[0753] For example, if a user has previously purchased many mystery novels, the AI ​​model will generate a summary of a new mystery novel and display it on their next visit. In this way, content tailored to the user's individual preferences and attributes is provided.

[0754] The flow of the specific processing in Example 2 will be explained using Figure 19.

[0755] Step 1:

[0756] The server collects user behavior information. Specifically, it stores user purchase history, browsing history, and rating information on online platforms in a database. The input is user behavior data, and the output is this data stored in the database. A database management system (e.g., MySQL) is used to systematically manage the data.

[0757] Step 2:

[0758] The server analyzes the collected behavioral information. The input is user behavioral information stored in a database, and the output is analysis results showing user preferences and trends. The Python pandas library is used to organize and aggregate the data to understand user behavior patterns. Specifically, it analyzes purchase frequency and browsing trends.

[0759] Step 3:

[0760] The server generates content using a generative AI model based on the analysis results. The input is analysis results indicating the user's preferences and tendencies, and the output is content tailored to the user. A prompt is input to the generative AI model (e.g., OpenAI's GPT-3) to generate content that will interest the user. A specific example uses the prompt: "Based on the user's past purchase history, recommend a mystery novel to read next."

[0761] Step 4:

[0762] The server provides the generated content to the user's terminal. The input is the generated content, and the output is the content displayed on the user's terminal. The generated summaries and testimonials are displayed in the user's browser via a web server (e.g., Apache). Specifically, when a user visits an online platform, the content is displayed in the browser.

[0763] (Application Example 2)

[0764] Next, we will describe application example 2 of form example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 as the "terminal".

[0765] It is difficult for users to choose the most suitable content from a vast amount of information. Furthermore, there is a need to provide personalized content based on user preferences and attributes, but traditional methods are insufficient to address this. Additionally, there is a lack of effective means to utilize user viewing history and rating data to accurately recommend the next content they should watch.

[0766] 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.

[0767] In this invention, the server includes means for customizing content based on user attributes and preferences, means for generating content using generative AI, and means for providing the generated content to the user. This makes it possible to collect the user's viewing history and evaluation data, input it into a generative AI model, learn the user's preferences, and generate summaries and recommendations of content that the user should watch next.

[0768] "User attributes" refer to personal characteristics of the user, such as age, gender, interests, and preferences.

[0769] "Preference" refers to a user's personal preference for specific content or genres.

[0770] "Means of customizing content" refers to methods and technologies for personalizing the content provided based on user attributes and preferences.

[0771] "Generative AI" refers to a system that automatically generates new content using artificial intelligence technology.

[0772] "Generated content" refers to information such as text, audio, video, and design created by generative AI.

[0773] "Viewing history" refers to a record of content that a user has watched in the past.

[0774] "Rating data" refers to information about the ratings and feedback that users have given to the content they have watched.

[0775] "Generative AI models" refer to the algorithms and learning models that form the core of generative AI.

[0776] A "summary" refers to information that concisely summarizes the main points of a piece of content.

[0777] A "recommendation letter" refers to a piece of writing created to recommend specific content to a user.

[0778] The system for implementing this invention collects the user's viewing history and evaluation data, inputs it into a generative AI model to learn the user's preferences, and generates summaries and recommendations for content the user should watch next.

[0779] The server stores user viewing history and rating data in a database. Cloud-based data storage such as Firebase can be used as the database. The server preprocesses this data and converts it into a format suitable for generative AI models. OpenAI's GPT-3 can be used as a generative AI model.

[0780] Generative AI models learn user preferences based on their viewing history and rating data. Based on this learning, they generate summaries and recommendations for content the user should watch next. The generated content is then sent to the user's device and provided to them.

[0781] As a concrete example, a generative AI model generates a summary of the next movie a user should watch, based on the genre and rating of the movies they have recently watched. An example of a prompt to input to the generative AI model might be: "The user recently watched an action movie with a high rating. Please generate a summary of the next movie you would recommend for this user."

[0782] The flow of a specific process in Application Example 2 will be explained using Figure 20.

[0783] Step 1:

[0784] The server collects user viewing history and rating data. Information about the content viewed by the user and their ratings are sent from the user's device to the server. The input is the user's viewing history and rating data, and the output is the storage of this data in a database.

[0785] Step 2:

[0786] The server preprocesses the collected viewing history and evaluation data. Specifically, it cleans and converts the data to a format suitable for the generative AI model. The input is the viewing history and evaluation data stored in the database, and the output is the preprocessed data.

[0787] Step 3:

[0788] The server inputs pre-processed data into a generative AI model. The generative AI model learns the user's preferences and generates summaries and recommendations for content to watch next. The input is pre-processed data, and the output is the generated summaries and recommendations.

[0789] Step 4:

[0790] The server sends the generated summary and recommendations to the user's device. The user receives information about the next content to watch through their device. The input is the generated summary and recommendations, and the output is the content information displayed on the user's device.

[0791] Step 5:

[0792] Users select the next content to watch based on summaries and recommendations displayed on their device. This selection is sent back to the server to influence future recommendations. The input is the user's selection, and the output is the updated viewing history and evaluation data.

[0793] (Example 3)

[0794] Next, we will describe Embodiment 3 of Embodiment Example 3. 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".

[0795] Traditional content delivery systems have a challenge in that they do not adequately provide content based on individual user preferences. In particular, there is a need to effectively utilize user behavior history and evaluation data to generate content that is optimal for each user.

[0796] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 3 is realized by the following means.

[0797] In this invention, the server includes means for collecting user behavior history, means for analyzing the collected behavior history and learning user preferences, and means for generating content based on user preferences using a generative AI model. This makes it possible to provide optimal content tailored to the individual preferences of each user.

[0798] "User activity history" refers to the record of a series of operations and choices that a user makes when using a system.

[0799] "Preferences" refer to the likes and tendencies that users show towards specific content or services.

[0800] A "generative AI model" is an algorithm or system that uses artificial intelligence technology to generate new content based on input data.

[0801] "Content" is a general term for information and media provided to users, such as music, videos, text, and images.

[0802] "Means of collection" refers to methods and technologies for acquiring and storing user behavior history and evaluation data.

[0803] "Means of analysis" refers to methods and techniques for processing collected data and understanding user preferences and behavioral patterns.

[0804] "Means of delivery" refers to the methods and technologies used to present and make available the generated content to users.

[0805] This invention is a system that provides content tailored to individual preferences based on the user's behavior history. The server collects and analyzes the user's behavior history to learn their preferences. Specifically, the server uses a database to store the user's playback history and evaluation data, and analyzes this data using machine learning algorithms. For analysis, the Python library Scikit-learn is used to perform clustering and classification.

[0806] The generative AI model used will utilize natural language processing technology. For example, by using OpenAI's GPT-3, it is possible to generate content based on user preferences. The generated content will be provided in various formats, including music, videos, and text.

[0807] The device displays content provided by the server to the user. The user can view, play, or use the content provided through the device. For example, if the user prefers rock music, the server will generate a playlist centered around rock music and display it on the device. An example of a prompt message would be, "Please generate a recommended playlist based on the user's playback history."

[0808] This system allows users to efficiently enjoy content tailored to their preferences. The flow of specific processing in Example 3 will be explained using Figure 21.

[0809] Step 1:

[0810] The server collects user activity history. Inputs include user playback history, playlists, and rating data. This data is stored in a database. Specifically, the server monitors user activity in real time and periodically updates the data.

[0811] Step 2:

[0812] The server analyzes the collected behavioral history to learn user preferences. The input is the data collected in Step 1. The server processes the data using machine learning algorithms to generate a user preference model. Specifically, it uses Python's Scikit-learn to perform clustering and classification to identify user preferences. The output is a model that shows the user's preferences.

[0813] Step 3:

[0814] The server generates content based on user preferences using a generative AI model. The input is the user preference model obtained in step 2. The generative AI model utilizes natural language processing techniques and generates content in response to prompt text. Specifically, it uses OpenAI's GPT-3 to generate playlists and recommendations that the user is likely to enjoy. The output is the generated content.

[0815] Step 4:

[0816] The device provides the generated content to the user. The input is the content generated in step 3. The device displays the content through the user interface, making it available to the user. Specifically, the device uses a notification function to inform the user of new content and displays playlists and recommendations on the screen. The output is the content available to the user.

[0817] (Application Example 3)

[0818] Next, we will describe application example 3 of form example 3. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 as the "terminal".

[0819] In modern content distribution services, providing content that caters to diverse user preferences is crucial. However, traditional systems struggle to fully utilize user behavior data, making it difficult to deliver content optimized for individual users. In particular, music streaming services require the automatic generation of personalized playlists based on users' listening history and ratings.

[0820] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means.

[0821] In this invention, the server includes means for customizing content based on user attributes and preferences, means for generating content using generative AI, means for providing the generated content to the user, means for collecting user behavior data and preprocessing it for input into a generative AI model, and means for analyzing the user's music playback history and ratings and automatically generating playlists that match the user's preferences. This makes it possible to provide personalized content that meets the individual preferences of each user.

[0822] "User attributes" refer to personal information about the user, such as age, gender, region, and interests.

[0823] "Preference" refers to a user's taste or preference for specific content or genres.

[0824] "Means of customizing content" refer to methods and technologies for individually adjusting the content provided based on the user's attributes and preferences.

[0825] "Generative AI" refers to a system that automatically generates new content using artificial intelligence technology.

[0826] "Means of generating content" refers to methods and technologies for creating new content such as music, videos, and text using generative AI.

[0827] "Means of delivering content" refers to the methods and technologies used to deliver generated content to users.

[0828] "Behavioral data" refers to data such as playback history, ratings, and click history that is generated when users use a service.

[0829] "Means of preprocessing" refer to methods and techniques for converting collected behavioral data into a format suitable for generative AI models.

[0830] "Music playback history" is a record of the music a user has played in the past.

[0831] "Rating" refers to the positive or negative feedback that users give to content.

[0832] "Methods for automatically generating playlists" refer to methods and technologies for automatically creating music lists based on the user's preferences.

[0833] The system for implementing this invention automatically generates personalized playlists using a generative AI model based on the user's music playback history and ratings. The system mainly consists of a server and the user's terminal.

[0834] The server collects user behavior data and preprocesses it for input into a generating AI model. Specifically, it uses Python to retrieve user playback history and rating data from music streaming service APIs. This data is used to analyze user preferences.

[0835] OpenAI's GPT-3 is used as a generative AI model. This model generates music playlists tailored to the user's preferences based on user behavior data. The generated playlists are provided to the user's device and displayed the next time they use the music streaming service.

[0836] For example, if the user's recently listened-to music genres are pop and rock, the generative AI model will suggest a playlist containing new artists and songs based on this. An example of a prompt might be, "The user's recently listened-to music genres are pop and rock. Please generate a new playlist based on this."

[0837] This system makes it easy for users to discover new songs that suit their musical preferences, enriching their musical experience.

[0838] The flow of the specific processing in Application Example 3 will be explained using Figure 22.

[0839] Step 1:

[0840] The server retrieves user playback history and rating data through the music streaming service's API. The input is the user ID, and the output is the user's playback history and rating data. This data serves as foundational information for analyzing the user's musical preferences.

[0841] Step 2:

[0842] The server preprocesses the acquired playback history and evaluation data. The input is the playback history and evaluation data, and the output is data converted into a format suitable for the generative AI model. Specifically, it performs data normalization and filtering to remove noise.

[0843] Step 3:

[0844] The server inputs pre-processed data into a generative AI model. The input is pre-processed data, and the output is a playlist based on the user's preferences. The generative AI model uses prompts to select music that matches the user's preferences.

[0845] Step 4:

[0846] The server sends the generated playlist to the user's device. The input is the generated playlist, and the output is the playlist displayed on the user's device. Specifically, the server converts the playlist to JSON format and sends it to the device via the API.

[0847] Step 5:

[0848] The user views the playlist displayed on their device and plays the music. The input is the playlist displayed on the device, and the output is the user's music experience. The user can discover and enjoy new music.

[0849] (Other examples)

[0850] Since this is the same as the specific processing described in the other embodiments of the first embodiment above, the explanation will be omitted.

[0851] 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.

[0852] The data generation model 58 is a form of so-called generative AI (Artificial Intelligence). One example of the data generation model 58 is ChatGPT (Internet Search).<URL: https: / / openai.com / blog / chatgpt> 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.

[0853] Other examples of generative AI include Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) are some examples.

[0854] 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.

[0855] [Third Embodiment]

[0856] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0857] 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.

[0858] 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).

[0859] 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.

[0860] 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.

[0861] 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).

[0862] 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.

[0863] 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.

[0864] 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.

[0865] 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.

[0866] 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.

[0867] Next, the identification process performed by the identification processing unit 290 of the data processing device 12 will be described.

[0868] "Example of form 1"

[0869] One embodiment of the present invention is a system comprising means for customizing content based on user attributes and preferences, means for generating content using generative AI, and means for providing the generated content to the user. Specifically, user attributes and preferences are learned from user behavior data and feedback, and the generative AI generates content based on the results. The generated content is provided in different formats such as text, audio, video, and design.

[0870] "Example of form 2"

[0871] As a concrete example, consider the case where a user purchases a book from an online bookstore. The system learns the user's preferences and attributes from behavioral data such as purchase history, browsing history, and reviews. Based on this learning, the generative AI generates summaries and recommendations of books that the user might be interested in. The generated content is then displayed and provided to the user the next time they visit the online bookstore.

[0872] "Example of form 3"

[0873] Another scenario is when users utilize music streaming services. The system learns the user's musical preferences and attributes from behavioral data such as their playback history, playlists, and ratings. Based on this learning, the generative AI generates playlists that the user is likely to enjoy. These generated playlists are then displayed and provided to the user the next time they use the music streaming service.

[0874] The following describes the processing flow for each example of the form.

[0875] "Example of form 1"

[0876] Step 1: Collect user behavior data and feedback.

[0877] Step 2: Learn user attributes and preferences from the collected data.

[0878] Step 3: Based on the learning results, the generative AI generates content.

[0879] Step 4: Provide the generated content to the user.

[0880] "Example of form 2"

[0881] Step 1: When a user visits an online bookstore, collect behavioral data such as their purchase history, browsing history, and reviews.

[0882] Step 2: Learn user preferences and attributes from the collected data.

[0883] Step 3: Based on the learning results, the generative AI generates summaries and recommendations of books that the user might be interested in.

[0884] Step 4: Display and provide the generated content to the user the next time they visit the online bookstore.

[0885] "Example of form 3"

[0886] Step 1: When a user uses a music streaming service, collect behavioral data such as the user's playback history, playlists, and ratings.

[0887] Step 2: Learn the user's music preferences and attributes from the collected data.

[0888] Step 3: Based on the learning results, the generative AI generates a playlist that the user is likely to like.

[0889] Step 4: Display and provide the generated playlist to the user the next time they use the music streaming service.

[0890] (Example 1)

[0891] Next, we will describe Embodiment 1 of 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."

[0892] In today's information-saturated society, users face the challenge of efficiently acquiring information that matches their characteristics and preferences. Furthermore, conventional systems are unable to fully utilize user behavior data and opinions, making personalized information delivery difficult.

[0893] 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.

[0894] In this invention, the server includes means for personalizing information based on the user's characteristics and preferences, means for generating information using generative artificial intelligence, and means for providing the generated information to the user. This enables the user to efficiently obtain information that suits their characteristics and preferences.

[0895] "User characteristics" refer to information that indicates a user's behavioral patterns, preferences, interests, and other related information.

[0896] "Preferences" refer to information that indicates the tendencies and preferences that a user particularly likes.

[0897] "Means of personalizing information" refer to methods and technologies for customizing information based on user characteristics and preferences, and providing it in a format suitable for each individual user.

[0898] "Generative artificial intelligence" is an artificial intelligence technology that has the ability to automatically generate new information and content based on given instructions and data.

[0899] "Means of generating information" refers to methods and technologies that use generative artificial intelligence to create information tailored to the user's characteristics and preferences.

[0900] "Means of providing information to users" refers to the methods and technologies used to deliver generated information to users.

[0901] "Behavioral information" refers to data such as the actions a user performs on the system and their browsing history.

[0902] "Opinions" refer to information such as feedback, ratings, and comments provided by users.

[0903] An "instruction statement" is a text used to give instructions to a generative artificial intelligence to generate information.

[0904] To implement this invention, the server needs to generate a program for personalizing information based on the user's characteristics and preferences. This program includes means for collecting user behavior information and opinions, and generating information using generative artificial intelligence based on that information.

[0905] The server uses a database management system (e.g., MySQL) to collect user behavior information and opinions. The collected data is analyzed using data analysis tools (e.g., Python's Pandas library) to identify user characteristics and preferences. Next, a generative artificial intelligence model (e.g., a large-scale language model) is used to generate information based on user characteristics and preferences. This generated information is provided in various formats, such as text, audio, video, and design.

[0906] As a concrete example, consider a scenario where a user requests information about movies. Based on the user's past viewing history and ratings, the server prompts a generative artificial intelligence model with the message, "Create a list of movies that the user might like." Based on this prompt, the generative AI generates a list of movies that match the user's preferences and provides it to the user through the terminal.

[0907] This system allows users to efficiently obtain information that matches their characteristics and preferences.

[0908] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0909] Step 1:

[0910] The server collects user behavior information and opinions. Specifically, it collects data such as user actions on websites and applications, browsing history, ratings, and comments. This data is entered into and stored in a database management system (e.g., MySQL).

[0911] Step 2:

[0912] The server analyzes the collected data. Specifically, it uses the Python Pandas library to organize the data and identify user behavior patterns and preferences. The input is the data collected in step 1, and the output is information about user characteristics and preferences. This analysis reveals trends in categories that users frequently view and content that they rate highly.

[0913] Step 3:

[0914] The server creates prompt statements to be input to the generative artificial intelligence model. Specifically, based on the user's characteristics and preferences obtained in step 2, it generates prompts such as, "Suggest content that the user might be interested in." The input is information about the user's characteristics and preferences, and the output is a prompt statement.

[0915] Step 4:

[0916] The server generates information using a generative artificial intelligence model. Specifically, it inputs the prompt text created in step 3 into a generative artificial intelligence model (e.g., a large-scale language model) to generate information that matches the user's characteristics and preferences. The input is a prompt text, and the output is information in the form of text, audio, video, design, etc.

[0917] Step 5:

[0918] The server sends the generated information to the terminal and provides it to the user. Specifically, it sends notifications to the user's device or displays the information within the application. The input is the information generated in step 4, and the output is the information provided in a format that the user can view. The user can view and enjoy the provided information.

[0919] (Application Example 1)

[0920] Next, we will describe Application Example 1 of Form 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."

[0921] In modern information distribution services, providing content tailored to diverse user preferences and attributes in real time is challenging. In particular, there is a need for technology that leverages user behavior history and feedback to generate and instantly deliver video and audio information optimized for individual users.

[0922] 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.

[0923] In this invention, the server includes means for customizing information based on user attributes and preferences, means for generating information using generative artificial intelligence, and means for providing the generated information to the user. This makes it possible to generate and deliver video and audio information optimized for individual users in real time, based on the user's behavior history and feedback.

[0924] "User attributes" refer to the unique characteristics and traits of each individual user, including age, gender, interests, and preferences.

[0925] "Preferences" refer to the things that users are particularly interested in or tend to prefer, and are inferred from their viewing history and feedback.

[0926] "Means of customizing information" refers to methods and technologies for individually adjusting the information provided based on the user's attributes and preferences.

[0927] "Generative artificial intelligence" refers to artificial intelligence technology that has the ability to automatically generate new information and content based on given data and instructions.

[0928] "Means of generating information" refers to methods and technologies that use generative artificial intelligence to create information tailored to the user's attributes and preferences.

[0929] "Means of providing information" refers to methods and technologies for delivering generated information to users, and includes distribution technologies and display technologies.

[0930] "Behavioral history" refers to a record of a user's past actions, including viewing history and search history.

[0931] "Feedback" refers to the evaluations and opinions provided by users, indicating their reaction to and satisfaction with the content.

[0932] "Means of real-time delivery" refers to methods and technologies for delivering generated information to users immediately, with the aim of providing information without delay.

[0933] The system for carrying out this invention customizes information based on user attributes and preferences, generates information using generative artificial intelligence, and provides the generated information to the user. The system includes a server, a user terminal, and a generative artificial intelligence model.

[0934] The server collects user behavior history and feedback, and uses this to analyze user attributes and preferences. Based on the analysis results, it sends prompts to a generative artificial intelligence (AI) to generate information optimized for the user. For example, OpenAI's GPT-4 is used as the generative AI.

[0935] User devices, such as smartphones and smart glasses, receive generated information in real time and provide it to the user. The information is delivered in video and audio formats, and content tailored to the user's preferences can be viewed instantly.

[0936] As a concrete example, based on the genres and ratings of movies a user has watched in the past, a generative artificial intelligence can generate and provide a new movie trailer to the user. An example of a prompt to input into the generative AI model would be, "Generate a new movie trailer that includes elements of action movies that the user likes."

[0937] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0938] Step 1:

[0939] The server collects user behavior history and feedback. It receives user viewing history and rating data as input and stores this in a database. As output, it generates analytical data showing user attributes and preferences. Specifically, the server executes database queries to aggregate users' past behavior.

[0940] Step 2:

[0941] The server analyzes user attributes and preferences based on the collected data. It uses the analysis data obtained in Step 1 as input and applies a machine learning algorithm. As output, it generates a profile that reflects the user's preferences. Specifically, the server uses a clustering algorithm to classify user interests.

[0942] Step 3:

[0943] The server sends a prompt to the generative artificial intelligence. It uses the user profile generated in step 2 as input to create the prompt text. It receives information generated by the generative artificial intelligence as output. Specifically, the server generates the prompt text in text format and sends it to the generative AI model.

[0944] Step 4:

[0945] Generative artificial intelligence generates information based on received prompts. It receives prompt text from a server as input and generates information using a generative AI model. It returns the generated video or audio information to the server as output. Specifically, the generative AI model generates content using natural language processing techniques.

[0946] Step 5:

[0947] The server delivers the generated information to the user's terminal. It receives the information generated in step 4 as input and sends it to the user's terminal. As output, it provides the information in a format viewable by the user. Specifically, the server delivers the information in real time using streaming technology.

[0948] Step 6:

[0949] The user terminal provides the user with the information it receives. As input, it receives information distributed from the server and displays or plays it for the user. As output, it provides content for the user to view. Specifically, the terminal launches a video or audio player and plays the content.

[0950] (Example 2)

[0951] Next, we will describe Example 2 of the morphological example. 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."

[0952] Traditional online content delivery systems have the challenge of not adequately providing personalized information based on individual user preferences. Furthermore, there is the difficulty in effectively utilizing user behavior data to generate information that is optimal for each user.

[0953] 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.

[0954] In this invention, the server includes means for collecting user behavior information, means for preprocessing the collected behavior information, and means for constructing a model that learns user preferences using the preprocessed behavior information. This makes it possible to provide personalized information based on the individual preferences of the user.

[0955] "User behavior information" refers to data about users' online activities, such as their purchase history, browsing history, and reviews.

[0956] "Preprocessing" refers to processes performed to convert collected data into an analyzable format, such as imputing missing values, removing outliers, and extracting features.

[0957] A "preference-learning model" is a machine learning model built to predict a user's preferences and interests based on their behavioral data.

[0958] A "generative AI model" is an artificial intelligence model that uses natural language processing technology to generate information tailored to the user.

[0959] A "prompt statement" is an instruction given to a generative AI model, used to specify the content and format of the information to be generated.

[0960] "Personalized information provision" refers to providing information that is customized based on the individual user's preferences and attributes.

[0961] This invention is a system that provides personalized information by utilizing user behavior data. The server collects behavioral information such as purchase history, browsing history, and reviews from users on online platforms. This data is stored in a database and used for later analysis.

[0962] The server uses programming languages ​​such as Python and R to preprocess the collected behavioral data. Specifically, it performs tasks such as imputing missing values ​​and removing outliers, and extracts user behavior information as features. This preprocessed data serves as the foundation for applying machine learning algorithms.

[0963] Next, the server uses the pre-processed data to build a model that learns user preferences. This model is built using libraries such as Scikit-learn and TensorFlow. The model analyzes user behavior patterns and predicts information that users are likely to be interested in.

[0964] The generative AI model generates information tailored to the user based on a pre-trained model. The server inputs prompts to the generative AI model, which then generates the information to provide to the user. For example, a prompt such as "Generate a summary of a new mystery novel that the user might be interested in" might be used.

[0965] The device provides the user with generated information. The next time the user visits the online platform, personalized information will be displayed on the screen. This allows users to easily obtain information tailored to their preferences.

[0966] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0967] Step 1:

[0968] The server collects user behavior information. Specifically, it collects data such as purchase history, browsing history, and reviews from users on online platforms and stores it in a database. The input for this step is user behavior information, and the output is the raw data stored in the database.

[0969] Step 2:

[0970] The server preprocesses the collected behavioral information. The input is raw data stored in a database. The server uses Python or R to impute missing values ​​and remove outliers, converting the data into an analyzable format. The output is preprocessed, clean data.

[0971] Step 3:

[0972] The server builds a model that learns user preferences using preprocessed data. The input is clean, preprocessed data. The server applies machine learning algorithms using Scikit-learn and TensorFlow to generate a model that predicts user behavior patterns. The output is the trained preference prediction model.

[0973] Step 4:

[0974] The server inputs prompt text into a generative AI model, which then generates information tailored to the user. The input consists of a trained preference prediction model and prompt text. The server uses the generative AI model to generate information that the user is likely to find interesting. The output is the generated, personalized information.

[0975] Step 5:

[0976] The device provides the user with generated information. The input is the generated, personalized information. The device displays this information on the screen the next time the user visits the online platform. The output is the personalized information displayed to the user.

[0977] (Application Example 2)

[0978] Next, we will describe application example 2 of form 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."

[0979] In today's information-saturated world, it is difficult for users to efficiently find products and information that suit their preferences. Furthermore, traditional recommendation systems struggle to provide accurate recommendations because they do not fully utilize diverse user behavior data.

[0980] 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.

[0981] In this invention, the server includes means for customizing information based on user attributes and preferences, means for generating information using generative artificial intelligence, and means for providing the generated information to the user. This makes it possible to provide individually customized product recommendations by utilizing user behavior data.

[0982] "User attributes" refer to information related to an individual, such as the user's age, gender, interests, and purchase history.

[0983] "Preference" refers to a user's taste or preference for a particular genre or style.

[0984] "Means of customizing information" refers to methods of individually adjusting and optimizing information based on the user's attributes and preferences.

[0985] "Generative artificial intelligence" refers to artificial intelligence technology that has the ability to generate new information and content based on data.

[0986] "Means of generating information" refers to methods of creating new information and content using generative artificial intelligence.

[0987] "Means of providing information" refers to the method of presenting and making available the generated information to the user.

[0988] "Behavioral data" refers to data that records a user's online activity history and operation history.

[0989] "Product recommendation" refers to suggesting products that a user might be interested in, based on their attributes and behavioral data.

[0990] A "communication terminal" is an electronic device, such as a smartphone or tablet, that can receive and display information.

[0991] The system for implementing this invention uses generative artificial intelligence to provide product recommendations based on user behavior data. The server customizes information based on user attributes and preferences and generates information using generative artificial intelligence. Specifically, it collects user behavior data and inputs it into a generative artificial intelligence model to generate optimal product recommendations for the user. The generated product recommendations are provided to the user via a communication terminal.

[0992] This system utilizes communication devices such as smartphones and tablets. The server is programmed using Python, and the Pandas library is used for data processing. OpenAI's GPT is employed as the generative artificial intelligence model. User behavior data (purchase history, browsing history, reviews, etc.) is processed using Pandas and input into the GPT model. The model generates product recommendations tailored to the user's preferences and displays them on the communication device.

[0993] As a concrete example, it's possible to recommend similar new books based on the genres and authors of books a user has previously purchased. An example of a prompt to the generative AI model might be, "Based on the user's past purchase history, please recommend new books that might interest them." In this way, users can efficiently find products that suit their preferences.

[0994] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0995] Step 1:

[0996] The server collects user behavior data. Specifically, it retrieves data such as user purchase history, browsing history, and reviews from a database. This data serves as input for understanding user attributes and preferences.

[0997] Step 2:

[0998] The server preprocesses the collected behavioral data using the Pandas library. Specifically, it cleans and formats the data, converting it into a format suitable for the generative AI model. This process prepares the input data for the generative AI model.

[0999] Step 3:

[1000] The server inputs pre-processed data into a generative AI model (OpenAI's GPT). The prompt used is, "Recommend new books that the user might be interested in, based on their past purchase history." The generative AI model uses this prompt and the input data to generate product recommendations that are best suited to the user.

[1001] Step 4:

[1002] The server receives product recommendations output from the generated AI model and sends them to the communication terminal. Specifically, it converts the generated recommendation results into a data format that can be displayed on the user's smartphone or tablet. This output data becomes the product recommendation information provided to the user.

[1003] Step 5:

[1004] The terminal displays product recommendations received from the server to the user. Specifically, it visually presents recommended product information on the terminal's screen, allowing the user to browse and select items. This step enables the user to efficiently find products that suit their preferences.

[1005] (Example 3)

[1006] Next, we will describe Embodiment 3 of Embodiment Example 3. 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."

[1007] Traditional content delivery systems fail to adequately provide content tailored to individual user preferences, highlighting the need for improved user experience. Furthermore, effectively utilizing user behavior data to generate highly accurate content remains a challenge.

[1008] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 3 is realized by the following means.

[1009] In this invention, the server includes means for collecting user behavior information, means for preprocessing the collected behavior information, means for learning user preferences from the preprocessed behavior information, means for generating content based on user preferences using a generative AI model, and means for providing the generated content to the user. This makes it possible to provide highly accurate content based on the individual preferences of the user.

[1010] "User behavior information" refers to data such as playback history, ratings, and playlists that are generated when users use the system.

[1011] "Preprocessing" refers to processes such as imputing missing values ​​and normalizing data, which are performed to convert collected data into a format that is easy to analyze.

[1012] "User preferences" refers to information that indicates the trends and patterns of content that users like.

[1013] A "generative AI model" refers to an artificial intelligence model used to generate content based on user preferences.

[1014] "Content" refers to information such as music, videos, and text provided to users.

[1015] The following systems are conceivable as embodiments for carrying out this invention.

[1016] The server collects user behavior information when users use music streaming services. This behavioral information includes playback history, ratings, and playlists. The server performs preprocessing, such as imputing missing values ​​and normalizing data, to convert the collected data into a format that is easy to analyze. Database management systems and data processing software can be used for this preprocessing.

[1017] Next, the server learns user preferences using pre-processed data. This learning process uses software that implements machine learning algorithms (e.g., TensorFlow or PyTorch). Based on the learning results, the server utilizes a generative AI model to generate content based on user preferences. This generative AI model uses natural language processing techniques to select music that matches the user's taste.

[1018] The generated content is delivered to the user through their device. The next time the user uses a music streaming service, the generated playlist will appear on their home screen. By playing the suggested playlist, the user can enjoy a new musical experience.

[1019] For example, if the server learns that the user prefers "rock" music, it will generate a playlist containing the latest rock songs. An example of a prompt might be, "Create a playlist based on the user's preferences." This enables the provision of highly accurate content based on the user's individual preferences. The specific processing flow in Example 3 will be explained using Figure 15.

[1020] Step 1:

[1021] The server collects behavioral information such as playback history, ratings, and playlists when users utilize music streaming services. It receives user operation logs and rating data as input and stores this information in a database. The output is a record of each user's behavioral information. Specifically, the server retrieves data in real time via an API and stores it in the database.

[1022] Step 2:

[1023] The server preprocesses the collected behavioral information. It receives the raw data collected in step 1 as input, and performs data imputation and normalization. The output is data converted into a format suitable for analysis. Specifically, the server uses data cleaning tools to impute missing values ​​with the mean and standardize numerical data.

[1024] Step 3:

[1025] The server learns user preferences using pre-processed data. It receives the pre-processed data from step 2 as input and applies a machine learning algorithm. The output is a model that represents user preferences. Specifically, the server uses a machine learning framework to perform collaborative filtering and content-based filtering.

[1026] Step 4:

[1027] The server uses a generative AI model based on the learning results to generate content based on the user's preferences. It receives the user preference model obtained in step 3 as input and inputs a prompt message into the generative AI model. The output is a playlist optimized for the user. Specifically, the server inputs the prompt message "Create a playlist based on the user's preferences" into the generative AI model and retrieves the generated playlist.

[1028] Step 5:

[1029] The terminal provides the generated playlist to the user. It receives the playlist generated in step 4 as input and displays it on the user's device. The output is the display of the playlist on the user's device. Specifically, the terminal displays the playlist through the user interface, allowing the user to start playback.

[1030] (Application Example 3)

[1031] Next, we will describe application example 3 of form example 3. In the following description, the data processing device 12 will be referred to as a "server," and the headset-type terminal 314 will be referred to as a "terminal."

[1032] In modern content delivery services, providing content tailored to diverse user preferences and attributes in real time is challenging. In particular, there is a need for technology that leverages user behavior history to quickly generate and deliver personalized, recommended content.

[1033] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means.

[1034] In this invention, the server includes means for customizing content based on user attributes and preferences, means for generating content using generative AI, means for providing the generated content to the user, means for analyzing the user's behavior history and generating recommended content in real time, and means for displaying the generated recommended content on the user's terminal. This makes it possible to provide content tailored to the user's preferences in real time.

[1035] "User attributes" refer to personal information about the user, such as age, gender, region, and interests.

[1036] "Preference" refers to a user's taste or preference for specific content or genres.

[1037] "Means of customizing content" refer to methods and technologies for adjusting and personalizing content based on user attributes and preferences.

[1038] "Generative AI" refers to a system that automatically generates new content using artificial intelligence technology.

[1039] "Means of generating content" refers to methods and technologies for creating new content using generative AI.

[1040] "Means of delivering content" refers to the methods and technologies used to deliver generated content to users.

[1041] "Behavioral history" refers to records of actions, choices, and viewing history that a user has performed in the past.

[1042] "Methods for generating recommended content in real time" refer to methods and technologies for instantly analyzing a user's behavior history and generating the most suitable content on the spot.

[1043] "Means of displaying on a device" refers to methods and technologies for visually presenting generated content on a user's device.

[1044] The system for implementing this invention customizes content based on user attributes and preferences, generates content using generative AI, and provides the generated content to the user. The server analyzes the user's behavior history and generates recommended content in real time. The generated recommended content is displayed on the user's device.

[1045] The server collects the user's music playback history and rating data, and uses this to learn the user's musical preferences. The learning results are input into a generative AI model, which generates a playlist tailored to the user. For example, OpenAI's GPT-3 is used as this generative AI model.

[1046] The user's device receives the generated playlist and displays it within the application. Each time the user opens the application, new recommended playlists are displayed in real time.

[1047] For example, if a user has recently been listening to a lot of jazz and classical music, the server will generate a new playlist focusing on these genres. An example of a prompt to the generation AI model would be: "Based on the user's recent listening history, we know they like jazz and classical music. Please generate a new playlist that the user will enjoy based on this."

[1048] In this way, users can easily find music that suits their preferences, resulting in a more fulfilling music experience.

[1049] The flow of the specific processing in Application Example 3 will be explained using Figure 16.

[1050] Step 1:

[1051] The server collects users' music playback history and rating data. The input is the user's past playback history data, and the output is a dataset for analysis. This dataset is used as foundational information to learn user preferences.

[1052] Step 2:

[1053] The server learns the user's music preferences based on the collected data. The input is the dataset obtained in step 1, and the output is a profile indicating the user's music preferences. As part of data processing, a machine learning algorithm is used to extract the user's preference patterns.

[1054] Step 3:

[1055] The server inputs the user's profile into the generative AI model and generates a playlist tailored to the user. The input is the user profile obtained in step 2, and the output is the recommended playlist. The generative AI model generates the playlist using prompts. An example of such a prompt is, "Based on the user's recent playback history, we know they like jazz and classical music. Based on this, please generate a new playlist that the user will enjoy."

[1056] Step 4:

[1057] The server sends the generated playlist to the user's device. The input is the playlist generated in step 3, and the output is the playlist displayed on the user's device. The device displays the received playlist in the application, allowing the user to play it immediately.

[1058] Step 5:

[1059] The user checks the playlist displayed on their device and starts playback. The input is the playlist displayed on the device, and the output is the user's music playback experience. Through the generated playlist, the user can discover and enjoy new music.

[1060] 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.

[1061] "Example of form 1"

[1062] One embodiment of the present invention involves a system that combines an emotion engine to recognize user emotions. This system learns user emotions from user behavior data and feedback, and a generative AI uses the results to generate content. Specifically, it grasps user emotions from behavior data such as when users browse books on online bookstores and their reactions to reviews, and based on those emotions, the generative AI generates book recommendations that the user is likely to enjoy.

[1063] "Example of form 2"

[1064] Another embodiment involves user emotion recognition in music streaming services. The system understands the user's emotions from behavioral data when they listen to a particular song and their evaluation of the song, and then a generative AI generates a playlist that the user is likely to like based on those emotions.

[1065] "Example of form 3"

[1066] Another embodiment involves user emotion recognition in a movie recommendation service. The system captures the user's emotions from behavioral data when they watch a movie and their ratings of the movie, and then a generative AI generates movie recommendations that the user is likely to enjoy based on those emotions.

[1067] The following describes the processing flow for each example of the form.

[1068] "Example of form 1"

[1069] Step 1: The user browses books at an online bookstore.

[1070] Step 2: The system collects user behavior data and responses to reviews.

[1071] Step 3: The emotion engine learns the user's emotions from the data it collects.

[1072] Step 4: The generative AI generates book recommendations that the user will likely enjoy, based on the emotions it has learned.

[1073] "Example of form 2"

[1074] Step 1: The user listens to a song on a music streaming service.

[1075] Step 2: The system collects user behavior data and song ratings.

[1076] Step 3: The emotion engine learns the user's emotions from the data it collects.

[1077] Step 4: The generative AI generates a playlist that the user is likely to like based on the emotions it has learned.

[1078] "Example of form 3"

[1079] Step 1: The user watches a movie using the movie recommendation service.

[1080] Step 2: The system collects user behavior data and movie ratings.

[1081] Step 3: The emotion engine learns the user's emotions from the data it collects.

[1082] Step 4: The generative AI generates movie recommendations that the user would likely enjoy, based on the emotions it has learned.

[1083] (Example 1)

[1084] Next, we will describe Embodiment 1 of 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."

[1085] Conventional information provision systems did not adequately customize information according to user attributes and preferences, making it difficult to provide users with the most relevant information. Furthermore, there was a lack of effective means to utilize user behavior data and feedback, resulting in challenges in generating information that met user needs.

[1086] 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.

[1087] In this invention, the server includes means for customizing information based on user attributes and preferences, means for generating information using generative artificial intelligence, and means for providing the generated information to the user. This makes it possible to provide information optimized for the user by utilizing user behavior data and feedback.

[1088] "User attributes" refer to personal characteristics of users, such as their age, gender, interests, and preferences.

[1089] "Preferences" refer to the user's particular interests, such as genres, themes, and styles.

[1090] "Means of customizing information" refers to methods and technologies for adjusting the content and format of information provided based on user attributes and preferences.

[1091] "Generative artificial intelligence" refers to artificial intelligence technology that uses machine learning and natural language processing techniques to automatically generate new information and content.

[1092] "Means of generating information" refers to methods and technologies that use generative artificial intelligence to create information tailored to the user's attributes and preferences.

[1093] "Means of providing information" refers to the methods and technologies used to deliver generated information to users.

[1094] "Behavioral data" refers to data such as the actions, choices, and browsing history that users perform online.

[1095] "Feedback" refers to the reactions and evaluations that users give to the information provided.

[1096] An "instruction statement" refers to a sentence used to tell a generative artificial intelligence what kind of information it should generate.

[1097] A server plays a central role in implementing this invention. The server collects user behavior data and feedback and stores it in a database. General-purpose database software can be used as the database management system. The collected data is analyzed using data analysis tools such as Python's Pandas library. This analysis identifies user attributes and preferences.

[1098] Next, the server creates a prompt message to input to the generative artificial intelligence based on the analysis results. For example, a model using natural language processing techniques can be used as the generative AI. The prompt message includes instructions for generating information tailored to the user's interests and preferences.

[1099] The generated information is provided to the user through the device. The device displays the information in a format accessible to the user via a web browser or mobile application. This allows users to easily obtain information that matches their interests.

[1100] For example, if a user frequently browses mystery novels on an online bookstore, the server analyzes this behavioral data and identifies that the user is interested in mystery novels. It then creates a prompt for the generative artificial intelligence (AI) stating, "The user is interested in mystery novels. Based on recent behavioral data, please generate book recommendations that may interest them." Based on this prompt, the AI ​​generates recommendations tailored to the user and provides them to the user through the terminal.

[1101] In this way, the system can efficiently provide information tailored to the user's attributes and preferences.

[1102] The flow of the specific processing in Example 1 will be explained using Figure 17.

[1103] Step 1:

[1104] The server collects user behavior data and feedback. Inputs include user actions, selections, and browsing history on online platforms. This data is stored in a database. Specifically, the server records user clicks and browsing time in real time. Output is the accumulation of user behavior data in the database.

[1105] Step 2:

[1106] The server analyzes the collected data. The input is user behavior data stored in a database. The server uses the Python Pandas library to analyze the data and identify user attributes and preferences. Specifically, the server aggregates data from the past month and identifies the most frequently viewed genres. The output identifies user attributes and preferences.

[1107] Step 3:

[1108] The server creates prompt text to input to the generative artificial intelligence based on the analysis results. The input includes information about the user's attributes and preferences. Based on this, the server creates a prompt text such as, "The user is interested in mystery novels. Based on recent behavioral data, please generate book recommendations that will interest them." In its specific operation, the server generates instructions that reflect the user's interests. The output is a prompt text to input to the generative artificial intelligence.

[1109] Step 4:

[1110] The server generates information using generative artificial intelligence. The input is a prompt. The generative AI uses natural language processing techniques to generate information relevant to the user. Specifically, the generative AI analyzes the prompt and generates book recommendations that the user might be interested in. The output is the information provided to the user.

[1111] Step 5:

[1112] The device provides the user with generated information. The input is information generated by a generative artificial intelligence. The device displays the information in a format accessible to the user via a web browser or mobile application. Specifically, when the user logs in, the device displays personalized recommended content on the homepage. The output is information that the user is interested in.

[1113] (Application Example 1)

[1114] Next, we will describe Application Example 1 of Form 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."

[1115] In modern information distribution services, it is difficult to appropriately provide content that caters to the diverse emotions and preferences of users. In particular, there is a demand for generating and delivering customized content based on users' emotions in real time, but conventional technologies cannot efficiently achieve this.

[1116] 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.

[1117] In this invention, the server includes means for customizing information based on user attributes and preferences, means for generating information using generative artificial intelligence, means for providing the generated information to the user, means for emotion analysis to recognize the user's emotions, and means for the generative artificial intelligence to generate information based on the emotions recognized by the emotion analysis means. This enables the real-time generation and provision of customized content that corresponds to the user's emotions and preferences.

[1118] "User attributes" refer to information related to an individual, such as the user's age, gender, interests, and preferences.

[1119] "Preferences" refer to information that indicates what a user is particularly interested in or what their preferences tend to be.

[1120] "Means of customizing information" refer to methods and technologies for adjusting and personalizing information based on the user's attributes and preferences.

[1121] "Generative artificial intelligence" refers to artificial intelligence technology that has the ability to automatically generate new information using machine learning and deep learning.

[1122] "Means of generating information" refers to methods and technologies for creating new information using generative artificial intelligence.

[1123] "Means of providing information" refers to the methods and technologies used to deliver generated information to users.

[1124] "Emotional analysis methods" refer to methods and techniques for recognizing and analyzing emotions from users' behavior and feedback.

[1125] "Means of generating information based on emotions" refers to methods and technologies for generative artificial intelligence to create information based on recognized emotions.

[1126] The system for implementing this invention mainly consists of a server and a terminal. The server has a database for customizing information based on user attributes and preferences, and generates information using generative artificial intelligence. The generated information is provided to the user through the terminal.

[1127] The server executes generative artificial intelligence models using software such as Python and TensorFlow. User behavior data and feedback are stored in a database and analyzed by sentiment analysis tools. It is possible to recognize user emotions in real time using libraries such as OpenCV.

[1128] The terminal is a device such as a smartphone or tablet, which displays information generated through its user interface. Users provide information to the system through viewing history and feedback, and the system uses this information to generate even more accurate information.

[1129] For example, if a user feels the need to relax, the emotion analysis system recognizes that emotion, and the generative artificial intelligence generates relaxing music or videos, which are then provided to the device. An example of a prompt message would be, "Generate music suitable for when the user feels the need to relax."

[1130] In this way, the system can generate and deliver customized content in real time that is tailored to the user's emotions and preferences.

[1131] The flow of a specific process in Application Example 1 will be explained using Figure 18.

[1132] Step 1:

[1133] The server receives behavioral data and feedback sent from the user's device. This input data includes information about the user's viewing history and emotions. The server stores this data in a database in preparation for later processing.

[1134] Step 2:

[1135] The server uses stored behavioral data and feedback to analyze the user's emotions using emotion analysis tools. Specifically, it uses libraries such as OpenCV to analyze the user's facial expressions and behavioral patterns to identify the user's emotional state. The output of this process is data indicating the user's current emotional state.

[1136] Step 3:

[1137] The server uses the user's emotional state, obtained through emotion analysis, as input to execute a generative artificial intelligence model. Using frameworks such as TensorFlow, it generates content appropriate to the user's emotions. The output of this process is customized content that matches the user's emotions.

[1138] Step 4:

[1139] The server sends the generated content to the user's device. The device displays the received content through a user interface. The user views the provided content and contributes to improving the system's accuracy by sending feedback to the server as needed.

[1140] Step 5:

[1141] The user sends feedback on the content they have watched from their device to the server. The server stores this feedback in a database and uses it to generate future content. This allows the system to provide content that is more tailored to the user's preferences and emotions.

[1142] (Example 2)

[1143] Next, we will describe Example 2 of the morphological example. 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."

[1144] Traditional content delivery systems have faced challenges in providing content that appeals to users because they do not adequately customize content based on individual user preferences and attributes. Furthermore, they have not been able to effectively utilize user behavior data to generate content using AI-generated models.

[1145] 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.

[1146] In this invention, the server includes means for collecting user behavior information, means for analyzing the collected behavior information, and means for generating content using a generative AI model based on the analysis results. This makes it possible to generate and provide content based on the individual preferences and attributes of the user.

[1147] "User behavior information" refers to data such as purchase history, browsing history, and ratings information that users have on online platforms.

[1148] "Means of collection" refers to the technical means of storing user behavior information in a database.

[1149] "Means of analysis" refers to technical means used to analyze collected user behavior information and understand user preferences and trends.

[1150] A "generative AI model" refers to an artificial intelligence model used to generate content based on user behavior data.

[1151] "Means of generating content" refers to technical means of creating user-friendly content using generative AI models.

[1152] "Means of providing to the terminal" refers to the technical means of displaying the generated content on the user's device.

[1153] A "prompt sentence" is an instruction sentence input into a generative AI model, and it refers to a sentence that determines the direction of content generation.

[1154] This invention is a system that generates and provides content to users using a generative AI model based on user behavior information. The server uses a database management system to collect user behavior information. Specifically, it uses a database such as MySQL to store user purchase history, browsing history, and rating information. This allows for the systematic management of user behavior information.

[1155] The server uses the Python pandas library to analyze the collected behavioral information. This allows for data analysis to understand user preferences and trends. The analysis results are then input into a generative AI model.

[1156] For example, OpenAI's GPT-3 is used as the generative AI model. The server generates prompt sentences based on the analysis results and inputs these prompt sentences into the generative AI model to generate content suitable for the user. An example of a prompt sentence is, "Based on the user's past purchase history, please recommend a mystery novel to read next."

[1157] The generated content is delivered to the user's device via a server. Specifically, the generated summaries and recommendations are displayed in the user's browser through a web server (e.g., Apache). This allows users to easily browse content that suits their preferences.

[1158] For example, if a user has previously purchased many mystery novels, the AI ​​model will generate a summary of a new mystery novel and display it on their next visit. In this way, content tailored to the user's individual preferences and attributes is provided.

[1159] The flow of the specific processing in Example 2 will be explained using Figure 19.

[1160] Step 1:

[1161] The server collects user behavior information. Specifically, it stores user purchase history, browsing history, and rating information on online platforms in a database. The input is user behavior data, and the output is this data stored in the database. A database management system (e.g., MySQL) is used to systematically manage the data.

[1162] Step 2:

[1163] The server analyzes the collected behavioral information. The input is user behavioral information stored in a database, and the output is analysis results showing user preferences and trends. The Python pandas library is used to organize and aggregate the data to understand user behavior patterns. Specifically, it analyzes purchase frequency and browsing trends.

[1164] Step 3:

[1165] The server generates content using a generative AI model based on the analysis results. The input is analysis results indicating the user's preferences and tendencies, and the output is content tailored to the user. A prompt is input to the generative AI model (e.g., OpenAI's GPT-3) to generate content that will interest the user. A specific example uses the prompt: "Based on the user's past purchase history, recommend a mystery novel to read next."

[1166] Step 4:

[1167] The server provides the generated content to the user's terminal. The input is the generated content, and the output is the content displayed on the user's terminal. The generated summaries and testimonials are displayed in the user's browser via a web server (e.g., Apache). Specifically, when a user visits an online platform, the content is displayed in the browser.

[1168] (Application Example 2)

[1169] Next, we will describe application example 2 of form 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."

[1170] It is difficult for users to choose the most suitable content from a vast amount of information. Furthermore, there is a need to provide personalized content based on user preferences and attributes, but traditional methods are insufficient to address this. Additionally, there is a lack of effective means to utilize user viewing history and rating data to accurately recommend the next content they should watch.

[1171] 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.

[1172] In this invention, the server includes means for customizing content based on user attributes and preferences, means for generating content using generative AI, and means for providing the generated content to the user. This makes it possible to collect the user's viewing history and evaluation data, input it into a generative AI model, learn the user's preferences, and generate summaries and recommendations of content that the user should watch next.

[1173] "User attributes" refer to personal characteristics of the user, such as age, gender, interests, and preferences.

[1174] "Preference" refers to a user's personal preference for specific content or genres.

[1175] "Means of customizing content" refers to methods and technologies for personalizing the content provided based on user attributes and preferences.

[1176] "Generative AI" refers to a system that automatically generates new content using artificial intelligence technology.

[1177] "Generated content" refers to information such as text, audio, video, and design created by generative AI.

[1178] "Viewing history" refers to a record of content that a user has watched in the past.

[1179] "Rating data" refers to information about the ratings and feedback that users have given to the content they have watched.

[1180] "Generative AI models" refer to the algorithms and learning models that form the core of generative AI.

[1181] A "summary" refers to information that concisely summarizes the main points of a piece of content.

[1182] A "recommendation letter" refers to a piece of writing created to recommend specific content to a user.

[1183] The system for implementing this invention collects the user's viewing history and evaluation data, inputs it into a generative AI model to learn the user's preferences, and generates summaries and recommendations for content the user should watch next.

[1184] The server stores user viewing history and rating data in a database. Cloud-based data storage such as Firebase can be used as the database. The server preprocesses this data and converts it into a format suitable for generative AI models. OpenAI's GPT-3 can be used as a generative AI model.

[1185] Generative AI models learn user preferences based on their viewing history and rating data. Based on this learning, they generate summaries and recommendations for content the user should watch next. The generated content is then sent to the user's device and provided to them.

[1186] As a concrete example, a generative AI model generates a summary of the next movie a user should watch, based on the genre and rating of the movies they have recently watched. An example of a prompt to input to the generative AI model might be: "The user recently watched an action movie with a high rating. Please generate a summary of the next movie you would recommend for this user."

[1187] The flow of a specific process in Application Example 2 will be explained using Figure 20.

[1188] Step 1:

[1189] The server collects user viewing history and rating data. Information about the content viewed by the user and their ratings are sent from the user's device to the server. The input is the user's viewing history and rating data, and the output is the storage of this data in a database.

[1190] Step 2:

[1191] The server preprocesses the collected viewing history and evaluation data. Specifically, it cleans and converts the data to a format suitable for the generative AI model. The input is the viewing history and evaluation data stored in the database, and the output is the preprocessed data.

[1192] Step 3:

[1193] The server inputs pre-processed data into a generative AI model. The generative AI model learns the user's preferences and generates summaries and recommendations for content to watch next. The input is pre-processed data, and the output is the generated summaries and recommendations.

[1194] Step 4:

[1195] The server sends the generated summary and recommendations to the user's device. The user receives information about the next content to watch through their device. The input is the generated summary and recommendations, and the output is the content information displayed on the user's device.

[1196] Step 5:

[1197] Users select the next content to watch based on summaries and recommendations displayed on their device. This selection is sent back to the server to influence future recommendations. The input is the user's selection, and the output is the updated viewing history and evaluation data.

[1198] (Example 3)

[1199] Next, we will describe Embodiment 3 of Embodiment Example 3. 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."

[1200] Traditional content delivery systems have a challenge in that they do not adequately provide content based on individual user preferences. In particular, there is a need to effectively utilize user behavior history and evaluation data to generate content that is optimal for each user.

[1201] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 3 is realized by the following means.

[1202] In this invention, the server includes means for collecting user behavior history, means for analyzing the collected behavior history and learning user preferences, and means for generating content based on user preferences using a generative AI model. This makes it possible to provide optimal content tailored to the individual preferences of each user.

[1203] "User activity history" refers to the record of a series of operations and choices that a user makes when using a system.

[1204] "Preferences" refer to the likes and tendencies that users show towards specific content or services.

[1205] A "generative AI model" is an algorithm or system that uses artificial intelligence technology to generate new content based on input data.

[1206] "Content" is a general term for information and media provided to users, such as music, videos, text, and images.

[1207] "Means of collection" refers to methods and technologies for acquiring and storing user behavior history and evaluation data.

[1208] "Means of analysis" refers to methods and techniques for processing collected data and understanding user preferences and behavioral patterns.

[1209] "Means of delivery" refers to the methods and technologies used to present and make available the generated content to users.

[1210] This invention is a system that provides content tailored to individual preferences based on the user's behavior history. The server collects and analyzes the user's behavior history to learn their preferences. Specifically, the server uses a database to store the user's playback history and evaluation data, and analyzes this data using machine learning algorithms. For analysis, the Python library Scikit-learn is used to perform clustering and classification.

[1211] The generative AI model used will utilize natural language processing technology. For example, by using OpenAI's GPT-3, it is possible to generate content based on user preferences. The generated content will be provided in various formats, including music, videos, and text.

[1212] The device displays content provided by the server to the user. The user can view, play, or use the content provided through the device. For example, if the user prefers rock music, the server will generate a playlist centered around rock music and display it on the device. An example of a prompt message would be, "Please generate a recommended playlist based on the user's playback history."

[1213] This system allows users to efficiently enjoy content tailored to their preferences. The flow of specific processing in Example 3 will be explained using Figure 21.

[1214] Step 1:

[1215] The server collects user activity history. Inputs include user playback history, playlists, and rating data. This data is stored in a database. Specifically, the server monitors user activity in real time and periodically updates the data.

[1216] Step 2:

[1217] The server analyzes the collected behavioral history to learn user preferences. The input is the data collected in Step 1. The server processes the data using machine learning algorithms to generate a user preference model. Specifically, it uses Python's Scikit-learn to perform clustering and classification to identify user preferences. The output is a model that shows the user's preferences.

[1218] Step 3:

[1219] The server generates content based on user preferences using a generative AI model. The input is the user preference model obtained in step 2. The generative AI model utilizes natural language processing techniques and generates content in response to prompt text. Specifically, it uses OpenAI's GPT-3 to generate playlists and recommendations that the user is likely to enjoy. The output is the generated content.

[1220] Step 4:

[1221] The device provides the generated content to the user. The input is the content generated in step 3. The device displays the content through the user interface, making it available to the user. Specifically, the device uses a notification function to inform the user of new content and displays playlists and recommendations on the screen. The output is the content available to the user.

[1222] (Application Example 3)

[1223] Next, we will describe application example 3 of form example 3. In the following description, the data processing device 12 will be referred to as a "server," and the headset-type terminal 314 will be referred to as a "terminal."

[1224] In modern content distribution services, providing content that caters to diverse user preferences is crucial. However, traditional systems struggle to fully utilize user behavior data, making it difficult to deliver content optimized for individual users. In particular, music streaming services require the automatic generation of personalized playlists based on users' listening history and ratings.

[1225] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means.

[1226] In this invention, the server includes means for customizing content based on user attributes and preferences, means for generating content using generative AI, means for providing the generated content to the user, means for collecting user behavior data and preprocessing it for input into a generative AI model, and means for analyzing the user's music playback history and ratings and automatically generating playlists that match the user's preferences. This makes it possible to provide personalized content that meets the individual preferences of each user.

[1227] "User attributes" refer to personal information about the user, such as age, gender, region, and interests.

[1228] "Preference" refers to a user's taste or preference for specific content or genres.

[1229] "Means of customizing content" refer to methods and technologies for individually adjusting the content provided based on the user's attributes and preferences.

[1230] "Generative AI" refers to a system that automatically generates new content using artificial intelligence technology.

[1231] "Means of generating content" refers to methods and technologies for creating new content such as music, videos, and text using generative AI.

[1232] "Means of delivering content" refers to the methods and technologies used to deliver generated content to users.

[1233] "Behavioral data" refers to data such as playback history, ratings, and click history that is generated when users use a service.

[1234] "Means of preprocessing" refer to methods and techniques for converting collected behavioral data into a format suitable for generative AI models.

[1235] "Music playback history" is a record of the music a user has played in the past.

[1236] "Rating" refers to the positive or negative feedback that users give to content.

[1237] "Methods for automatically generating playlists" refer to methods and technologies for automatically creating music lists based on the user's preferences.

[1238] The system for implementing this invention automatically generates personalized playlists using a generative AI model based on the user's music playback history and ratings. The system mainly consists of a server and the user's terminal.

[1239] The server collects user behavior data and preprocesses it for input into a generating AI model. Specifically, it uses Python to retrieve user playback history and rating data from music streaming service APIs. This data is used to analyze user preferences.

[1240] OpenAI's GPT-3 is used as a generative AI model. This model generates music playlists tailored to the user's preferences based on user behavior data. The generated playlists are provided to the user's device and displayed the next time they use the music streaming service.

[1241] For example, if the user's recently listened-to music genres are pop and rock, the generative AI model will suggest a playlist containing new artists and songs based on this. An example of a prompt might be, "The user's recently listened-to music genres are pop and rock. Please generate a new playlist based on this."

[1242] This system makes it easy for users to discover new songs that suit their musical preferences, enriching their musical experience.

[1243] The flow of the specific processing in Application Example 3 will be explained using Figure 22.

[1244] Step 1:

[1245] The server retrieves user playback history and rating data through the music streaming service's API. The input is the user ID, and the output is the user's playback history and rating data. This data serves as foundational information for analyzing the user's musical preferences.

[1246] Step 2:

[1247] The server preprocesses the acquired playback history and evaluation data. The input is the playback history and evaluation data, and the output is data converted into a format suitable for the generative AI model. Specifically, it performs data normalization and filtering to remove noise.

[1248] Step 3:

[1249] The server inputs pre-processed data into a generative AI model. The input is pre-processed data, and the output is a playlist based on the user's preferences. The generative AI model uses prompts to select music that matches the user's preferences.

[1250] Step 4:

[1251] The server sends the generated playlist to the user's device. The input is the generated playlist, and the output is the playlist displayed on the user's device. Specifically, the server converts the playlist to JSON format and sends it to the device via the API.

[1252] Step 5:

[1253] The user views the playlist displayed on their device and plays the music. The input is the playlist displayed on the device, and the output is the user's music experience. The user can discover and enjoy new music.

[1254] (Other examples)

[1255] Since this is the same as the specific processing described in the other embodiments of the first embodiment above, the explanation will be omitted.

[1256] 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.

[1257] The data generation model 58 is a form of so-called generative AI (Artificial Intelligence). One example of the data generation model 58 is ChatGPT (Internet Search).<URL: https: / / openai.com / blog / chatgpt> 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.

[1258] Other examples of generative AI include Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) are some examples.

[1259] 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.

[1260] [Fourth Embodiment]

[1261] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1262] 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.

[1263] 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).

[1264] 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.

[1265] 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.

[1266] 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).

[1267] 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.

[1268] 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.

[1269] 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.

[1270] 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.

[1271] 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.

[1272] 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.

[1273] Next, the identification process performed by the identification processing unit 290 of the data processing device 12 will be described.

[1274] "Example of form 1"

[1275] One embodiment of the present invention is a system comprising means for customizing content based on user attributes and preferences, means for generating content using generative AI, and means for providing the generated content to the user. Specifically, user attributes and preferences are learned from user behavior data and feedback, and the generative AI generates content based on the results. The generated content is provided in different formats such as text, audio, video, and design.

[1276] "Example of form 2"

[1277] As a concrete example, consider the case where a user purchases a book from an online bookstore. The system learns the user's preferences and attributes from behavioral data such as purchase history, browsing history, and reviews. Based on this learning, the generative AI generates summaries and recommendations of books that the user might be interested in. The generated content is then displayed and provided to the user the next time they visit the online bookstore.

[1278] "Example of form 3"

[1279] Another scenario is when users utilize music streaming services. The system learns the user's musical preferences and attributes from behavioral data such as their playback history, playlists, and ratings. Based on this learning, the generative AI generates playlists that the user is likely to enjoy. These generated playlists are then displayed and provided to the user the next time they use the music streaming service.

[1280] The following describes the processing flow for each example of the form.

[1281] "Example of form 1"

[1282] Step 1: Collect user behavior data and feedback.

[1283] Step 2: Learn user attributes and preferences from the collected data.

[1284] Step 3: Based on the learning results, the generative AI generates content.

[1285] Step 4: Provide the generated content to the user.

[1286] "Example of form 2"

[1287] Step 1: When a user visits an online bookstore, collect behavioral data such as their purchase history, browsing history, and reviews.

[1288] Step 2: Learn user preferences and attributes from the collected data.

[1289] Step 3: Based on the learning results, the generative AI generates summaries and recommendations of books that the user might be interested in.

[1290] Step 4: Display and provide the generated content to the user the next time they visit the online bookstore.

[1291] "Example of form 3"

[1292] Step 1: When a user uses a music streaming service, collect behavioral data such as the user's playback history, playlists, and ratings.

[1293] Step 2: Learn the user's music preferences and attributes from the collected data.

[1294] Step 3: Based on the learning results, the generative AI generates a playlist that the user is likely to like.

[1295] Step 4: Display and provide the generated playlist to the user the next time they use the music streaming service.

[1296] (Example 1)

[1297] Next, we will describe Embodiment 1 of Example Form 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1298] In today's information-saturated society, users face the challenge of efficiently acquiring information that matches their characteristics and preferences. Furthermore, conventional systems are unable to fully utilize user behavior data and opinions, making personalized information delivery difficult.

[1299] 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.

[1300] In this invention, the server includes means for personalizing information based on the user's characteristics and preferences, means for generating information using generative artificial intelligence, and means for providing the generated information to the user. This enables the user to efficiently obtain information that suits their characteristics and preferences.

[1301] "User characteristics" refer to information that indicates a user's behavioral patterns, preferences, interests, and other related information.

[1302] "Preferences" refer to information that indicates the tendencies and preferences that a user particularly likes.

[1303] "Means of personalizing information" refer to methods and technologies for customizing information based on user characteristics and preferences, and providing it in a format suitable for each individual user.

[1304] "Generative artificial intelligence" is an artificial intelligence technology that has the ability to automatically generate new information and content based on given instructions and data.

[1305] "Means of generating information" refers to methods and technologies that use generative artificial intelligence to create information tailored to the user's characteristics and preferences.

[1306] "Means of providing information to users" refers to the methods and technologies used to deliver generated information to users.

[1307] "Behavioral information" refers to data such as the actions a user performs on the system and their browsing history.

[1308] "Opinions" refer to information such as feedback, ratings, and comments provided by users.

[1309] An "instruction statement" is a text used to give instructions to a generative artificial intelligence to generate information.

[1310] To implement this invention, the server needs to generate a program for personalizing information based on the user's characteristics and preferences. This program includes means for collecting user behavior information and opinions, and generating information using generative artificial intelligence based on that information.

[1311] The server uses a database management system (e.g., MySQL) to collect user behavior information and opinions. The collected data is analyzed using data analysis tools (e.g., Python's Pandas library) to identify user characteristics and preferences. Next, a generative artificial intelligence model (e.g., a large-scale language model) is used to generate information based on user characteristics and preferences. This generated information is provided in various formats, such as text, audio, video, and design.

[1312] As a concrete example, consider a scenario where a user requests information about movies. Based on the user's past viewing history and ratings, the server prompts a generative artificial intelligence model with the message, "Create a list of movies that the user might like." Based on this prompt, the generative AI generates a list of movies that match the user's preferences and provides it to the user through the terminal.

[1313] This system allows users to efficiently obtain information that matches their characteristics and preferences.

[1314] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1315] Step 1:

[1316] The server collects user behavior information and opinions. Specifically, it collects data such as user actions on websites and applications, browsing history, ratings, and comments. This data is entered into and stored in a database management system (e.g., MySQL).

[1317] Step 2:

[1318] The server analyzes the collected data. Specifically, it uses the Python Pandas library to organize the data and identify user behavior patterns and preferences. The input is the data collected in step 1, and the output is information about user characteristics and preferences. This analysis reveals trends in categories that users frequently view and content that they rate highly.

[1319] Step 3:

[1320] The server creates prompt statements to be input to the generative artificial intelligence model. Specifically, based on the user's characteristics and preferences obtained in step 2, it generates prompts such as, "Suggest content that the user might be interested in." The input is information about the user's characteristics and preferences, and the output is a prompt statement.

[1321] Step 4:

[1322] The server generates information using a generative artificial intelligence model. Specifically, it inputs the prompt text created in step 3 into a generative artificial intelligence model (e.g., a large-scale language model) to generate information that matches the user's characteristics and preferences. The input is a prompt text, and the output is information in the form of text, audio, video, design, etc.

[1323] Step 5:

[1324] The server sends the generated information to the terminal and provides it to the user. Specifically, it sends notifications to the user's device or displays the information within the application. The input is the information generated in step 4, and the output is the information provided in a format that the user can view. The user can view and enjoy the provided information.

[1325] (Application Example 1)

[1326] Next, we will describe Application Example 1 of Form 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".

[1327] In modern information distribution services, providing content tailored to diverse user preferences and attributes in real time is challenging. In particular, there is a need for technology that leverages user behavior history and feedback to generate and instantly deliver video and audio information optimized for individual users.

[1328] 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.

[1329] In this invention, the server includes means for customizing information based on user attributes and preferences, means for generating information using generative artificial intelligence, and means for providing the generated information to the user. This makes it possible to generate and deliver video and audio information optimized for individual users in real time, based on the user's behavior history and feedback.

[1330] "User attributes" refer to the unique characteristics and traits of each individual user, including age, gender, interests, and preferences.

[1331] "Preferences" refer to the things that users are particularly interested in or tend to prefer, and are inferred from their viewing history and feedback.

[1332] "Means of customizing information" refers to methods and technologies for individually adjusting the information provided based on the user's attributes and preferences.

[1333] "Generative artificial intelligence" refers to artificial intelligence technology that has the ability to automatically generate new information and content based on given data and instructions.

[1334] "Means of generating information" refers to methods and technologies that use generative artificial intelligence to create information tailored to the user's attributes and preferences.

[1335] "Means of providing information" refers to methods and technologies for delivering generated information to users, and includes distribution technologies and display technologies.

[1336] "Behavioral history" refers to a record of a user's past actions, including viewing history and search history.

[1337] "Feedback" refers to the evaluations and opinions provided by users, indicating their reaction to and satisfaction with the content.

[1338] "Means of real-time delivery" refers to methods and technologies for delivering generated information to users immediately, with the aim of providing information without delay.

[1339] The system for carrying out this invention customizes information based on user attributes and preferences, generates information using generative artificial intelligence, and provides the generated information to the user. The system includes a server, a user terminal, and a generative artificial intelligence model.

[1340] The server collects user behavior history and feedback, and uses this to analyze user attributes and preferences. Based on the analysis results, it sends prompts to a generative artificial intelligence (AI) to generate information optimized for the user. For example, OpenAI's GPT-4 is used as the generative AI.

[1341] User devices, such as smartphones and smart glasses, receive generated information in real time and provide it to the user. The information is delivered in video and audio formats, and content tailored to the user's preferences can be viewed instantly.

[1342] As a concrete example, based on the genres and ratings of movies a user has watched in the past, a generative artificial intelligence can generate and provide a new movie trailer to the user. An example of a prompt to input into the generative AI model would be, "Generate a new movie trailer that includes elements of action movies that the user likes."

[1343] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1344] Step 1:

[1345] The server collects user behavior history and feedback. It receives user viewing history and rating data as input and stores this in a database. As output, it generates analytical data showing user attributes and preferences. Specifically, the server executes database queries to aggregate users' past behavior.

[1346] Step 2:

[1347] The server analyzes user attributes and preferences based on the collected data. It uses the analysis data obtained in Step 1 as input and applies a machine learning algorithm. As output, it generates a profile that reflects the user's preferences. Specifically, the server uses a clustering algorithm to classify user interests.

[1348] Step 3:

[1349] The server sends a prompt to the generative artificial intelligence. It uses the user profile generated in step 2 as input to create the prompt text. It receives information generated by the generative artificial intelligence as output. Specifically, the server generates the prompt text in text format and sends it to the generative AI model.

[1350] Step 4:

[1351] Generative artificial intelligence generates information based on received prompts. It receives prompt text from a server as input and generates information using a generative AI model. It returns the generated video or audio information to the server as output. Specifically, the generative AI model generates content using natural language processing techniques.

[1352] Step 5:

[1353] The server delivers the generated information to the user's terminal. It receives the information generated in step 4 as input and sends it to the user's terminal. As output, it provides the information in a format viewable by the user. Specifically, the server delivers the information in real time using streaming technology.

[1354] Step 6:

[1355] The user terminal provides the user with the information it receives. As input, it receives information distributed from the server and displays or plays it for the user. As output, it provides content for the user to view. Specifically, the terminal launches a video or audio player and plays the content.

[1356] (Example 2)

[1357] Next, we will describe Example 2 of the morphological example. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1358] Traditional online content delivery systems have the challenge of not adequately providing personalized information based on individual user preferences. Furthermore, there is the difficulty in effectively utilizing user behavior data to generate information that is optimal for each user.

[1359] 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.

[1360] In this invention, the server includes means for collecting user behavior information, means for preprocessing the collected behavior information, and means for constructing a model that learns user preferences using the preprocessed behavior information. This makes it possible to provide personalized information based on the individual preferences of the user.

[1361] "User behavior information" refers to data about users' online activities, such as their purchase history, browsing history, and reviews.

[1362] "Preprocessing" refers to processes performed to convert collected data into an analyzable format, such as imputing missing values, removing outliers, and extracting features.

[1363] A "preference-learning model" is a machine learning model built to predict a user's preferences and interests based on their behavioral data.

[1364] A "generative AI model" is an artificial intelligence model that uses natural language processing technology to generate information tailored to the user.

[1365] A "prompt statement" is an instruction given to a generative AI model, used to specify the content and format of the information to be generated.

[1366] "Personalized information provision" refers to providing information that is customized based on the individual user's preferences and attributes.

[1367] This invention is a system that provides personalized information by utilizing user behavior data. The server collects behavioral information such as purchase history, browsing history, and reviews from users on online platforms. This data is stored in a database and used for later analysis.

[1368] The server uses programming languages ​​such as Python and R to preprocess the collected behavioral data. Specifically, it performs tasks such as imputing missing values ​​and removing outliers, and extracts user behavior information as features. This preprocessed data serves as the foundation for applying machine learning algorithms.

[1369] Next, the server uses the pre-processed data to build a model that learns user preferences. This model is built using libraries such as Scikit-learn and TensorFlow. The model analyzes user behavior patterns and predicts information that users are likely to be interested in.

[1370] The generative AI model generates information tailored to the user based on a pre-trained model. The server inputs prompts to the generative AI model, which then generates the information to provide to the user. For example, a prompt such as "Generate a summary of a new mystery novel that the user might be interested in" might be used.

[1371] The device provides the user with generated information. The next time the user visits the online platform, personalized information will be displayed on the screen. This allows users to easily obtain information tailored to their preferences.

[1372] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1373] Step 1:

[1374] The server collects user behavior information. Specifically, it collects data such as purchase history, browsing history, and reviews from users on online platforms and stores it in a database. The input for this step is user behavior information, and the output is the raw data stored in the database.

[1375] Step 2:

[1376] The server preprocesses the collected behavioral information. The input is raw data stored in a database. The server uses Python or R to impute missing values ​​and remove outliers, converting the data into an analyzable format. The output is preprocessed, clean data.

[1377] Step 3:

[1378] The server builds a model that learns user preferences using preprocessed data. The input is clean, preprocessed data. The server applies machine learning algorithms using Scikit-learn and TensorFlow to generate a model that predicts user behavior patterns. The output is the trained preference prediction model.

[1379] Step 4:

[1380] The server inputs prompt text into a generative AI model, which then generates information tailored to the user. The input consists of a trained preference prediction model and prompt text. The server uses the generative AI model to generate information that the user is likely to find interesting. The output is the generated, personalized information.

[1381] Step 5:

[1382] The device provides the user with generated information. The input is the generated, personalized information. The device displays this information on the screen the next time the user visits the online platform. The output is the personalized information displayed to the user.

[1383] (Application Example 2)

[1384] Next, we will describe application example 2 of form 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".

[1385] In today's information-saturated world, it is difficult for users to efficiently find products and information that suit their preferences. Furthermore, traditional recommendation systems struggle to provide accurate recommendations because they do not fully utilize diverse user behavior data.

[1386] 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.

[1387] In this invention, the server includes means for customizing information based on user attributes and preferences, means for generating information using generative artificial intelligence, and means for providing the generated information to the user. This makes it possible to provide individually customized product recommendations by utilizing user behavior data.

[1388] "User attributes" refer to information related to an individual, such as the user's age, gender, interests, and purchase history.

[1389] "Preference" refers to a user's taste or preference for a particular genre or style.

[1390] "Means of customizing information" refers to methods of individually adjusting and optimizing information based on the user's attributes and preferences.

[1391] "Generative artificial intelligence" refers to artificial intelligence technology that has the ability to generate new information and content based on data.

[1392] "Means of generating information" refers to methods of creating new information and content using generative artificial intelligence.

[1393] "Means of providing information" refers to the method of presenting and making available the generated information to the user.

[1394] "Behavioral data" refers to data that records a user's online activity history and operation history.

[1395] "Product recommendation" refers to suggesting products that a user might be interested in, based on their attributes and behavioral data.

[1396] A "communication terminal" is an electronic device, such as a smartphone or tablet, that can receive and display information.

[1397] The system for implementing this invention uses generative artificial intelligence to provide product recommendations based on user behavior data. The server customizes information based on user attributes and preferences and generates information using generative artificial intelligence. Specifically, it collects user behavior data and inputs it into a generative artificial intelligence model to generate optimal product recommendations for the user. The generated product recommendations are provided to the user via a communication terminal.

[1398] This system utilizes communication devices such as smartphones and tablets. The server is programmed using Python, and the Pandas library is used for data processing. OpenAI's GPT is employed as the generative artificial intelligence model. User behavior data (purchase history, browsing history, reviews, etc.) is processed using Pandas and input into the GPT model. The model generates product recommendations tailored to the user's preferences and displays them on the communication device.

[1399] As a concrete example, it's possible to recommend similar new books based on the genres and authors of books a user has previously purchased. An example of a prompt to the generative AI model might be, "Based on the user's past purchase history, please recommend new books that might interest them." In this way, users can efficiently find products that suit their preferences.

[1400] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1401] Step 1:

[1402] The server collects user behavior data. Specifically, it retrieves data such as user purchase history, browsing history, and reviews from a database. This data serves as input for understanding user attributes and preferences.

[1403] Step 2:

[1404] The server preprocesses the collected behavioral data using the Pandas library. Specifically, it cleans and formats the data, converting it into a format suitable for the generative AI model. This process prepares the input data for the generative AI model.

[1405] Step 3:

[1406] The server inputs pre-processed data into a generative AI model (OpenAI's GPT). The prompt used is, "Recommend new books that the user might be interested in, based on their past purchase history." The generative AI model uses this prompt and the input data to generate product recommendations that are best suited to the user.

[1407] Step 4:

[1408] The server receives product recommendations output from the generated AI model and sends them to the communication terminal. Specifically, it converts the generated recommendation results into a data format that can be displayed on the user's smartphone or tablet. This output data becomes the product recommendation information provided to the user.

[1409] Step 5:

[1410] The terminal displays product recommendations received from the server to the user. Specifically, it visually presents recommended product information on the terminal's screen, allowing the user to browse and select items. This step enables the user to efficiently find products that suit their preferences.

[1411] (Example 3)

[1412] Next, we will describe Embodiment 3 of Example 3. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1413] Traditional content delivery systems fail to adequately provide content tailored to individual user preferences, highlighting the need for improved user experience. Furthermore, effectively utilizing user behavior data to generate highly accurate content remains a challenge.

[1414] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 3 is realized by the following means.

[1415] In this invention, the server includes means for collecting user behavior information, means for preprocessing the collected behavior information, means for learning user preferences from the preprocessed behavior information, means for generating content based on user preferences using a generative AI model, and means for providing the generated content to the user. This makes it possible to provide highly accurate content based on the individual preferences of the user.

[1416] "User behavior information" refers to data such as playback history, ratings, and playlists that are generated when users use the system.

[1417] "Preprocessing" refers to processes such as imputing missing values ​​and normalizing data, which are performed to convert collected data into a format that is easy to analyze.

[1418] "User preferences" refers to information that indicates the trends and patterns of content that users like.

[1419] A "generative AI model" refers to an artificial intelligence model used to generate content based on user preferences.

[1420] "Content" refers to information such as music, videos, and text provided to users.

[1421] The following systems are conceivable as embodiments for carrying out this invention.

[1422] The server collects user behavior information when users use music streaming services. This behavioral information includes playback history, ratings, and playlists. The server performs preprocessing, such as imputing missing values ​​and normalizing data, to convert the collected data into a format that is easy to analyze. Database management systems and data processing software can be used for this preprocessing.

[1423] Next, the server learns user preferences using pre-processed data. This learning process uses software that implements machine learning algorithms (e.g., TensorFlow or PyTorch). Based on the learning results, the server utilizes a generative AI model to generate content based on user preferences. This generative AI model uses natural language processing techniques to select music that matches the user's taste.

[1424] The generated content is delivered to the user through their device. The next time the user uses a music streaming service, the generated playlist will appear on their home screen. By playing the suggested playlist, the user can enjoy a new musical experience.

[1425] For example, if the server learns that the user prefers "rock" music, it will generate a playlist containing the latest rock songs. An example of a prompt might be, "Create a playlist based on the user's preferences." This enables the provision of highly accurate content based on the user's individual preferences. The specific processing flow in Example 3 will be explained using Figure 15.

[1426] Step 1:

[1427] The server collects behavioral information such as playback history, ratings, and playlists when users utilize music streaming services. It receives user operation logs and rating data as input and stores this information in a database. The output is a record of each user's behavioral information. Specifically, the server retrieves data in real time via an API and stores it in the database.

[1428] Step 2:

[1429] The server preprocesses the collected behavioral information. It receives the raw data collected in step 1 as input, and performs data imputation and normalization. The output is data converted into a format suitable for analysis. Specifically, the server uses data cleaning tools to impute missing values ​​with the mean and standardize numerical data.

[1430] Step 3:

[1431] The server learns user preferences using pre-processed data. It receives the pre-processed data from step 2 as input and applies a machine learning algorithm. The output is a model that represents user preferences. Specifically, the server uses a machine learning framework to perform collaborative filtering and content-based filtering.

[1432] Step 4:

[1433] The server uses a generative AI model based on the learning results to generate content based on the user's preferences. It receives the user preference model obtained in step 3 as input and inputs a prompt message into the generative AI model. The output is a playlist optimized for the user. Specifically, the server inputs the prompt message "Create a playlist based on the user's preferences" into the generative AI model and retrieves the generated playlist.

[1434] Step 5:

[1435] The terminal provides the generated playlist to the user. It receives the playlist generated in step 4 as input and displays it on the user's device. The output is the display of the playlist on the user's device. Specifically, the terminal displays the playlist through the user interface, allowing the user to start playback.

[1436] (Application Example 3)

[1437] Next, we will describe application example 3 of form example 3. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1438] In modern content delivery services, providing content tailored to diverse user preferences and attributes in real time is challenging. In particular, there is a need for technology that leverages user behavior history to quickly generate and deliver personalized, recommended content.

[1439] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means.

[1440] In this invention, the server includes means for customizing content based on user attributes and preferences, means for generating content using generative AI, means for providing the generated content to the user, means for analyzing the user's behavior history and generating recommended content in real time, and means for displaying the generated recommended content on the user's terminal. This makes it possible to provide content tailored to the user's preferences in real time.

[1441] "User attributes" refer to personal information about the user, such as age, gender, region, and interests.

[1442] "Preference" refers to a user's taste or preference for specific content or genres.

[1443] "Means of customizing content" refer to methods and technologies for adjusting and personalizing content based on user attributes and preferences.

[1444] "Generative AI" refers to a system that automatically generates new content using artificial intelligence technology.

[1445] "Means of generating content" refers to methods and technologies for creating new content using generative AI.

[1446] "Means of delivering content" refers to the methods and technologies used to deliver generated content to users.

[1447] "Behavioral history" refers to records of actions, choices, and viewing history that a user has performed in the past.

[1448] "Methods for generating recommended content in real time" refer to methods and technologies for instantly analyzing a user's behavior history and generating the most suitable content on the spot.

[1449] "Means of displaying on a device" refers to methods and technologies for visually presenting generated content on a user's device.

[1450] The system for implementing this invention customizes content based on user attributes and preferences, generates content using generative AI, and provides the generated content to the user. The server analyzes the user's behavior history and generates recommended content in real time. The generated recommended content is displayed on the user's device.

[1451] The server collects the user's music playback history and rating data, and uses this to learn the user's musical preferences. The learning results are input into a generative AI model, which generates a playlist tailored to the user. For example, OpenAI's GPT-3 is used as this generative AI model.

[1452] The user's device receives the generated playlist and displays it within the application. Each time the user opens the application, new recommended playlists are displayed in real time.

[1453] For example, if a user has recently been listening to a lot of jazz and classical music, the server will generate a new playlist focusing on these genres. An example of a prompt to the generation AI model would be: "Based on the user's recent listening history, we know they like jazz and classical music. Please generate a new playlist that the user will enjoy based on this."

[1454] In this way, users can easily find music that suits their preferences, resulting in a more fulfilling music experience.

[1455] The flow of the specific processing in Application Example 3 will be explained using Figure 16.

[1456] Step 1:

[1457] The server collects users' music playback history and rating data. The input is the user's past playback history data, and the output is a dataset for analysis. This dataset is used as foundational information to learn user preferences.

[1458] Step 2:

[1459] The server learns the user's music preferences based on the collected data. The input is the dataset obtained in step 1, and the output is a profile indica...

Claims

[Claim 1] Equipped with a processor, The aforementioned processor, The system collects the user's activity history on designated services and feedback evaluating the user's level of usefulness and interest in the information provided to the user, and stores the collected data in a database. The aforementioned behavioral history includes, if the service is a streaming service, playback history, playlists, and rating data; and if the service is an online bookstore, purchase history, browsing history, and reviews. The numerical data of the behavioral history collected above is standardized by imputing missing values ​​and normalizing, When analyzing the collected data and analyzing the user's behavioral information and opinions, the user's interests are classified by clustering the data using a machine learning algorithm, thereby identifying the user's characteristics. Based on identified characteristics, prompt statements are automatically generated to instruct the generative AI model to generate information, and the instructions for information generation include the generation of at least one of the following: video, audio information, or text in the form of a list or summary. Based on the prompt statement, the AI ​​model is used to generate the content corresponding to the service. The generated content is delivered to the user's terminal. system.

Images