System

Abstract

A system is provided.SOLUTION: A system, comprising: means for collecting user interest data; means for generating a keyword map based on the interest data; means for "reverse" filtering to exclude keywords located at a center of the keyword map and extract new keywords from surrounding keywords; means for collecting new information based on the new keywords; means for generating a thumbnail image associated with the new information; means for providing the new information and the thumbnail image to a user; and means for collecting feedback from the user.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] On the modern Internet, information providers optimize content based on users' interests, reducing opportunities for users to stray from their own range of interests. As a result, phenomena known as filter bubbles and echo chambers have become more pronounced, reducing users' opportunities to be exposed to multifaceted information and new perspectives. Addressing this issue requires providing users with a means to access new information and perspectives. [Means for solving the problem]

[0005] To solve this problem, the present invention provides the following means: a system including a means for collecting user interest data, a means for generating a keyword map based on the interest data, a "reverse" filtering means for excluding keywords located at the center of the keyword map and extracting new keywords from peripheral keywords, a means for collecting new information based on the new keywords, a means for generating thumbnail images related to the new information, a means for providing the new information and the thumbnail images to the user, and a means for collecting feedback from the user. This makes it possible to provide new information that deviates appropriately from the user's known areas of interest, creating opportunities for exposure to diverse perspectives and new knowledge.

[0006] "Interest Data" means data containing information about a User's interests and concerns.

[0007] A "keyword map" is a structured display of key keywords extracted from user interest data.

[0008] "Inverse filtering" is a filtering method that eliminates central keywords and extracts new keywords from peripheral keywords.

[0009] "New information" is information that is novel to the user and related to the keywords extracted by inverse filtering.

[0010] A "thumbnail image" is a small visual image associated with new information.

[0011] A "natural language processing model" is a model that includes algorithms and techniques for analyzing and understanding natural language.

[0012] An "artificial intelligence model" is a computational model that uses artificial intelligence techniques to perform a specific task.

[0013] "Feedback" refers to data that refers to ratings, opinions, and reactions provided by users. [Brief explanation of the drawings]

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

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

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

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

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

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

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

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

[0022] [First embodiment]

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

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

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

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

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

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

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

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

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

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

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

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

[0035] The embodiments for carrying out the present invention will be described in detail below.

[0036] User data collection (server)

[0037] The server collects users' access history, search history, bookmark data, etc. This includes the websites they have visited, the keywords they have searched for, the articles they have viewed, the pages they have bookmarked, etc. The collected data is stored in a database for each individual user and used for later analysis.

[0038] Generate Keyword Mapping (Server)

[0039] The server uses a natural language processing model to extract key keywords from the collected user data. For example, if a user frequently views articles related to "technology," "artificial intelligence," and "programming," these keywords are extracted. The extracted keywords are then mapped based on their relevance to generate a keyword map for each user. This keyword map has a graph structure that displays keywords near the center as the user's main interests and places less relevant keywords on the periphery.

[0040] "Reverse" filtering implementation (server)

[0041] The server analyzes the generated keyword map and eliminates central keywords. For example, if the user's primary interests are "technology," "artificial intelligence," and "programming," these keywords are excluded. It then extracts more peripheral keywords (e.g., "biology," "art," and "psychology"). This "reverse" filtering extracts keywords that the user would not normally be interested in, but which may provide relevant new information.

[0042] Selection of new information (server)

[0043] The server searches and collects relevant information from the Internet based on the new keywords extracted through "reverse" filtering, such as articles, news, and research papers related to the new keywords. The collected information is stored in a database and later provided to users.

[0044] Thumbnail generation (server)

[0045] The server uses artificial intelligence models to generate thumbnail images related to new information collected. For example, for articles related to "biology," the AI ​​model might generate diagrams of DNA or images of cells. The generated thumbnail images are then associated with the new information, increasing visual engagement.

[0046] Provide information and thumbnails (device)

[0047] The terminal displays new information and thumbnail images sent from the server to the user. Through the user interface, the user can access new information and view related thumbnail images, which attracts the user's interest and encourages access to the information.

[0048] Collecting user feedback (device and server)

[0049] The user evaluates the new information and thumbnails provided and enters feedback. The device collects the user's feedback data (e.g., interesting / uninteresting, useful / unhelpful, etc.) and sends it to the server. The server analyzes this feedback and updates the user's profile. Based on the user's feedback, the selection process for the next information offering is optimized.

[0050] Specific examples

[0051] If User A's areas of interest are "technology," "artificial intelligence," and "programming," the user's interest data is collected and key keywords are extracted using a natural language processing model. New keywords such as "biology," "art," and "psychology" are selected using "reverse" filtering, and new information is collected based on these keywords. Visual thumbnail images of this new information are generated by AI and provided to User A. User A views the new information and thumbnail images and provides feedback on whether or not they are interested. The server uses this feedback to provide more appropriate information next time.

[0052] The processing flow will be explained below.

[0053] Step 1:

[0054] The server collects information about users' interests, such as their access history, search history, and bookmark data, including the websites they visit, the keywords they search for, the articles they read, and the pages they bookmark. The collected data is stored in a database for each individual user.

[0055] Step 2:

[0056] The server uses natural language processing models to extract key keywords from the collected user data. For example, if a user frequently views articles related to "technology," "artificial intelligence," and "programming," these keywords will be extracted. The extracted keywords are then mapped based on their relevance, generating a keyword map for each user.

[0057] Step 3:

[0058] The server analyzes the generated keyword map and eliminates central keywords, including the user's primary interests: "technology," "artificial intelligence," and "programming." It then extracts peripheral keywords (e.g., "biology," "art," and "psychology"). This "reverse" filtering generates a new keyword list.

[0059] Step 4:

[0060] The server searches and collects relevant information from the Internet based on the new keywords extracted by "reverse" filtering. It selects articles, news, research papers, etc. related to the new keywords from the search results and collects appropriate information. The collected information is stored in a database.

[0061] Step 5:

[0062] The server uses artificial intelligence models to generate thumbnail images related to the new information collected. For example, for an article related to "biology," the AI ​​might generate a diagram of DNA or an image of a cell. The generated thumbnail images are then associated with the new information and stored.

[0063] Step 6:

[0064] The terminal displays new information and thumbnail images sent from the server to the user. Through the user interface, the user can access new information and view related thumbnail images, which attracts the user's interest and encourages access to the information.

[0065] Step 7:

[0066] Users rate the new information and thumbnails provided and provide feedback, including interest, relevance, and visual opinion about the information.

[0067] Step 8:

[0068] The device sends feedback data from the user to the server, which adds the collected feedback data to a profile for each user.

[0069] Step 9:

[0070] The server analyzes the feedback data and updates the user's profile, reflecting the user's new trends and interests and optimizing the selection process for the next offering. The analysis results are used for future "reverse" filtering and information gathering processes.

[0071] Example 1

[0072] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0073] Conventional information provision systems provide information tailored to users' interests, but it has been difficult to provide new information that expands users' interests. Furthermore, there is a lack of methods to attract users' attention by using visually appealing thumbnail images. Furthermore, there are also challenges in effectively collecting user feedback and optimizing the information provided based on that feedback.

[0074] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0075] In this invention, the server includes a means for collecting user information, a means for generating a keyword map based on the user information, a means for inverse filtering that excludes keywords at the center of the keyword map and extracts new keywords from peripheral keywords, a means for collecting new information based on the new keywords, a means for generating images related to the new information, a means for providing the new information and the images to the user, and a means for collecting user ratings. This allows the server to efficiently provide new information that is not of interest to the user and generate visually appealing thumbnail images related to that information, thereby attracting the user's interest. Furthermore, by collecting and analyzing user feedback, the server can provide optimized information the next time it provides information.

[0076] "User information" refers to data such as a user's access history, search history, articles viewed, and bookmarked pages.

[0077] A "keyword map" is a graph structure that arranges key keywords extracted based on user interests in a relevance manner.

[0078] "Inverse filtering" is a technique that removes the main keywords in the center of a keyword map and extracts new, relevant keywords that are located on the periphery.

[0079] "New information" refers to content such as articles, news, and research papers that are collected based on new keywords obtained through reverse filtering.

[0080] "Image" refers to a visual thumbnail image generated in association with new information.

[0081] A "generative artificial intelligence model" is an artificial intelligence model that generates new images or text based on input data provided by humans.

[0082] "Evaluation" refers to feedback data that reflects whether the user is interested in the information or images provided, whether it is useful, etc.

[0083] The present invention provides an information provision system based on a user's interest, which is implemented through the following steps.

[0084] The server first collects user information, including the history of websites the user has visited, keywords they have searched for, articles they have viewed, and pages they have bookmarked. This data is collected using, for example, Google Analytics or a dedicated tracking code. The collected data is associated with each user and stored in a database (for example, MySQL or PostgreSQL).

[0085] The server then analyzes the collected user information and extracts key keywords using a natural language processing model (e.g., Hugging Face's BERT model). This analysis may also utilize Term Frequency-Inverse Document Frequency (TF-IDF). For example, if a user frequently views articles related to "programming," "machine learning," and "data science," these keywords will be extracted as their key interests.

[0086] These extracted key keywords are then mapped based on their relevance to generate a keyword map, which has a graph structure with the key keyword at the center and related keywords around it.

[0087] Additionally, the server performs reverse filtering, which removes key keywords (e.g., "programming," "machine learning," etc.) from the center of the keyword map and extracts peripheral keywords (e.g., "psychology," "biology," etc.). This reverse filtering allows for the extraction of new, relevant keywords that are outside the user's usual interests.

[0088] The server collects new information from the Internet based on the new keywords obtained through reverse filtering. Specifically, it uses the Google News API, PubMed API, etc. to collect related articles, news, research papers, etc. This collected information is stored in a database and managed as candidates for information to be provided to users.

[0089] As new information is collected, the server uses generative AI models (such as DALL-E 2 or StyleGAN) to generate thumbnail images related to that information. For example, for articles related to "biology," it might generate diagrams of DNA or images of cells. The generated thumbnail images are then associated with the new information and presented to the user in a visually appealing format.

[0090] The device displays new information and thumbnail images sent from the server to the user. The front-end user interface uses React or Vue.js, and the user is visually captivated by the information and thumbnail images.

[0091] Finally, the user rates the provided information and thumbnail images and enters their feedback. The device collects this feedback data and sends it to the server. For example, it can be in the form of a simple rating such as "useful" or "not interesting." The server analyzes the feedback and updates the user's profile, enabling it to provide more optimized information the next time information is provided.

[0092] Specific examples

[0093] For example, if User A is interested in "technology," "artificial intelligence," and "programming," the server collects related data and uses the BERT model to extract key keywords. It then performs reverse filtering to identify new keywords, such as "biology," "art," and "psychology." Based on these keywords, related information is collected using the Google News API and PubMed API. "Thumbnail images related to biology" generated by DALL-E 2 are provided to User A as visual aids. User A provides feedback on the information provided, and the server uses that feedback to optimize the information provided.

[0094] Prompt Sentence Examples

[0095] "Extract key keywords based on user data, then use reverse filtering to gather new information related to the resulting keywords. Then generate thumbnail images related to the new information."

[0096] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0097] Step 1: Collecting User Data (Server)

[0098] The server collects data such as the history of websites visited by the user, keywords searched, articles viewed, and pages bookmarked. This data is collected using Google Analytics or a dedicated tracking code. The server receives each user's action data and uses this data to create a detailed history database for each user. The collected data is associated with each user and stored in a database (e.g., MySQL, PostgreSQL).

[0099] Step 2: Generate Keyword Mapping (Server)

[0100] The server receives the collected user information as input and uses a natural language processing model (e.g., Hugging Face's BERT model) to extract key keywords. This analysis also uses Term Frequency-Inverse Document Frequency (TF-IDF) scoring of the text data. The output is a list of keywords that indicate the user's main interests. For example, "programming," "machine learning," and "data science" may be extracted.

[0101] Step 3: Generate a Keyword Map (Server)

[0102] The server receives the extracted keywords as input and generates a keyword map based on their relevance. This keyword map has a graph structure with the main keyword at the center and related keywords around it. The output is a graph-structured keyword map.

[0103] Step 4: Performing Reverse Filtering (Server)

[0104] The server receives the generated keyword map as input, eliminates the main keywords located in the center, and extracts peripheral keywords. This process is called inverse filtering. The output is a list of new keywords that are outside the user's usual interests. For example, "psychology," "biology," and "philosophy" are extracted.

[0105] Step 5: Selecting new information (server)

[0106] The server receives the new keywords obtained through reverse filtering as input and collects new information from the Internet based on them, using APIs such as Google News and PubMed. The output is a dataset of articles, news articles, and research papers related to the new keywords. This information is then stored in a database.

[0107] Step 6: Thumbnail generation (server)

[0108] The server takes the collected new information as input and uses a generative AI model (e.g., DALL-E 2 or StyleGAN) to generate relevant thumbnail images. The output is a visually appealing thumbnail image that corresponds to the new information. For example, an article on "biology" might generate a diagram of DNA or an image of a cell.

[0109] Step 7: Provide information and thumbnail (device)

[0110] The terminal receives new information and thumbnail images provided by the server as input and displays them to the user. The front-end user interface uses React or Vue.js. The output is a visually appealing display of information provided to the user. The user can browse these displays.

[0111] Step 8: Collecting User Feedback (Devices and Servers)

[0112] Users rate the provided information and thumbnail images and enter their feedback. The input includes ratings such as "useful" or "not interesting." The device collects this feedback data and sends it to the server. The server outputs the analyzed feedback data, which is used to update the user's profile. This optimizes the next information provided.

[0113] (Application example 1)

[0114] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0115] Conventional information provision systems only provide information based on users' existing interests, making it difficult to stimulate new interests. Furthermore, the methods for providing information based on user data are limited, preventing effective provision of information that will interest users in local stores. Furthermore, there is a lack of information provision based on the user's location information, making it insufficient to improve the user experience in physical stores.

[0116] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0117] In this invention, the server includes means for collecting user interest data, means for generating a keyword map based on the interest data, means for "reverse" filtering to exclude keywords located at the center of the keyword map and extract new keywords from peripheral keywords, means for collecting new information based on the new keywords, means for generating thumbnail images related to the new information, means for providing the new information and the thumbnail images to the user, means for collecting feedback from the user, means for recognizing the user within the terminal, and means for the terminal to provide information based on the user's location, thereby enabling the server to elicit new interests beyond the user's existing interests and improve the in-store experience.

[0118] "User Interest Data" is data about a user's interests, such as the user's website visit history, search history, and bookmarks.

[0119] A "keyword map" is a graph structure that associates key keywords based on user interest data, with the main interest at the center and less relevant keywords at the periphery.

[0120] "Inverse filtering" is a technique that eliminates the main keywords located at the center of the keyword map and extracts new keywords from the remaining peripheral keywords.

[0121] A "thumbnail image" is a small visual image generated in association with new information collected and used to attract user attention.

[0122] "Device" refers to an electronic device used by a user to receive information, including smartphones, tablets, and personal computers.

[0123] "Means of recognizing users" refers to technology that enables devices to detect and identify the user's presence and location within a store, and uses beacons, Wi-Fi, etc.

[0124] User Data Collection

[0125] The server collects user interest data, including user website visit history, search history, and bookmark data, which is stored in a database for each user and used for later analysis.

[0126] Generate Keyword Mapping

[0127] The server uses a natural language processing model to extract key keywords from the collected user data. For example, if a user frequently views articles related to "technology," "artificial intelligence," and "programming," these keywords are extracted. The extracted keywords are then mapped based on their relevance to generate a keyword map for each user. This keyword map has a graph structure that displays keywords near the center as the user's main interests and places less relevant keywords on the periphery.

[0128] Implementing "reverse" filtering

[0129] The server analyzes the generated keyword map and eliminates central keywords. For example, if the user's primary interests are "technology," "artificial intelligence," and "programming," these keywords are excluded. It then extracts more peripheral keywords (e.g., "biology," "art," and "psychology"). This "reverse" filtering extracts keywords that the user would not normally be interested in, but which may provide relevant new information.

[0130] Selection of new information

[0131] The server searches and collects relevant information from the Internet based on the new keywords extracted through "reverse" filtering, such as articles, news, and research papers related to the new keywords. The collected information is stored in a database and later provided to users.

[0132] Thumbnail Generation

[0133] The server uses artificial intelligence models to generate thumbnail images related to new information collected. For example, for articles related to "biology," the AI ​​model might generate diagrams of DNA or images of cells. The generated thumbnail images are then associated with the new information, increasing visual engagement.

[0134] Providing information and thumbnails

[0135] The device displays new information and thumbnail images sent from the server to the user. Through the user interface, the user can access new information and view related thumbnail images, which can attract the user's interest and encourage access to the information. It also recognizes the user within the device and provides more relevant information based on the user's location.

[0136] Collecting user feedback

[0137] The user evaluates the new information and thumbnails provided and enters feedback. The device collects the user's feedback data (e.g., interesting / uninteresting, useful / unhelpful, etc.) and sends it to the server. The server analyzes this feedback and updates the user's profile. Based on the user's feedback, the selection process for the next information offering is optimized.

[0138] Specific examples

[0139] If User A's areas of interest are "technology," "artificial intelligence," and "programming," the user's interest data is collected and key keywords are extracted using a natural language processing model. New keywords such as "biology," "art," and "psychology" are selected using "reverse" filtering, and new information is collected based on these keywords. Visual thumbnail images of this new information are generated by AI and provided to User A. User A views the new information and thumbnail images and provides feedback on whether or not they are interested. The server uses this feedback to provide more appropriate information next time.

[0140] Prompt Sentence Examples

[0141] For users with an interest in "Technology," suggest recent research articles in "Biology" or "Psychology" that are relevant but may spark new interest. Also generate visual thumbnails associated with each article.

[0142] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0143] Step 1:

[0144] The server collects user website visit history, search history, and bookmark data. This data is stored in a database for each user and converted into a format for analysis. The input is the user's online activity history, and the output is formatted user data.

[0145] Step 2:

[0146] The server uses a natural language processing model to extract key keywords from the formatted user data. The input is the user data organized in the previous step, and the output is a list of key keywords. Specifically, the data is input into the natural language processing algorithm to extract frequently occurring keywords.

[0147] Step 3:

[0148] The server visualizes the extracted keywords as a keyword map based on their relevance. The input is a list of key keywords, and the output is a graph-structured map of the keywords. Specifically, it calculates the relevance between keywords and generates the map using a graph library.

[0149] Step 4:

[0150] The server analyzes the generated keyword map, eliminates the main keywords located in the center, and extracts new keywords from the periphery using a "reverse" filtering method. The input is the keyword map, and the output is a list of new keywords. Specifically, the server reduces the weight of the central node and selects new keywords from the periphery.

[0151] Step 5:

[0152] The server searches and collects relevant information from the Internet based on the new keywords extracted by "reverse" filtering. It takes a list of new keywords as input and a list of new information as output. Specifically, it uses a search API to query related information on the web and organizes the results.

[0153] Step 6:

[0154] The server uses an artificial intelligence model to generate thumbnail images related to the new information collected. The input is a list of new information, and the output is a list of thumbnail images. Specifically, the content of the information is input into the AI ​​model, which generates related visual thumbnails.

[0155] Step 7:

[0156] The terminal displays the new information and thumbnail image sent from the server to the user. The input is the new information and thumbnail image, and the output is the information and image displayed on the user's display. The specific operation is to arrange the information and image in the terminal's user interface.

[0157] Step 8:

[0158] Users rate the provided new information and thumbnails and enter their feedback. The input is the user's rating data, and the output is the feedback data. The specific operation is for users to enter their interest and rating through the interface.

[0159] Step 9:

[0160] The terminal collects feedback data from the user and sends it to the server. The input is the user's feedback data, and the output is the feedback data sent to the server. Specifically, the terminal uses the sending function to upload the feedback data to the server.

[0161] Step 10:

[0162] The server analyzes the feedback data from the user and updates the user profile. The input is the feedback data sent from the device, and the output is the updated user profile. Specifically, the server reevaluates the user's interest trends based on the feedback data and optimizes the next information provision.

[0163] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0164] The following describes in detail the mode for carrying out the present invention. In the present invention, a system is constructed that provides new information that the user would not normally come across based on the user's interest data, and further recognizes the user's emotions to optimize the feedback.

[0165] User data collection (server)

[0166] The server collects user access history, search history, bookmark data, and other interest data, including the websites the user has visited, the keywords they have searched for, the articles they have viewed, the pages they have bookmarked, etc. The collected data is stored in a database for each individual user.

[0167] Generate Keyword Mapping (Server)

[0168] The server uses natural language processing models to extract key keywords from the collected user data. For example, if a user frequently views articles about "technology," "artificial intelligence," and "programming," these keywords will be extracted. The extracted keywords are then mapped based on their relevance to generate a keyword map for each user.

[0169] "Reverse" filtering implementation (server)

[0170] The server analyzes the generated keyword map and eliminates central keywords, including the user's primary interests: "technology," "artificial intelligence," and "programming." It then extracts peripheral keywords (e.g., "biology," "art," and "psychology"). This "reverse" filtering generates a new keyword list.

[0171] Selection of new information (server)

[0172] The server searches and collects relevant information from the Internet based on the new keywords extracted by "reverse" filtering. It selects articles, news, research papers, etc. related to the new keywords from the search results and collects appropriate information. The collected information is stored in a database.

[0173] Thumbnail generation (server)

[0174] The server uses artificial intelligence models to generate thumbnail images related to the new information collected. For example, for an article related to "biology," the AI ​​might generate a diagram of DNA or an image of a cell. The generated thumbnail images are then associated with the new information and stored.

[0175] Incorporating emotion recognition (server)

[0176] The server uses an emotion engine to recognize emotions based on user input data (e.g., feedback and voice input). The emotion engine inputs the user's ratings and feedback into an emotion analysis algorithm to analyze the user's emotional state (e.g., happy, interesting, bored, etc.). This emotion data is added to the user's profile.

[0177] Provide information and thumbnails (device)

[0178] The terminal displays new information and thumbnail images sent from the server to the user. Through the user interface, the user can access new information and view related thumbnail images, which attracts the user's interest and encourages access to the information.

[0179] Collecting user feedback (device and server)

[0180] Users rate the new information and thumbnails provided and provide feedback, including interest, relevance, visual opinion, and emotional state (e.g., enjoyable, interesting, boring, etc.).

[0181] The device sends feedback data from the user to the server, which adds the collected feedback data to a profile for each user.

[0182] The server analyzes the feedback data and updates the user's profile, reflecting the user's new trends and interests and optimizing the selection process for the next offering. The analysis results are used for future "reverse" filtering and information gathering processes.

[0183] Specific examples

[0184] If User A's areas of interest are "technology," "artificial intelligence," and "programming," the user's interest data is collected and key keywords are extracted using a natural language processing model. New keywords such as "biology," "art," and "psychology" are selected using "reverse" filtering, and new information is collected based on these keywords. Visual thumbnail images of this new information are generated by AI and provided to User A.

[0185] User A browses the new information and thumbnail images and provides feedback on whether or not they are interested. The emotion engine then analyzes User A's emotions based on their feedback and voice input to understand their emotional state. The server uses this feedback and emotion data to update User A's profile to provide more appropriate information next time.

[0186] This allows users to be exposed to new perspectives and knowledge, and also provides them with the most appropriate information based on their emotions.

[0187] The processing flow will be explained below.

[0188] Step 1:

[0189] The server collects user access history, search history, bookmark data, and other interest data, including the websites the user has visited, the keywords they have searched for, the articles they have viewed, the pages they have bookmarked, etc. The collected data is stored in a database for each individual user.

[0190] Step 2:

[0191] The server uses natural language processing models to extract key keywords from the collected user data. For example, if a user frequently views articles about "technology," "artificial intelligence," and "programming," these keywords will be extracted. The extracted keywords are then mapped based on their relevance to generate a keyword map for each user.

[0192] Step 3:

[0193] The server analyzes the generated keyword map and eliminates central keywords, including the user's primary interests: "technology," "artificial intelligence," and "programming." It then extracts peripheral keywords (e.g., "biology," "art," and "psychology"). This "reverse" filtering generates a new keyword list.

[0194] Step 4:

[0195] The server searches and collects relevant information from the Internet based on the new keywords extracted by "reverse" filtering. It searches for articles, news, research papers, etc. related to the new keywords to collect appropriate information. The collected information is stored in a database.

[0196] Step 5:

[0197] The server uses artificial intelligence models to generate thumbnail images related to the new information collected. For example, for an article related to "biology," the AI ​​might generate a diagram of DNA or an image of a cell. The generated thumbnail images are then associated with the new information and stored.

[0198] Step 6:

[0199] The server uses an emotion engine to recognize emotions based on user input data (e.g., feedback and voice input). The emotion engine inputs the user's ratings and feedback into an emotion analysis algorithm to analyze the user's emotional state (e.g., happy, interesting, bored, etc.). This emotion data is added to the user's profile.

[0200] Step 7:

[0201] The terminal displays new information and thumbnail images sent from the server to the user. Through the user interface, the user can access new information and view related thumbnail images, which attracts the user's interest and encourages access to the information.

[0202] Step 8:

[0203] Users rate the new information and thumbnails provided and provide feedback, including interest and relevance of the information, visual opinion, and emotional state (e.g., enjoyable, interesting, boring, etc.).

[0204] Step 9:

[0205] The device sends feedback data from the user to the server, which adds the collected feedback data to a profile for each user.

[0206] Step 10:

[0207] The server analyzes the feedback data and updates the user's profile, reflecting the user's new trends and interests and optimizing the selection process for the next offering. The analysis results are used for future "reverse" filtering and information gathering processes.

[0208] Specific examples

[0209] As a specific example, we will explain the case where User A's areas of interest are "technology," "artificial intelligence," and "programming."

[0210] Step 1:

[0211] The server collects user A's past access history (e.g., technews.com), search history (e.g., "latest AI algorithm"), bookmarks (e.g., programmingblog.com), etc. and stores them in a database.

[0212] Step 2:

[0213] The server uses a natural language processing model to extract key keywords such as "technology," "artificial intelligence," and "programming" from the collected data, and generates a keyword map based on this.

[0214] Step 3:

[0215] The server removes the central keywords "technology," "artificial intelligence," and "programming" from the generated keyword map and extracts peripheral keywords such as "biology," "art," and "psychology."

[0216] Step 4:

[0217] The server searches the Internet for relevant articles and news based on the extracted new keywords, collects the necessary information, and stores it in a database.

[0218] Step 5:

[0219] The server uses artificial intelligence models to generate thumbnail images related to the new information collected. For example, for an article related to "biology," the AI ​​model generates a diagram of DNA or an image of a cell, which is then stored along with the new information.

[0220] Step 6:

[0221] The server analyzes emotions using an emotion engine based on user feedback and voice input. User A's emotions when viewing information (fun, interesting, bored, etc.) are input into the emotion analysis algorithm, and the analysis results are added to the user profile.

[0222] Step 7:

[0223] The terminal displays the new information and thumbnail image sent from the server and provides them to User A. User A views the new information and thumbnail image.

[0224] Step 8:

[0225] User A evaluates the new information and thumbnails provided and enters feedback (interesting / uninteresting, useful / unhelpful, etc.).

[0226] Step 9:

[0227] The device sends feedback data from user A to the server, which adds this feedback data to the profile for each user A.

[0228] Step 10:

[0229] The server analyzes the feedback data and updates the profile of User A. Based on the feedback and sentiment data, the server optimizes the selection process for the next information to be provided and utilizes it in the selection of new keywords and information gathering process.

[0230] As a result, User A is exposed to new perspectives and knowledge, and is provided with information that is optimally tailored to their emotions.

[0231] Example 2

[0232] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0233] Modern information systems mainly present information based on the user's existing interests, limiting opportunities for users to discover new interests and knowledge. Furthermore, information is often provided in a uniform manner without considering the user's emotional state. This makes it difficult to capture the user's attention, resulting in low information receptivity.

[0234] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting user interest data, means for generating a keyword map based on the interest data, a "reverse" filtering means for excluding keywords located at the center of the keyword map and extracting new keywords from peripheral keywords, means for searching and collecting new information based on the new keywords, means for generating visual content related to the new information, means for providing the new information and the visual content to the user, means for collecting feedback and emotional state from the user, and means for updating a user profile based on the feedback and emotional data. This allows the user to discover new information that they have not encountered before and provides optimal information according to their emotional state.

[0235] "User interest data" is a general term for behavioral history that indicates a user's interests and concerns, such as websites visited by the user, keywords searched for, articles viewed, and pages bookmarked.

[0236] A "keyword map" is a visual or digital mapping of key keywords extracted from user interest data and their relationships.

[0237] "Inverse filtering" is the process of eliminating the main keywords located in the center of the keyword map and extracting new keywords located around them.

[0238] "New information" is related information collected from the Internet and databases based on new keywords extracted by reverse filtering.

[0239] "Visual content" refers to images and visual materials related to new information generated using artificial intelligence models, etc.

[0240] "Feedback" refers to the evaluations and impressions users make of new information and visual content.

[0241] "Emotional state" refers to the user's emotional response or state, as analyzed from user feedback, voice input, etc.

[0242] The embodiments of the present invention will be described in detail. In this invention, new information that the user would not normally come across is provided based on the user's interest data. Also, a system is constructed that recognizes the user's emotions and optimizes feedback.

[0243] The server first collects interest data, such as the websites visited by the user, keywords searched, articles viewed, pages bookmarked, etc. This data is collected using technologies such as cookies and local storage data from the browser, and the collected data is stored in a database for each individual user.

[0244] The server then uses the collected user data to extract key keywords using a natural language processing (NLP) model, such as BERT or GPT, and maps the extracted keywords based on their relevance to generate a keyword map for each user.

[0245] In the process of analyzing the generated keyword map, the server will eliminate the main keywords located in the center and extract new keywords located in the periphery. This "reverse filtering" process will reveal keywords in areas that users are not usually interested in.

[0246] The server then searches and collects relevant information from the Internet based on the new keywords extracted by reverse filtering, using Google Search API and News API to retrieve articles and research papers related to the keywords and store them in a database.

[0247] The server then uses artificial intelligence models such as DALL-E and MidJourney to generate visual content related to the collected information: for example, articles related to "biology" will generate models of DNA and microscopic images of cells.

[0248] The terminal displays new information and visual content sent from the server to the user, and through a web browser or application interface, the user can access the new information and view the associated visual content.

[0249] Users provide feedback on the new information and visual content provided, including ratings of interest and relevance of the information. The device sends the feedback data to a server, where it is added to each user's profile.

[0250] The server uses an emotion engine to analyze emotions from user feedback and voice input. The feedback text is fed into the emotion analysis algorithm to determine the user's emotional state. This emotion data is also added to the user profile.

[0251] Finally, the server analyzes the feedback and sentiment data and updates the user's profile to reflect new trends and interests and optimize the next round of information delivery.

[0252] Specific examples

[0253] For example, if User A's areas of interest are "technology," "artificial intelligence," and "programming," these interest data are collected and key keywords are extracted using the NLP model. New keywords such as "biology," "art," and "psychology" are selected through reverse filtering, and new information is collected based on these keywords. Visual thumbnail images of this new information are generated by the AI ​​model and provided to User A.

[0254] Prompt Sentence Examples

[0255] "Suggest relevant articles and visual content based on the user's emerging areas of interest."

[0256] "Generate thumbnail images of new information based on reverse-filtered keywords."

[0257] "Use user feedback and sentiment data to suggest ways to optimize your next offering."

[0258] This allows users to access new perspectives and knowledge, and provides optimal information tailored to their emotions.

[0259] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0260] Step 1: Collecting User Data (Server)

[0261] The server collects behavioral data from the user's browser, such as the websites visited, keywords searched, articles viewed, and pages bookmarked, and uses the browser's cookies and local storage to obtain the data and store it in a database for each individual user.

[0262] Input: User behavior history data (website URL, search keywords, viewed articles, bookmark data)

[0263] Output: Database entries per user

[0264] Step 2: Generate Keyword Mapping (Server)

[0265] The server extracts key keywords from the collected user data using natural language processing (NLP) models such as BERT and GPT, which analyze and extract important keywords from text data, and then generates a keyword map based on the relevance of the keywords.

[0266] Input: User behavior history data

[0267] Output: Keyword map per user

[0268] Step 3: Performing "reverse" filtering (server)

[0269] The server analyzes the generated keyword map, eliminates the main keywords located in the center, and extracts new keywords located around these eliminated keywords, thereby obtaining keywords in areas that users are not usually interested in.

[0270] Input: User-specific keyword map

[0271] Output: New keyword list

[0272] Step 4: Selecting new information (server)

[0273] The server searches and collects relevant information from the Internet based on the new keywords extracted by reverse filtering, using Google Search API and News API to retrieve articles, research papers, etc. related to the keywords.

[0274] Input: New keyword list

[0275] Output: New information collected (articles, news, research papers, etc.)

[0276] Step 5: Generate visual content (server)

[0277] The server uses artificial intelligence models to generate visual content related to the collected information, for example, image generation AI such as DALL-E or MidJourney to generate visual content related to keywords.

[0278] Input: New information (articles, news, research papers, etc.)

[0279] Output: Relevant visual content (e.g. thumbnail image)

[0280] Step 6: Providing information and visual content (device)

[0281] The device displays new information and visual content sent from the server to the user. Through a web browser or application interface, the user can access new information and view related visual content (such as thumbnail images).

[0282] Input: New information and visual content sent from the server

[0283] Output: Information and visual content displayed in the user interface

[0284] Step 7: Collecting User Feedback (Devices and Servers)

[0285] The user inputs ratings and feedback for the new information and visual content provided, including interest in the information, relevance, visual opinion, and emotional state. The device then transmits the feedback data to the server.

[0286] Input: User feedback data

[0287] Output: Feedback data sent to the server

[0288] Step 8: Integrating Emotion Recognition (Server)

[0289] The server uses an emotion engine to analyze emotions from the user's feedback and voice input, and uses an emotion analysis algorithm to determine the user's emotional state from the feedback text.

[0290] Input: User feedback and voice input data

[0291] Output: Parsed user sentiment data

[0292] Step 9: Update Profile (Server)

[0293] The server analyzes the feedback and sentiment data and updates the user profile, so that new trends and interests are reflected in the profile and the next information delivery is optimized.

[0294] Input: User feedback data, sentiment data

[0295] Output: Updated user profile

[0296] The above are the specific processing steps of this system.

[0297] (Application example 2)

[0298] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0299] Conventional information delivery systems tend to only provide information that users are already interested in, limiting opportunities to encounter new perspectives and knowledge. Furthermore, it is difficult to provide optimal information based on the user's interests and emotions, resulting in a lack of improvement in the user experience. Furthermore, the lack of feedback optimization incorporating emotion recognition can reduce the accuracy and relevance of information provided.

[0300] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0301] In this invention, the server includes means for collecting user interest data, means for generating a keyword map based on the interest data, a "reverse" filtering means for excluding keywords located at the center of the keyword map and extracting new keywords from peripheral keywords, means for collecting new information based on the new keywords, means for generating thumbnail images related to the new information, means for providing the new information and the thumbnail images to the user, means for collecting feedback from the user, emotion recognition means for analyzing the user's emotions based on the feedback, and means for optimizing the next information to be provided based on the analyzed emotions. This allows the user to be exposed to new perspectives and knowledge and to be provided with optimal information according to their emotions.

[0302] "User interest data" refers to data that reflects a user's interests, such as user access history, search history, and bookmark data.

[0303] A "keyword map" is a map of key keyword relationships generated using natural language processing models based on user interest data.

[0304] "Inverse filtering" is a technique that eliminates the main keywords located at the center of the keyword map and extracts new keywords from the surrounding areas.

[0305] "New information" is information related to new perspectives and knowledge that is collected based on new keywords extracted by reverse filtering.

[0306] A "thumbnail image" is a small visual image associated with the new information collected.

[0307] "User feedback" refers to users' ratings and impressions of the provided information and thumbnail images.

[0308] An "emotion recognition means" is an engine or algorithm that analyzes emotions based on user feedback and understands their emotional state.

[0309] "Information optimization means" refers to the means of adjusting and optimizing the content of the next information provided based on the analyzed user sentiment.

[0310] A "natural language processing model" is an algorithm or machine learning model for extracting and analyzing meaning and keywords from text data.

[0311] An "artificial intelligence model" is an algorithm or system that learns from large amounts of data and automates specific tasks.

[0312] The present invention provides a technology for constructing a system that provides new information that a user would not normally come across, and further recognizes the user's emotions and optimizes feedback. Specific embodiments of the present invention will be described below.

[0313] System configuration

[0314] This system consists of a server that collects user interest data, a server that generates a keyword map based on the interest data, a server that performs "reverse" filtering to eliminate major keywords from the generated keyword map and extract new keywords, a server that collects new information and generates thumbnail images, a terminal that provides information and thumbnail images to users, a server that collects user feedback and performs emotion recognition, and a server that optimizes the information to be provided next time.

[0315] Collection and Creation

[0316] The server collects interest data such as user access history, search history, and bookmark data, including the websites the user has visited, the keywords they have searched for, the articles they have viewed, the pages they have bookmarked, etc. The collected data is stored in a database for each user.

[0317] The server then uses natural language processing models to extract key keywords from the collected user data. For example, if a user frequently views articles about "technology," "artificial intelligence," and "programming," these keywords will be extracted. The extracted keywords are then mapped based on their relevance to generate a keyword map for each user.

[0318] Reverse filtering and new information gathering

[0319] The generated keyword map is analyzed and the main keywords located in the center are eliminated. This elimination includes the user's main interests, such as "technology," "artificial intelligence," and "programming." New keywords located in the periphery (e.g., "biology," "art," and "psychology") are then extracted. This "reverse" filtering generates a new keyword list.

[0320] The server searches and collects relevant information from the Internet or specific information sources based on the new keywords extracted by "reverse" filtering. It selects articles, news, research papers, etc. related to the new keywords from the search results and collects appropriate information. The collected information is stored in a database.

[0321] Thumbnail generation and serving

[0322] The server uses artificial intelligence models to generate thumbnail images related to the new information collected. For example, for an article related to "biology," the generated thumbnail image might be a diagram of DNA or an image of a cell. The generated thumbnail image is associated with the new information and stored.

[0323] The terminal displays the new information and thumbnail images sent from the server to the user. Through the user interface, the user can access the new information and view the related thumbnail images.

[0324] Feedback collection and emotion recognition

[0325] The user evaluates the provided new information and thumbnails and enters feedback, including interest, relevance, visual evaluation, and emotional state (e.g., fun, interesting, boring, etc.). The device then transmits this feedback data to the server.

[0326] The server uses an emotion engine to recognize emotions based on user input data. The emotion engine inputs user ratings and feedback into an emotion analysis algorithm to analyze the user's emotional state. This emotion data is added to the user's profile.

[0327] Optimizing next advertisement

[0328] The server analyzes the feedback data and sentiment data and updates the user's profile to reflect the user's new trends and interests and optimize the next information offering.

[0329] Examples of specific examples and prompts

[0330] As a concrete example, assume that User A is interested in "technology," "artificial intelligence," and "programming." This user's interest data is collected, and key keywords are extracted using a natural language processing model. New keywords such as "biology," "art," and "psychology" are selected using a "reverse" filtering method, and new information is collected based on these keywords. Visual thumbnail images of this new information are generated by AI and provided to User A. User A views the new information and thumbnail images and provides feedback on whether or not they are interested. Furthermore, an emotion engine analyzes emotions based on User A's feedback and voice input, and grasps the user's emotional state. The server uses this feedback and emotion data to update User A's profile to provide more appropriate information next time.

[0331] An example prompt might be, "The user is interested in technology, artificial intelligence, and programming. After reverse filtering, the keywords are biology, art, and psychology. Based on these keywords, please find and provide relevant, up-to-date articles and information."

[0332] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0333] Step 1:

[0334] Server collection of user interest data

[0335] The server collects interest data such as user access history, search history, and bookmark data, including the websites the user visits, keywords searched, articles viewed, and pages bookmarked. The collected data is stored in a database as individual user profiles.

[0336] Input: Access history, search history, bookmark data

[0337] Output: Save interest data in user profile

[0338] Step 2:

[0339] Server-generated keyword mappings

[0340] The server extracts key keywords from the collected interest data using natural language processing models (e.g., TF-IDF-based vectorization and clustering). For example, if a user frequently views articles related to "technology," "artificial intelligence," and "programming," these keywords will be extracted. The extracted keywords are then mapped to a keyword map based on their relevance.

[0341] Input: Interest data (user profile)

[0342] Output: Keyword map (including a list of primary keywords)

[0343] Step 3:

[0344] Server performs "reverse" filtering

[0345] The server analyzes the generated keyword map and eliminates central keywords, including the user's main interests: "technology," "artificial intelligence," and "programming." It then extracts peripheral new keywords (e.g., "biology," "art," and "psychology").

[0346] Input: Keyword Map

[0347] Output: New "reverse" filtered keyword list

[0348] Step 4:

[0349] Server selection and collection of new information

[0350] The server searches and collects relevant information from the Internet and specific information sources based on the new keywords extracted through "reverse" filtering, and stores the information, such as articles, news, and research papers, related to the new keywords in a database.

[0351] Input: New "reverse" filtered keyword list

[0352] Output: New information (articles, news, research papers, etc.)

[0353] Step 5:

[0354] Server-generated thumbnail images

[0355] The server uses artificial intelligence models (e.g., image generation AI) to generate thumbnail images related to the new information collected. For example, for an article related to "biology," the generated thumbnail images might be diagrams of DNA or images of cells.

[0356] Input: New information

[0357] Output: Thumbnail image

[0358] Step 6:

[0359] Device provides information and thumbnail images

[0360] The terminal displays the new information and thumbnail images sent from the server to the user, and through the user interface, the user can access the new information and visually check the related thumbnail images.

[0361] Input: New information, thumbnail image

[0362] Output: User interface display (new information and thumbnail image)

[0363] Step 7:

[0364] Collecting and entering user feedback

[0365] Users input feedback about the new information and thumbnail images provided, such as interest, relevance, rating, visual opinion, and emotional state, which is then sent to the server via the device.

[0366] Input: User feedback on new information and thumbnail images

[0367] Output: Feedback data (ratings, emotional state, etc.)

[0368] Step 8:

[0369] Emotion recognition and analysis by the server

[0370] The server uses an emotion recognition engine to analyze the user's emotional state based on the feedback data received from the user, and the analysis results are added to the user's profile.

[0371] Input: Feedback data

[0372] Output: Emotion data (fun, interesting, bored, etc.)

[0373] Step 9:

[0374] Server optimization of next information provision

[0375] The server analyzes the feedback and sentiment data and updates the user's profile to reflect new trends and interests and optimize the content of the next information provided.

[0376] Input: Feedback data, emotion data

[0377] Output: Updated user profile, optimization plan for next information provided

[0378] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0379] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0380] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0381] [Second embodiment]

[0382] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0383] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0384] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0385] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0386] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0387] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0388] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0389] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0390] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0391] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0392] In the smart glasses 214, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0393] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0394] The embodiments for carrying out the present invention will be described in detail below.

[0395] User data collection (server)

[0396] The server collects users' access history, search history, bookmark data, etc. This includes the websites they have visited, the keywords they have searched for, the articles they have viewed, the pages they have bookmarked, etc. The collected data is stored in a database for each individual user and used for later analysis.

[0397] Generate Keyword Mapping (Server)

[0398] The server uses a natural language processing model to extract key keywords from the collected user data. For example, if a user frequently views articles related to "technology," "artificial intelligence," and "programming," these keywords are extracted. The extracted keywords are then mapped based on their relevance to generate a keyword map for each user. This keyword map has a graph structure that displays keywords near the center as the user's main interests and places less relevant keywords on the periphery.

[0399] "Reverse" filtering implementation (server)

[0400] The server analyzes the generated keyword map and eliminates central keywords. For example, if the user's primary interests are "technology," "artificial intelligence," and "programming," these keywords are excluded. It then extracts more peripheral keywords (e.g., "biology," "art," and "psychology"). This "reverse" filtering extracts keywords that the user would not normally be interested in, but which may provide relevant new information.

[0401] Selection of new information (server)

[0402] The server searches and collects relevant information from the Internet based on the new keywords extracted through "reverse" filtering, such as articles, news, and research papers related to the new keywords. The collected information is stored in a database and later provided to users.

[0403] Thumbnail generation (server)

[0404] The server uses artificial intelligence models to generate thumbnail images related to new information collected. For example, for articles related to "biology," the AI ​​model might generate diagrams of DNA or images of cells. The generated thumbnail images are then associated with the new information, increasing visual engagement.

[0405] Provide information and thumbnails (device)

[0406] The terminal displays new information and thumbnail images sent from the server to the user. Through the user interface, the user can access new information and view related thumbnail images, which attracts the user's interest and encourages access to the information.

[0407] Collecting user feedback (device and server)

[0408] The user evaluates the new information and thumbnails provided and enters feedback. The device collects the user's feedback data (e.g., interesting / uninteresting, useful / unhelpful, etc.) and sends it to the server. The server analyzes this feedback and updates the user's profile. Based on the user's feedback, the selection process for the next information offering is optimized.

[0409] Specific examples

[0410] If User A's areas of interest are "technology," "artificial intelligence," and "programming," the user's interest data is collected and key keywords are extracted using a natural language processing model. New keywords such as "biology," "art," and "psychology" are selected using "reverse" filtering, and new information is collected based on these keywords. Visual thumbnail images of this new information are generated by AI and provided to User A. User A views the new information and thumbnail images and provides feedback on whether or not they are interested. The server uses this feedback to provide more appropriate information next time.

[0411] The processing flow will be explained below.

[0412] Step 1:

[0413] The server collects information about users' interests, such as their access history, search history, and bookmark data, including the websites they visit, the keywords they search for, the articles they read, and the pages they bookmark. The collected data is stored in a database for each individual user.

[0414] Step 2:

[0415] The server uses natural language processing models to extract key keywords from the collected user data. For example, if a user frequently views articles related to "technology," "artificial intelligence," and "programming," these keywords will be extracted. The extracted keywords are then mapped based on their relevance, generating a keyword map for each user.

[0416] Step 3:

[0417] The server analyzes the generated keyword map and eliminates central keywords, including the user's primary interests: "technology," "artificial intelligence," and "programming." It then extracts peripheral keywords (e.g., "biology," "art," and "psychology"). This "reverse" filtering generates a new keyword list.

[0418] Step 4:

[0419] The server searches and collects relevant information from the Internet based on the new keywords extracted by "reverse" filtering. It selects articles, news, research papers, etc. related to the new keywords from the search results and collects appropriate information. The collected information is stored in a database.

[0420] Step 5:

[0421] The server uses artificial intelligence models to generate thumbnail images related to the new information collected. For example, for an article related to "biology," the AI ​​might generate a diagram of DNA or an image of a cell. The generated thumbnail images are then associated with the new information and stored.

[0422] Step 6:

[0423] The terminal displays new information and thumbnail images sent from the server to the user. Through the user interface, the user can access new information and view related thumbnail images, which attracts the user's interest and encourages access to the information.

[0424] Step 7:

[0425] Users rate the new information and thumbnails provided and provide feedback, including interest, relevance, and visual opinion about the information.

[0426] Step 8:

[0427] The device sends feedback data from the user to the server, which adds the collected feedback data to a profile for each user.

[0428] Step 9:

[0429] The server analyzes the feedback data and updates the user's profile, reflecting the user's new trends and interests and optimizing the selection process for the next offering. The analysis results are used for future "reverse" filtering and information gathering processes.

[0430] Example 1

[0431] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0432] Conventional information provision systems provide information tailored to users' interests, but it has been difficult to provide new information that expands users' interests. Furthermore, there is a lack of methods to attract users' attention by using visually appealing thumbnail images. Furthermore, there are also challenges in effectively collecting user feedback and optimizing the information provided based on that feedback.

[0433] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0434] In this invention, the server includes a means for collecting user information, a means for generating a keyword map based on the user information, a means for inverse filtering that excludes keywords at the center of the keyword map and extracts new keywords from peripheral keywords, a means for collecting new information based on the new keywords, a means for generating images related to the new information, a means for providing the new information and the images to the user, and a means for collecting user ratings. This allows the server to efficiently provide new information that is not of interest to the user and generate visually appealing thumbnail images related to that information, thereby attracting the user's interest. Furthermore, by collecting and analyzing user feedback, the server can provide optimized information the next time it provides information.

[0435] "User information" refers to data such as a user's access history, search history, articles viewed, and bookmarked pages.

[0436] A "keyword map" is a graph structure that arranges key keywords extracted based on user interests in a relevance manner.

[0437] "Inverse filtering" is a technique that removes the main keywords in the center of a keyword map and extracts new, relevant keywords that are located on the periphery.

[0438] "New information" refers to content such as articles, news, and research papers that are collected based on new keywords obtained through reverse filtering.

[0439] "Image" refers to a visual thumbnail image generated in association with new information.

[0440] A "generative artificial intelligence model" is an artificial intelligence model that generates new images or text based on input data provided by humans.

[0441] "Evaluation" refers to feedback data that reflects whether the user is interested in the information or images provided, whether it is useful, etc.

[0442] The present invention provides an information provision system based on a user's interest, which is implemented through the following steps.

[0443] The server first collects user information, including the history of websites the user has visited, keywords they have searched for, articles they have viewed, and pages they have bookmarked. This data is collected using, for example, Google Analytics or a dedicated tracking code. The collected data is associated with each user and stored in a database (for example, MySQL or PostgreSQL).

[0444] The server then analyzes the collected user information and extracts key keywords using a natural language processing model (e.g., Hugging Face's BERT model). This analysis may also utilize Term Frequency-Inverse Document Frequency (TF-IDF). For example, if a user frequently views articles related to "programming," "machine learning," and "data science," these keywords will be extracted as their key interests.

[0445] These extracted key keywords are then mapped based on their relevance to generate a keyword map, which has a graph structure with the key keyword at the center and related keywords around it.

[0446] Additionally, the server performs reverse filtering, which removes key keywords (e.g., "programming," "machine learning," etc.) from the center of the keyword map and extracts peripheral keywords (e.g., "psychology," "biology," etc.). This reverse filtering allows for the extraction of new, relevant keywords that are outside the user's usual interests.

[0447] The server collects new information from the Internet based on the new keywords obtained through reverse filtering. Specifically, it uses the Google News API, PubMed API, etc. to collect related articles, news, research papers, etc. This collected information is stored in a database and managed as candidates for information to be provided to users.

[0448] As new information is collected, the server uses generative AI models (such as DALL-E 2 or StyleGAN) to generate thumbnail images related to that information. For example, for articles related to "biology," it might generate diagrams of DNA or images of cells. The generated thumbnail images are then associated with the new information and presented to the user in a visually appealing format.

[0449] The device displays new information and thumbnail images sent from the server to the user. The front-end user interface uses React or Vue.js, and the user is visually captivated by the information and thumbnail images.

[0450] Finally, the user rates the provided information and thumbnail images and enters their feedback. The device collects this feedback data and sends it to the server. For example, it can be in the form of a simple rating such as "useful" or "not interesting." The server analyzes the feedback and updates the user's profile, enabling it to provide more optimized information the next time information is provided.

[0451] Specific examples

[0452] For example, if User A is interested in "technology," "artificial intelligence," and "programming," the server collects related data and uses the BERT model to extract key keywords. It then performs reverse filtering to identify new keywords, such as "biology," "art," and "psychology." Based on these keywords, related information is collected using the Google News API and PubMed API. "Thumbnail images related to biology" generated by DALL-E 2 are provided to User A as visual aids. User A provides feedback on the information provided, and the server uses that feedback to optimize the information provided.

[0453] Prompt Sentence Examples

[0454] "Extract key keywords based on user data, then use reverse filtering to gather new information related to the resulting keywords. Then generate thumbnail images related to the new information."

[0455] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0456] Step 1: Collecting User Data (Server)

[0457] The server collects data such as the history of websites visited by the user, keywords searched, articles viewed, and pages bookmarked. This data is collected using Google Analytics or a dedicated tracking code. The server receives each user's action data and uses this data to create a detailed history database for each user. The collected data is associated with each user and stored in a database (e.g., MySQL, PostgreSQL).

[0458] Step 2: Generate Keyword Mapping (Server)

[0459] The server receives the collected user information as input and uses a natural language processing model (e.g., Hugging Face's BERT model) to extract key keywords. This analysis also uses Term Frequency-Inverse Document Frequency (TF-IDF) scoring of the text data. The output is a list of keywords that indicate the user's main interests. For example, "programming," "machine learning," and "data science" may be extracted.

[0460] Step 3: Generate a Keyword Map (Server)

[0461] The server receives the extracted keywords as input and generates a keyword map based on their relevance. This keyword map has a graph structure with the main keyword at the center and related keywords around it. The output is a graph-structured keyword map.

[0462] Step 4: Performing Reverse Filtering (Server)

[0463] The server receives the generated keyword map as input, eliminates the main keywords located in the center, and extracts peripheral keywords. This process is called inverse filtering. The output is a list of new keywords that are outside the user's usual interests. For example, "psychology," "biology," and "philosophy" are extracted.

[0464] Step 5: Selecting new information (server)

[0465] The server receives the new keywords obtained through reverse filtering as input and collects new information from the Internet based on them, using APIs such as Google News and PubMed. The output is a dataset of articles, news articles, and research papers related to the new keywords. This information is then stored in a database.

[0466] Step 6: Thumbnail generation (server)

[0467] The server takes the collected new information as input and uses a generative AI model (e.g., DALL-E 2 or StyleGAN) to generate relevant thumbnail images. The output is a visually appealing thumbnail image that corresponds to the new information. For example, an article on "biology" might generate a diagram of DNA or an image of a cell.

[0468] Step 7: Provide information and thumbnail (device)

[0469] The terminal receives new information and thumbnail images provided by the server as input and displays them to the user. The front-end user interface uses React or Vue.js. The output is a visually appealing display of information provided to the user. The user can browse these displays.

[0470] Step 8: Collecting User Feedback (Devices and Servers)

[0471] Users rate the provided information and thumbnail images and enter their feedback. The input includes ratings such as "useful" or "not interesting." The device collects this feedback data and sends it to the server. The server outputs the analyzed feedback data, which is used to update the user's profile. This optimizes the next information provided.

[0472] (Application example 1)

[0473] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0474] Conventional information provision systems only provide information based on users' existing interests, making it difficult to stimulate new interests. Furthermore, the methods for providing information based on user data are limited, preventing effective provision of information that will interest users in local stores. Furthermore, there is a lack of information provision based on the user's location information, making it insufficient to improve the user experience in physical stores.

[0475] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0476] In this invention, the server includes means for collecting user interest data, means for generating a keyword map based on the interest data, means for "reverse" filtering to exclude keywords located at the center of the keyword map and extract new keywords from peripheral keywords, means for collecting new information based on the new keywords, means for generating thumbnail images related to the new information, means for providing the new information and the thumbnail images to the user, means for collecting feedback from the user, means for recognizing the user within the terminal, and means for the terminal to provide information based on the user's location, thereby enabling the server to elicit new interests beyond the user's existing interests and improve the in-store experience.

[0477] "User Interest Data" is data about a user's interests, such as the user's website visit history, search history, and bookmarks.

[0478] A "keyword map" is a graph structure that associates key keywords based on user interest data, with the main interest at the center and less relevant keywords at the periphery.

[0479] "Inverse filtering" is a technique that eliminates the main keywords located at the center of the keyword map and extracts new keywords from the remaining peripheral keywords.

[0480] A "thumbnail image" is a small visual image generated in association with new information collected and used to attract user attention.

[0481] "Device" refers to an electronic device used by a user to receive information, including smartphones, tablets, and personal computers.

[0482] "Means of recognizing users" refers to technology that enables devices to detect and identify the user's presence and location within a store, and uses beacons, Wi-Fi, etc.

[0483] User Data Collection

[0484] The server collects user interest data, including user website visit history, search history, and bookmark data, which is stored in a database for each user and used for later analysis.

[0485] Generate Keyword Mapping

[0486] The server uses a natural language processing model to extract key keywords from the collected user data. For example, if a user frequently views articles related to "technology," "artificial intelligence," and "programming," these keywords are extracted. The extracted keywords are then mapped based on their relevance to generate a keyword map for each user. This keyword map has a graph structure that displays keywords near the center as the user's main interests and places less relevant keywords on the periphery.

[0487] Implementing "reverse" filtering

[0488] The server analyzes the generated keyword map and eliminates central keywords. For example, if the user's primary interests are "technology," "artificial intelligence," and "programming," these keywords are excluded. It then extracts more peripheral keywords (e.g., "biology," "art," and "psychology"). This "reverse" filtering extracts keywords that the user would not normally be interested in, but which may provide relevant new information.

[0489] Selection of new information

[0490] The server searches and collects relevant information from the Internet based on the new keywords extracted through "reverse" filtering, such as articles, news, and research papers related to the new keywords. The collected information is stored in a database and later provided to users.

[0491] Thumbnail Generation

[0492] The server uses artificial intelligence models to generate thumbnail images related to new information collected. For example, for articles related to "biology," the AI ​​model might generate diagrams of DNA or images of cells. The generated thumbnail images are then associated with the new information, increasing visual engagement.

[0493] Providing information and thumbnails

[0494] The device displays new information and thumbnail images sent from the server to the user. Through the user interface, the user can access new information and view related thumbnail images, which can attract the user's interest and encourage access to the information. It also recognizes the user within the device and provides more relevant information based on the user's location.

[0495] Collecting user feedback

[0496] The user evaluates the new information and thumbnails provided and enters feedback. The device collects the user's feedback data (e.g., interesting / uninteresting, useful / unhelpful, etc.) and sends it to the server. The server analyzes this feedback and updates the user's profile. Based on the user's feedback, the selection process for the next information offering is optimized.

[0497] Specific examples

[0498] If User A's areas of interest are "technology," "artificial intelligence," and "programming," the user's interest data is collected and key keywords are extracted using a natural language processing model. New keywords such as "biology," "art," and "psychology" are selected using "reverse" filtering, and new information is collected based on these keywords. Visual thumbnail images of this new information are generated by AI and provided to User A. User A views the new information and thumbnail images and provides feedback on whether or not they are interested. The server uses this feedback to provide more appropriate information next time.

[0499] Prompt Sentence Examples

[0500] For users with an interest in "Technology," suggest recent research articles in "Biology" or "Psychology" that are relevant but may spark new interest. Also generate visual thumbnails associated with each article.

[0501] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0502] Step 1:

[0503] The server collects user website visit history, search history, and bookmark data. This data is stored in a database for each user and converted into a format for analysis. The input is the user's online activity history, and the output is formatted user data.

[0504] Step 2:

[0505] The server uses a natural language processing model to extract key keywords from the formatted user data. The input is the user data organized in the previous step, and the output is a list of key keywords. Specifically, the data is input into the natural language processing algorithm to extract frequently occurring keywords.

[0506] Step 3:

[0507] The server visualizes the extracted keywords as a keyword map based on their relevance. The input is a list of key keywords, and the output is a graph-structured map of the keywords. Specifically, it calculates the relevance between keywords and generates the map using a graph library.

[0508] Step 4:

[0509] The server analyzes the generated keyword map, eliminates the main keywords located in the center, and extracts new keywords from the periphery using a "reverse" filtering method. The input is the keyword map, and the output is a list of new keywords. Specifically, the server reduces the weight of the central node and selects new keywords from the periphery.

[0510] Step 5:

[0511] The server searches and collects relevant information from the Internet based on the new keywords extracted by "reverse" filtering. It takes a list of new keywords as input and a list of new information as output. Specifically, it uses a search API to query related information on the web and organizes the results.

[0512] Step 6:

[0513] The server uses an artificial intelligence model to generate thumbnail images related to the new information collected. The input is a list of new information, and the output is a list of thumbnail images. Specifically, the content of the information is input into the AI ​​model, which generates related visual thumbnails.

[0514] Step 7:

[0515] The terminal displays the new information and thumbnail image sent from the server to the user. The input is the new information and thumbnail image, and the output is the information and image displayed on the user's display. The specific operation is to arrange the information and image in the terminal's user interface.

[0516] Step 8:

[0517] Users rate the provided new information and thumbnails and enter their feedback. The input is the user's rating data, and the output is the feedback data. The specific operation is for users to enter their interest and rating through the interface.

[0518] Step 9:

[0519] The terminal collects feedback data from the user and sends it to the server. The input is the user's feedback data, and the output is the feedback data sent to the server. Specifically, the terminal uses the sending function to upload the feedback data to the server.

[0520] Step 10:

[0521] The server analyzes the feedback data from the user and updates the user profile. The input is the feedback data sent from the device, and the output is the updated user profile. Specifically, the server reevaluates the user's interest trends based on the feedback data and optimizes the next information provision.

[0522] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0523] The following describes in detail the mode for carrying out the present invention. In the present invention, a system is constructed that provides new information that the user would not normally come across based on the user's interest data, and further recognizes the user's emotions to optimize the feedback.

[0524] User data collection (server)

[0525] The server collects user access history, search history, bookmark data, and other interest data, including the websites the user has visited, the keywords they have searched for, the articles they have viewed, the pages they have bookmarked, etc. The collected data is stored in a database for each individual user.

[0526] Generate Keyword Mapping (Server)

[0527] The server uses natural language processing models to extract key keywords from the collected user data. For example, if a user frequently views articles about "technology," "artificial intelligence," and "programming," these keywords will be extracted. The extracted keywords are then mapped based on their relevance to generate a keyword map for each user.

[0528] "Reverse" filtering implementation (server)

[0529] The server analyzes the generated keyword map and eliminates central keywords, including the user's primary interests: "technology," "artificial intelligence," and "programming." It then extracts peripheral keywords (e.g., "biology," "art," and "psychology"). This "reverse" filtering generates a new keyword list.

[0530] Selection of new information (server)

[0531] The server searches and collects relevant information from the Internet based on the new keywords extracted by "reverse" filtering. It selects articles, news, research papers, etc. related to the new keywords from the search results and collects appropriate information. The collected information is stored in a database.

[0532] Thumbnail generation (server)

[0533] The server uses artificial intelligence models to generate thumbnail images related to the new information collected. For example, for an article related to "biology," the AI ​​might generate a diagram of DNA or an image of a cell. The generated thumbnail images are then associated with the new information and stored.

[0534] Incorporating emotion recognition (server)

[0535] The server uses an emotion engine to recognize emotions based on user input data (e.g., feedback and voice input). The emotion engine inputs the user's ratings and feedback into an emotion analysis algorithm to analyze the user's emotional state (e.g., happy, interesting, bored, etc.). This emotion data is added to the user's profile.

[0536] Provide information and thumbnails (device)

[0537] The terminal displays new information and thumbnail images sent from the server to the user. Through the user interface, the user can access new information and view related thumbnail images, which attracts the user's interest and encourages access to the information.

[0538] Collecting user feedback (device and server)

[0539] Users rate the new information and thumbnails provided and provide feedback, including interest, relevance, visual opinion, and emotional state (e.g., enjoyable, interesting, boring, etc.).

[0540] The device sends feedback data from the user to the server, which adds the collected feedback data to a profile for each user.

[0541] The server analyzes the feedback data and updates the user's profile, reflecting the user's new trends and interests and optimizing the selection process for the next offering. The analysis results are used for future "reverse" filtering and information gathering processes.

[0542] Specific examples

[0543] If User A's areas of interest are "technology," "artificial intelligence," and "programming," the user's interest data is collected and key keywords are extracted using a natural language processing model. New keywords such as "biology," "art," and "psychology" are selected using "reverse" filtering, and new information is collected based on these keywords. Visual thumbnail images of this new information are generated by AI and provided to User A.

[0544] User A browses the new information and thumbnail images and provides feedback on whether or not they are interested. The emotion engine then analyzes User A's emotions based on their feedback and voice input to understand their emotional state. The server uses this feedback and emotion data to update User A's profile to provide more appropriate information next time.

[0545] This allows users to be exposed to new perspectives and knowledge, and also provides them with the most appropriate information based on their emotions.

[0546] The processing flow will be explained below.

[0547] Step 1:

[0548] The server collects user access history, search history, bookmark data, and other interest data, including the websites the user has visited, the keywords they have searched for, the articles they have viewed, the pages they have bookmarked, etc. The collected data is stored in a database for each individual user.

[0549] Step 2:

[0550] The server uses natural language processing models to extract key keywords from the collected user data. For example, if a user frequently views articles about "technology," "artificial intelligence," and "programming," these keywords will be extracted. The extracted keywords are then mapped based on their relevance to generate a keyword map for each user.

[0551] Step 3:

[0552] The server analyzes the generated keyword map and eliminates central keywords, including the user's primary interests: "technology," "artificial intelligence," and "programming." It then extracts peripheral keywords (e.g., "biology," "art," and "psychology"). This "reverse" filtering generates a new keyword list.

[0553] Step 4:

[0554] The server searches and collects relevant information from the Internet based on the new keywords extracted by "reverse" filtering. It searches for articles, news, research papers, etc. related to the new keywords to collect appropriate information. The collected information is stored in a database.

[0555] Step 5:

[0556] The server uses artificial intelligence models to generate thumbnail images related to the new information collected. For example, for an article related to "biology," the AI ​​might generate a diagram of DNA or an image of a cell. The generated thumbnail images are then associated with the new information and stored.

[0557] Step 6:

[0558] The server uses an emotion engine to recognize emotions based on user input data (e.g., feedback and voice input). The emotion engine inputs the user's ratings and feedback into an emotion analysis algorithm to analyze the user's emotional state (e.g., happy, interesting, bored, etc.). This emotion data is added to the user's profile.

[0559] Step 7:

[0560] The terminal displays new information and thumbnail images sent from the server to the user. Through the user interface, the user can access new information and view related thumbnail images, which attracts the user's interest and encourages access to the information.

[0561] Step 8:

[0562] Users rate the new information and thumbnails provided and provide feedback, including interest and relevance of the information, visual opinion, and emotional state (e.g., enjoyable, interesting, boring, etc.).

[0563] Step 9:

[0564] The device sends feedback data from the user to the server, which adds the collected feedback data to a profile for each user.

[0565] Step 10:

[0566] The server analyzes the feedback data and updates the user's profile, reflecting the user's new trends and interests and optimizing the selection process for the next offering. The analysis results are used for future "reverse" filtering and information gathering processes.

[0567] Specific examples

[0568] As a specific example, we will explain the case where User A's areas of interest are "technology," "artificial intelligence," and "programming."

[0569] Step 1:

[0570] The server collects user A's past access history (e.g., technews.com), search history (e.g., "latest AI algorithm"), bookmarks (e.g., programmingblog.com), etc. and stores them in a database.

[0571] Step 2:

[0572] The server uses a natural language processing model to extract key keywords such as "technology," "artificial intelligence," and "programming" from the collected data, and generates a keyword map based on this.

[0573] Step 3:

[0574] The server removes the central keywords "technology," "artificial intelligence," and "programming" from the generated keyword map and extracts peripheral keywords such as "biology," "art," and "psychology."

[0575] Step 4:

[0576] The server searches the Internet for relevant articles and news based on the extracted new keywords, collects the necessary information, and stores it in a database.

[0577] Step 5:

[0578] The server uses artificial intelligence models to generate thumbnail images related to the new information collected. For example, for an article related to "biology," the AI ​​model generates a diagram of DNA or an image of a cell, which is then stored along with the new information.

[0579] Step 6:

[0580] The server analyzes emotions using an emotion engine based on user feedback and voice input. User A's emotions when viewing information (fun, interesting, bored, etc.) are input into the emotion analysis algorithm, and the analysis results are added to the user profile.

[0581] Step 7:

[0582] The terminal displays the new information and thumbnail image sent from the server and provides them to User A. User A views the new information and thumbnail image.

[0583] Step 8:

[0584] User A evaluates the new information and thumbnails provided and enters feedback (interesting / uninteresting, useful / unhelpful, etc.).

[0585] Step 9:

[0586] The device sends feedback data from user A to the server, which adds this feedback data to the profile for each user A.

[0587] Step 10:

[0588] The server analyzes the feedback data and updates the profile of User A. Based on the feedback and sentiment data, the server optimizes the selection process for the next information to be provided and utilizes it in the selection of new keywords and information gathering process.

[0589] As a result, User A is exposed to new perspectives and knowledge, and is provided with information that is optimally tailored to their emotions.

[0590] Example 2

[0591] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0592] Modern information systems mainly present information based on the user's existing interests, limiting opportunities for users to discover new interests and knowledge. Furthermore, information is often provided in a uniform manner without considering the user's emotional state. This makes it difficult to capture the user's attention, resulting in low information receptivity.

[0593] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting user interest data, means for generating a keyword map based on the interest data, a "reverse" filtering means for excluding keywords located at the center of the keyword map and extracting new keywords from peripheral keywords, means for searching and collecting new information based on the new keywords, means for generating visual content related to the new information, means for providing the new information and the visual content to the user, means for collecting feedback and emotional state from the user, and means for updating a user profile based on the feedback and emotional data. This allows the user to discover new information that they have not encountered before and provides optimal information according to their emotional state.

[0594] "User interest data" is a general term for behavioral history that indicates a user's interests and concerns, such as websites visited by the user, keywords searched for, articles viewed, and pages bookmarked.

[0595] A "keyword map" is a visual or digital mapping of key keywords extracted from user interest data and their relationships.

[0596] "Inverse filtering" is the process of eliminating the main keywords located in the center of the keyword map and extracting new keywords located around them.

[0597] "New information" is related information collected from the Internet and databases based on new keywords extracted by reverse filtering.

[0598] "Visual content" refers to images and visual materials related to new information generated using artificial intelligence models, etc.

[0599] "Feedback" refers to the evaluations and impressions users make of new information and visual content.

[0600] "Emotional state" refers to the user's emotional response or state, as analyzed from user feedback, voice input, etc.

[0601] The embodiments of the present invention will be described in detail. In this invention, new information that the user would not normally come across is provided based on the user's interest data. Also, a system is constructed that recognizes the user's emotions and optimizes feedback.

[0602] The server first collects interest data, such as the websites visited by the user, keywords searched, articles viewed, pages bookmarked, etc. This data is collected using technologies such as cookies and local storage data from the browser, and the collected data is stored in a database for each individual user.

[0603] The server then uses the collected user data to extract key keywords using a natural language processing (NLP) model, such as BERT or GPT, and maps the extracted keywords based on their relevance to generate a keyword map for each user.

[0604] In the process of analyzing the generated keyword map, the server will eliminate the main keywords located in the center and extract new keywords located in the periphery. This "reverse filtering" process will reveal keywords in areas that users are not usually interested in.

[0605] The server then searches and collects relevant information from the Internet based on the new keywords extracted by reverse filtering, using Google Search API and News API to retrieve articles and research papers related to the keywords and store them in a database.

[0606] The server then uses artificial intelligence models such as DALL-E and MidJourney to generate visual content related to the collected information: for example, articles related to "biology" will generate models of DNA and microscopic images of cells.

[0607] The terminal displays new information and visual content sent from the server to the user, and through a web browser or application interface, the user can access the new information and view the associated visual content.

[0608] Users provide feedback on the new information and visual content provided, including ratings of interest and relevance of the information. The device sends the feedback data to a server, where it is added to each user's profile.

[0609] The server uses an emotion engine to analyze emotions from user feedback and voice input. The feedback text is fed into the emotion analysis algorithm to determine the user's emotional state. This emotion data is also added to the user profile.

[0610] Finally, the server analyzes the feedback and sentiment data and updates the user's profile to reflect new trends and interests and optimize the next round of information delivery.

[0611] Specific examples

[0612] For example, if User A's areas of interest are "technology," "artificial intelligence," and "programming," these interest data are collected and key keywords are extracted using the NLP model. New keywords such as "biology," "art," and "psychology" are selected through reverse filtering, and new information is collected based on these keywords. Visual thumbnail images of this new information are generated by the AI ​​model and provided to User A.

[0613] Prompt Sentence Examples

[0614] "Suggest relevant articles and visual content based on the user's emerging areas of interest."

[0615] "Generate thumbnail images of new information based on reverse-filtered keywords."

[0616] "Use user feedback and sentiment data to suggest ways to optimize your next offering."

[0617] This allows users to access new perspectives and knowledge, and provides optimal information tailored to their emotions.

[0618] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0619] Step 1: Collecting User Data (Server)

[0620] The server collects behavioral data from the user's browser, such as the websites visited, keywords searched, articles viewed, and pages bookmarked, and uses the browser's cookies and local storage to obtain the data and store it in a database for each individual user.

[0621] Input: User behavior history data (website URL, search keywords, viewed articles, bookmark data)

[0622] Output: Database entries per user

[0623] Step 2: Generate Keyword Mapping (Server)

[0624] The server extracts key keywords from the collected user data using natural language processing (NLP) models such as BERT and GPT, which analyze and extract important keywords from text data, and then generates a keyword map based on the relevance of the keywords.

[0625] Input: User behavior history data

[0626] Output: Keyword map per user

[0627] Step 3: Performing "reverse" filtering (server)

[0628] The server analyzes the generated keyword map, eliminates the main keywords located in the center, and extracts new keywords located around these eliminated keywords, thereby obtaining keywords in areas that users are not usually interested in.

[0629] Input: User-specific keyword map

[0630] Output: New keyword list

[0631] Step 4: Selecting new information (server)

[0632] The server searches and collects relevant information from the Internet based on the new keywords extracted by reverse filtering, using Google Search API and News API to retrieve articles, research papers, etc. related to the keywords.

[0633] Input: New keyword list

[0634] Output: New information collected (articles, news, research papers, etc.)

[0635] Step 5: Generate visual content (server)

[0636] The server uses artificial intelligence models to generate visual content related to the collected information, for example, image generation AI such as DALL-E or MidJourney to generate visual content related to keywords.

[0637] Input: New information (articles, news, research papers, etc.)

[0638] Output: Relevant visual content (e.g. thumbnail image)

[0639] Step 6: Providing information and visual content (device)

[0640] The device displays new information and visual content sent from the server to the user. Through a web browser or application interface, the user can access new information and view related visual content (such as thumbnail images).

[0641] Input: New information and visual content sent from the server

[0642] Output: Information and visual content displayed in the user interface

[0643] Step 7: Collecting User Feedback (Devices and Servers)

[0644] The user inputs ratings and feedback for the new information and visual content provided, including interest in the information, relevance, visual opinion, and emotional state. The device then transmits the feedback data to the server.

[0645] Input: User feedback data

[0646] Output: Feedback data sent to the server

[0647] Step 8: Integrating Emotion Recognition (Server)

[0648] The server uses an emotion engine to analyze emotions from the user's feedback and voice input, and uses an emotion analysis algorithm to determine the user's emotional state from the feedback text.

[0649] Input: User feedback and voice input data

[0650] Output: Parsed user sentiment data

[0651] Step 9: Update Profile (Server)

[0652] The server analyzes the feedback and sentiment data and updates the user profile, so that new trends and interests are reflected in the profile and the next information delivery is optimized.

[0653] Input: User feedback data, sentiment data

[0654] Output: Updated user profile

[0655] The above are the specific processing steps of this system.

[0656] (Application example 2)

[0657] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0658] Conventional information delivery systems tend to only provide information that users are already interested in, limiting opportunities to encounter new perspectives and knowledge. Furthermore, it is difficult to provide optimal information based on the user's interests and emotions, resulting in a lack of improvement in the user experience. Furthermore, the lack of feedback optimization incorporating emotion recognition can reduce the accuracy and relevance of information provided.

[0659] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0660] In this invention, the server includes means for collecting user interest data, means for generating a keyword map based on the interest data, a "reverse" filtering means for excluding keywords located at the center of the keyword map and extracting new keywords from peripheral keywords, means for collecting new information based on the new keywords, means for generating thumbnail images related to the new information, means for providing the new information and the thumbnail images to the user, means for collecting feedback from the user, emotion recognition means for analyzing the user's emotions based on the feedback, and means for optimizing the next information to be provided based on the analyzed emotions. This allows the user to be exposed to new perspectives and knowledge and to be provided with optimal information according to their emotions.

[0661] "User interest data" refers to data that reflects a user's interests, such as user access history, search history, and bookmark data.

[0662] A "keyword map" is a map showing key keyword relationships generated using natural language processing models based on user interest data.

[0663] "Inverse filtering" is a technique that eliminates the main keywords located at the center of the keyword map and extracts new keywords from the periphery.

[0664] "New information" is information related to new perspectives and knowledge that is collected based on new keywords extracted by reverse filtering.

[0665] A "thumbnail image" is a small visual image associated with the new information collected.

[0666] "User feedback" refers to users' ratings and impressions of the provided information and thumbnail images.

[0667] An "emotion recognition means" is an engine or algorithm that analyzes emotions based on user feedback and understands their emotional state.

[0668] "Information optimization means" refers to the means of adjusting and optimizing the content of the next information provided based on the analyzed user sentiment.

[0669] A "natural language processing model" is an algorithm or machine learning model for extracting and analyzing meaning and keywords from text data.

[0670] An "artificial intelligence model" is an algorithm or system that learns from large amounts of data and automates specific tasks.

[0671] The present invention provides a technology for constructing a system that provides new information that a user would not normally come across, and further recognizes the user's emotions and optimizes feedback. Specific embodiments of the present invention will be described below.

[0672] System configuration

[0673] This system consists of a server that collects user interest data, a server that generates a keyword map based on the interest data, a server that performs "reverse" filtering to eliminate major keywords from the generated keyword map and extract new keywords, a server that collects new information and generates thumbnail images, a terminal that provides information and thumbnail images to users, a server that collects user feedback and performs emotion recognition, and a server that optimizes the information to be provided next time.

[0674] Collection and Creation

[0675] The server collects interest data such as user access history, search history, and bookmark data, including the websites the user has visited, the keywords they have searched for, the articles they have viewed, the pages they have bookmarked, etc. The collected data is stored in a database for each user.

[0676] The server then uses natural language processing models to extract key keywords from the collected user data. For example, if a user frequently views articles about "technology," "artificial intelligence," and "programming," these keywords will be extracted. The extracted keywords are then mapped based on their relevance to generate a keyword map for each user.

[0677] Reverse filtering and new information gathering

[0678] The generated keyword map is analyzed and the main keywords located in the center are eliminated. This elimination includes the user's main interests, such as "technology," "artificial intelligence," and "programming." New keywords located in the periphery (e.g., "biology," "art," and "psychology") are then extracted. This "reverse" filtering generates a new keyword list.

[0679] The server searches and collects relevant information from the Internet or specific information sources based on the new keywords extracted by "reverse" filtering. It selects articles, news, research papers, etc. related to the new keywords from the search results and collects appropriate information. The collected information is stored in a database.

[0680] Thumbnail generation and serving

[0681] The server uses artificial intelligence models to generate thumbnail images related to the new information collected. For example, for an article related to "biology," the generated thumbnail image might be a diagram of DNA or an image of a cell. The generated thumbnail image is associated with the new information and stored.

[0682] The terminal displays the new information and thumbnail images sent from the server to the user. Through the user interface, the user can access the new information and view the related thumbnail images.

[0683] Feedback collection and emotion recognition

[0684] The user evaluates the provided new information and thumbnails and enters feedback, including interest, relevance, visual evaluation, and emotional state (e.g., fun, interesting, boring, etc.). The device then transmits this feedback data to the server.

[0685] The server uses an emotion engine to recognize emotions based on user input data. The emotion engine inputs user ratings and feedback into an emotion analysis algorithm to analyze the user's emotional state. This emotion data is added to the user's profile.

[0686] Optimizing next advertisement

[0687] The server analyzes the feedback data and sentiment data and updates the user's profile to reflect the user's new trends and interests and optimize the next information offering.

[0688] Examples of specific examples and prompts

[0689] As a concrete example, assume that User A is interested in "technology," "artificial intelligence," and "programming." This user's interest data is collected, and key keywords are extracted using a natural language processing model. New keywords such as "biology," "art," and "psychology" are selected using a "reverse" filtering method, and new information is collected based on these keywords. Visual thumbnail images of this new information are generated by AI and provided to User A. User A views the new information and thumbnail images and provides feedback on whether or not they are interested. Furthermore, an emotion engine analyzes emotions based on User A's feedback and voice input, and grasps the user's emotional state. The server uses this feedback and emotion data to update User A's profile to provide more appropriate information next time.

[0690] An example prompt might be, "The user is interested in technology, artificial intelligence, and programming. After reverse filtering, the keywords are biology, art, and psychology. Based on these keywords, please find and provide relevant, up-to-date articles and information."

[0691] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0692] Step 1:

[0693] Server collection of user interest data

[0694] The server collects interest data such as user access history, search history, and bookmark data, including the websites the user visits, keywords searched, articles viewed, and pages bookmarked. The collected data is stored in a database as individual user profiles.

[0695] Input: Access history, search history, bookmark data

[0696] Output: Save interest data in user profile

[0697] Step 2:

[0698] Server-generated keyword mappings

[0699] The server extracts key keywords from the collected interest data using natural language processing models (e.g., TF-IDF-based vectorization and clustering). For example, if a user frequently views articles related to "technology," "artificial intelligence," and "programming," these keywords will be extracted. The extracted keywords are then mapped to a keyword map based on their relevance.

[0700] Input: Interest data (user profile)

[0701] Output: Keyword map (including a list of primary keywords)

[0702] Step 3:

[0703] Server performs "reverse" filtering

[0704] The server analyzes the generated keyword map and eliminates central keywords, including the user's main interests: "technology," "artificial intelligence," and "programming." It then extracts peripheral new keywords (e.g., "biology," "art," and "psychology").

[0705] Input: Keyword Map

[0706] Output: New "reverse" filtered keyword list

[0707] Step 4:

[0708] Server selection and collection of new information

[0709] The server searches and collects relevant information from the Internet and specific information sources based on the new keywords extracted through "reverse" filtering, and stores the information, such as articles, news, and research papers, related to the new keywords in a database.

[0710] Input: New "reverse" filtered keyword list

[0711] Output: New information (articles, news, research papers, etc.)

[0712] Step 5:

[0713] Server-generated thumbnail images

[0714] The server uses artificial intelligence models (e.g., image generation AI) to generate thumbnail images related to the new information collected. For example, for an article related to "biology," the generated thumbnail images might be diagrams of DNA or images of cells.

[0715] Input: New information

[0716] Output: Thumbnail image

[0717] Step 6:

[0718] Device provides information and thumbnail images

[0719] The terminal displays the new information and thumbnail images sent from the server to the user, and through the user interface, the user can access the new information and visually check the related thumbnail images.

[0720] Input: New information, thumbnail image

[0721] Output: User interface display (new information and thumbnail image)

[0722] Step 7:

[0723] Collecting and entering user feedback

[0724] Users input feedback about the new information and thumbnail images provided, such as interest, relevance, rating, visual opinion, and emotional state, which is then sent to the server via the device.

[0725] Input: User feedback on new information and thumbnail images

[0726] Output: Feedback data (ratings, emotional state, etc.)

[0727] Step 8:

[0728] Emotion recognition and analysis by the server

[0729] The server uses an emotion recognition engine to analyze the user's emotional state based on the feedback data received from the user, and the analysis results are added to the user's profile.

[0730] Input: Feedback data

[0731] Output: Emotion data (fun, interesting, bored, etc.)

[0732] Step 9:

[0733] Server optimization of next information provision

[0734] The server analyzes the feedback and sentiment data and updates the user's profile to reflect new trends and interests and optimize the content of the next information provided.

[0735] Input: Feedback data, emotion data

[0736] Output: Updated user profile, optimization plan for next information provided

[0737] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0738] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0739] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0740] [Third embodiment]

[0741] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0742] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0743] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0744] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0745] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0746] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0747] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0748] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0749] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0750] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0751] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0752] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0753] The embodiments for carrying out the present invention will be described in detail below.

[0754] User data collection (server)

[0755] The server collects users' access history, search history, bookmark data, etc. This includes the websites they have visited, the keywords they have searched for, the articles they have viewed, the pages they have bookmarked, etc. The collected data is stored in a database for each individual user and used for later analysis.

[0756] Generate Keyword Mapping (Server)

[0757] The server uses a natural language processing model to extract key keywords from the collected user data. For example, if a user frequently views articles related to "technology," "artificial intelligence," and "programming," these keywords are extracted. The extracted keywords are then mapped based on their relevance to generate a keyword map for each user. This keyword map has a graph structure that displays keywords near the center as the user's main interests and places less relevant keywords on the periphery.

[0758] "Reverse" filtering implementation (server)

[0759] The server analyzes the generated keyword map and eliminates central keywords. For example, if the user's primary interests are "technology," "artificial intelligence," and "programming," these keywords are excluded. It then extracts more peripheral keywords (e.g., "biology," "art," and "psychology"). This "reverse" filtering extracts keywords that the user would not normally be interested in, but which may provide relevant new information.

[0760] Selection of new information (server)

[0761] The server searches and collects relevant information from the Internet based on the new keywords extracted through "reverse" filtering, such as articles, news, and research papers related to the new keywords. The collected information is stored in a database and later provided to users.

[0762] Thumbnail generation (server)

[0763] The server uses artificial intelligence models to generate thumbnail images related to new information collected. For example, for articles related to "biology," the AI ​​model might generate diagrams of DNA or images of cells. The generated thumbnail images are then associated with the new information, increasing visual engagement.

[0764] Provide information and thumbnails (device)

[0765] The terminal displays new information and thumbnail images sent from the server to the user. Through the user interface, the user can access new information and view related thumbnail images, which attracts the user's interest and encourages access to the information.

[0766] Collecting user feedback (device and server)

[0767] The user evaluates the new information and thumbnails provided and enters feedback. The device collects the user's feedback data (e.g., interesting / uninteresting, useful / unhelpful, etc.) and sends it to the server. The server analyzes this feedback and updates the user's profile. Based on the user's feedback, the selection process for the next information offering is optimized.

[0768] Specific examples

[0769] If User A's areas of interest are "technology," "artificial intelligence," and "programming," the user's interest data is collected and key keywords are extracted using a natural language processing model. New keywords such as "biology," "art," and "psychology" are selected using "reverse" filtering, and new information is collected based on these keywords. Visual thumbnail images of this new information are generated by AI and provided to User A. User A views the new information and thumbnail images and provides feedback on whether or not they are interested. The server uses this feedback to provide more appropriate information next time.

[0770] The processing flow will be explained below.

[0771] Step 1:

[0772] The server collects information about users' interests, such as their access history, search history, and bookmark data, including the websites they visit, the keywords they search for, the articles they read, and the pages they bookmark. The collected data is stored in a database for each individual user.

[0773] Step 2:

[0774] The server uses natural language processing models to extract key keywords from the collected user data. For example, if a user frequently views articles related to "technology," "artificial intelligence," and "programming," these keywords will be extracted. The extracted keywords are then mapped based on their relevance, generating a keyword map for each user.

[0775] Step 3:

[0776] The server analyzes the generated keyword map and eliminates central keywords, including the user's primary interests: "technology," "artificial intelligence," and "programming." It then extracts peripheral keywords (e.g., "biology," "art," and "psychology"). This "reverse" filtering generates a new keyword list.

[0777] Step 4:

[0778] The server searches and collects relevant information from the Internet based on the new keywords extracted by "reverse" filtering. It selects articles, news, research papers, etc. related to the new keywords from the search results and collects appropriate information. The collected information is stored in a database.

[0779] Step 5:

[0780] The server uses artificial intelligence models to generate thumbnail images related to the new information collected. For example, for an article related to "biology," the AI ​​might generate a diagram of DNA or an image of a cell. The generated thumbnail images are then associated with the new information and stored.

[0781] Step 6:

[0782] The terminal displays new information and thumbnail images sent from the server to the user. Through the user interface, the user can access new information and view related thumbnail images, which attracts the user's interest and encourages access to the information.

[0783] Step 7:

[0784] Users rate the new information and thumbnails provided and provide feedback, including interest, relevance, and visual opinion about the information.

[0785] Step 8:

[0786] The device sends feedback data from the user to the server, which adds the collected feedback data to a profile for each user.

[0787] Step 9:

[0788] The server analyzes the feedback data and updates the user's profile, reflecting the user's new trends and interests and optimizing the selection process for the next offering. The analysis results are used for future "reverse" filtering and information gathering processes.

[0789] Example 1

[0790] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0791] Conventional information provision systems provide information tailored to users' interests, but it has been difficult to provide new information that expands users' interests. Furthermore, there is a lack of methods to attract users' attention by using visually appealing thumbnail images. Furthermore, there are also challenges in effectively collecting user feedback and optimizing the information provided based on that feedback.

[0792] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0793] In this invention, the server includes a means for collecting user information, a means for generating a keyword map based on the user information, a means for inverse filtering that excludes keywords at the center of the keyword map and extracts new keywords from peripheral keywords, a means for collecting new information based on the new keywords, a means for generating images related to the new information, a means for providing the new information and the images to the user, and a means for collecting user ratings. This allows the server to efficiently provide new information that is not of interest to the user and generate visually appealing thumbnail images related to that information, thereby attracting the user's interest. Furthermore, by collecting and analyzing user feedback, the server can provide optimized information the next time it provides information.

[0794] "User information" refers to data such as a user's access history, search history, articles viewed, and bookmarked pages.

[0795] A "keyword map" is a graph structure that arranges key keywords extracted based on user interests in a relevance manner.

[0796] "Inverse filtering" is a technique that removes the main keywords in the center of a keyword map and extracts new, relevant keywords that are located on the periphery.

[0797] "New information" refers to content such as articles, news, and research papers that are collected based on new keywords obtained through reverse filtering.

[0798] "Image" refers to a visual thumbnail image generated in association with new information.

[0799] A "generative artificial intelligence model" is an artificial intelligence model that generates new images or text based on input data provided by humans.

[0800] "Evaluation" refers to feedback data that reflects whether the user is interested in the information or images provided, whether it is useful, etc.

[0801] The present invention provides an information provision system based on a user's interest, which is implemented through the following steps.

[0802] The server first collects user information, including the history of websites the user has visited, keywords they have searched for, articles they have viewed, and pages they have bookmarked. This data is collected using, for example, Google Analytics or a dedicated tracking code. The collected data is associated with each user and stored in a database (for example, MySQL or PostgreSQL).

[0803] The server then analyzes the collected user information and extracts key keywords using a natural language processing model (e.g., Hugging Face's BERT model). This analysis may also utilize Term Frequency-Inverse Document Frequency (TF-IDF). For example, if a user frequently views articles related to "programming," "machine learning," and "data science," these keywords will be extracted as their key interests.

[0804] These extracted key keywords are then mapped based on their relevance to generate a keyword map, which has a graph structure with the key keyword at the center and related keywords around it.

[0805] Additionally, the server performs reverse filtering, which removes key keywords (e.g., "programming," "machine learning," etc.) from the center of the keyword map and extracts peripheral keywords (e.g., "psychology," "biology," etc.). This reverse filtering allows for the extraction of new, relevant keywords that are outside the user's usual interests.

[0806] The server collects new information from the Internet based on the new keywords obtained through reverse filtering. Specifically, it uses the Google News API, PubMed API, etc. to collect related articles, news, research papers, etc. This collected information is stored in a database and managed as candidates for information to be provided to users.

[0807] As new information is collected, the server uses generative AI models (such as DALL-E 2 or StyleGAN) to generate thumbnail images related to that information. For example, for articles related to "biology," it might generate diagrams of DNA or images of cells. The generated thumbnail images are then associated with the new information and presented to the user in a visually appealing format.

[0808] The device displays new information and thumbnail images sent from the server to the user. The front-end user interface uses React or Vue.js, and the user is visually captivated by the information and thumbnail images.

[0809] Finally, the user rates the provided information and thumbnail images and enters their feedback. The device collects this feedback data and sends it to the server. For example, it can be in the form of a simple rating such as "useful" or "not interesting." The server analyzes the feedback and updates the user's profile, enabling it to provide more optimized information the next time information is provided.

[0810] Specific examples

[0811] For example, if User A is interested in "technology," "artificial intelligence," and "programming," the server collects related data and uses the BERT model to extract key keywords. It then performs reverse filtering to identify new keywords, such as "biology," "art," and "psychology." Based on these keywords, related information is collected using the Google News API and PubMed API. "Thumbnail images related to biology" generated by DALL-E 2 are provided to User A as visual aids. User A provides feedback on the information provided, and the server uses that feedback to optimize the information provided.

[0812] Prompt Sentence Examples

[0813] "Extract key keywords based on user data, then use reverse filtering to gather new information related to the resulting keywords. Then generate thumbnail images related to the new information."

[0814] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0815] Step 1: Collecting User Data (Server)

[0816] The server collects data such as the history of websites visited by the user, keywords searched, articles viewed, and pages bookmarked. This data is collected using Google Analytics or a dedicated tracking code. The server receives each user's action data and uses this data to create a detailed history database for each user. The collected data is associated with each user and stored in a database (e.g., MySQL, PostgreSQL).

[0817] Step 2: Generate Keyword Mapping (Server)

[0818] The server receives the collected user information as input and uses a natural language processing model (e.g., Hugging Face's BERT model) to extract key keywords. This analysis also uses Term Frequency-Inverse Document Frequency (TF-IDF) scoring of the text data. The output is a list of keywords that indicate the user's main interests. For example, "programming," "machine learning," and "data science" may be extracted.

[0819] Step 3: Generate a Keyword Map (Server)

[0820] The server receives the extracted keywords as input and generates a keyword map based on their relevance. This keyword map has a graph structure with the main keyword at the center and related keywords around it. The output is a graph-structured keyword map.

[0821] Step 4: Performing Reverse Filtering (Server)

[0822] The server receives the generated keyword map as input, eliminates the main keywords located in the center, and extracts peripheral keywords. This process is called inverse filtering. The output is a list of new keywords that are outside the user's usual interests. For example, "psychology," "biology," and "philosophy" are extracted.

[0823] Step 5: Selecting new information (server)

[0824] The server receives the new keywords obtained through reverse filtering as input and collects new information from the Internet based on them, using APIs such as Google News and PubMed. The output is a dataset of articles, news articles, and research papers related to the new keywords. This information is then stored in a database.

[0825] Step 6: Thumbnail generation (server)

[0826] The server takes the collected new information as input and uses a generative AI model (e.g., DALL-E 2 or StyleGAN) to generate relevant thumbnail images. The output is a visually appealing thumbnail image that corresponds to the new information. For example, an article on "biology" might generate a diagram of DNA or an image of a cell.

[0827] Step 7: Provide information and thumbnail (device)

[0828] The terminal receives new information and thumbnail images provided by the server as input and displays them to the user. The front-end user interface uses React or Vue.js. The output is a visually appealing display of information provided to the user. The user can browse these displays.

[0829] Step 8: Collecting User Feedback (Devices and Servers)

[0830] Users rate the provided information and thumbnail images and enter their feedback. The input includes ratings such as "useful" or "not interesting." The device collects this feedback data and sends it to the server. The server outputs the analyzed feedback data, which is used to update the user's profile. This optimizes the next information provided.

[0831] (Application example 1)

[0832] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0833] Conventional information provision systems only provide information based on users' existing interests, making it difficult to stimulate new interests. Furthermore, the methods for providing information based on user data are limited, preventing effective provision of information that will interest users in local stores. Furthermore, there is a lack of information provision based on the user's location information, making it insufficient to improve the user experience in physical stores.

[0834] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0835] In this invention, the server includes means for collecting user interest data, means for generating a keyword map based on the interest data, means for "reverse" filtering to exclude keywords located at the center of the keyword map and extract new keywords from peripheral keywords, means for collecting new information based on the new keywords, means for generating thumbnail images related to the new information, means for providing the new information and the thumbnail images to the user, means for collecting feedback from the user, means for recognizing the user within the terminal, and means for the terminal to provide information based on the user's location, thereby enabling the server to elicit new interests beyond the user's existing interests and improve the in-store experience.

[0836] "User Interest Data" is data about a user's interests, such as the user's website visit history, search history, and bookmarks.

[0837] A "keyword map" is a graph structure that associates key keywords based on user interest data, with the main interest at the center and less relevant keywords at the periphery.

[0838] "Inverse filtering" is a technique that eliminates the main keywords located at the center of the keyword map and extracts new keywords from the remaining peripheral keywords.

[0839] A "thumbnail image" is a small visual image generated in association with new information collected and used to attract user attention.

[0840] "Device" refers to an electronic device used by a user to receive information, including smartphones, tablets, and personal computers.

[0841] "Means of recognizing users" refers to technology that enables devices to detect and identify the user's presence and location within a store, and uses beacons, Wi-Fi, etc.

[0842] User Data Collection

[0843] The server collects user interest data, including user website visit history, search history, and bookmark data, which is stored in a database for each user and used for later analysis.

[0844] Generate Keyword Mapping

[0845] The server uses a natural language processing model to extract key keywords from the collected user data. For example, if a user frequently views articles related to "technology," "artificial intelligence," and "programming," these keywords are extracted. The extracted keywords are then mapped based on their relevance to generate a keyword map for each user. This keyword map has a graph structure that displays keywords near the center as the user's main interests and places less relevant keywords on the periphery.

[0846] Implementing "reverse" filtering

[0847] The server analyzes the generated keyword map and eliminates central keywords. For example, if the user's primary interests are "technology," "artificial intelligence," and "programming," these keywords are excluded. It then extracts more peripheral keywords (e.g., "biology," "art," and "psychology"). This "reverse" filtering extracts keywords that the user would not normally be interested in, but which may provide relevant new information.

[0848] Selection of new information

[0849] The server searches and collects relevant information from the Internet based on the new keywords extracted through "reverse" filtering, such as articles, news, and research papers related to the new keywords. The collected information is stored in a database and later provided to users.

[0850] Thumbnail Generation

[0851] The server uses artificial intelligence models to generate thumbnail images related to new information collected. For example, for articles related to "biology," the AI ​​model might generate diagrams of DNA or images of cells. The generated thumbnail images are then associated with the new information, increasing visual engagement.

[0852] Providing information and thumbnails

[0853] The device displays new information and thumbnail images sent from the server to the user. Through the user interface, the user can access new information and view related thumbnail images, which can attract the user's interest and encourage access to the information. It also recognizes the user within the device and provides more relevant information based on the user's location.

[0854] Collecting user feedback

[0855] The user evaluates the new information and thumbnails provided and enters feedback. The device collects the user's feedback data (e.g., interesting / uninteresting, useful / unhelpful, etc.) and sends it to the server. The server analyzes this feedback and updates the user's profile. Based on the user's feedback, the selection process for the next information offering is optimized.

[0856] Specific examples

[0857] If User A's areas of interest are "technology," "artificial intelligence," and "programming," the user's interest data is collected and key keywords are extracted using a natural language processing model. New keywords such as "biology," "art," and "psychology" are selected using "reverse" filtering, and new information is collected based on these keywords. Visual thumbnail images of this new information are generated by AI and provided to User A. User A views the new information and thumbnail images and provides feedback on whether or not they are interested. The server uses this feedback to provide more appropriate information next time.

[0858] Prompt Sentence Examples

[0859] For users with an interest in "Technology," suggest recent research articles in "Biology" or "Psychology" that are relevant but may spark new interest. Also generate visual thumbnails associated with each article.

[0860] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0861] Step 1:

[0862] The server collects user website visit history, search history, and bookmark data. This data is stored in a database for each user and converted into a format for analysis. The input is the user's online activity history, and the output is formatted user data.

[0863] Step 2:

[0864] The server uses a natural language processing model to extract key keywords from the formatted user data. The input is the user data organized in the previous step, and the output is a list of key keywords. Specifically, the data is input into the natural language processing algorithm to extract frequently occurring keywords.

[0865] Step 3:

[0866] The server visualizes the extracted keywords as a keyword map based on their relevance. The input is a list of key keywords, and the output is a graph-structured map of the keywords. Specifically, it calculates the relevance between keywords and generates the map using a graph library.

[0867] Step 4:

[0868] The server analyzes the generated keyword map, eliminates the main keywords located in the center, and extracts new keywords from the periphery using a "reverse" filtering method. The input is the keyword map, and the output is a list of new keywords. Specifically, the server reduces the weight of the central node and selects new keywords from the periphery.

[0869] Step 5:

[0870] The server searches and collects relevant information from the Internet based on the new keywords extracted by "reverse" filtering. It takes a list of new keywords as input and a list of new information as output. Specifically, it uses a search API to query related information on the web and organizes the results.

[0871] Step 6:

[0872] The server uses an artificial intelligence model to generate thumbnail images related to the new information collected. The input is a list of new information, and the output is a list of thumbnail images. Specifically, the content of the information is input into the AI ​​model, which generates related visual thumbnails.

[0873] Step 7:

[0874] The terminal displays the new information and thumbnail image sent from the server to the user. The input is the new information and thumbnail image, and the output is the information and image displayed on the user's display. The specific operation is to arrange the information and image in the terminal's user interface.

[0875] Step 8:

[0876] Users rate the provided new information and thumbnails and enter their feedback. The input is the user's rating data, and the output is the feedback data. The specific operation is for users to enter their interest and rating through the interface.

[0877] Step 9:

[0878] The terminal collects feedback data from the user and sends it to the server. The input is the user's feedback data, and the output is the feedback data sent to the server. Specifically, the terminal uses the sending function to upload the feedback data to the server.

[0879] Step 10:

[0880] The server analyzes the feedback data from the user and updates the user profile. The input is the feedback data sent from the device, and the output is the updated user profile. Specifically, the server reevaluates the user's interest trends based on the feedback data and optimizes the next information provision.

[0881] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0882] The following describes in detail the mode for carrying out the present invention. In the present invention, a system is constructed that provides new information that the user would not normally come across based on the user's interest data, and further recognizes the user's emotions to optimize the feedback.

[0883] User data collection (server)

[0884] The server collects user access history, search history, bookmark data, and other interest data, including the websites the user has visited, the keywords they have searched for, the articles they have viewed, the pages they have bookmarked, etc. The collected data is stored in a database for each individual user.

[0885] Generate Keyword Mapping (Server)

[0886] The server uses natural language processing models to extract key keywords from the collected user data. For example, if a user frequently views articles about "technology," "artificial intelligence," and "programming," these keywords will be extracted. The extracted keywords are then mapped based on their relevance to generate a keyword map for each user.

[0887] "Reverse" filtering implementation (server)

[0888] The server analyzes the generated keyword map and eliminates central keywords, including the user's primary interests: "technology," "artificial intelligence," and "programming." It then extracts peripheral keywords (e.g., "biology," "art," and "psychology"). This "reverse" filtering generates a new keyword list.

[0889] Selection of new information (server)

[0890] The server searches and collects relevant information from the Internet based on the new keywords extracted by "reverse" filtering. It selects articles, news, research papers, etc. related to the new keywords from the search results and collects appropriate information. The collected information is stored in a database.

[0891] Thumbnail generation (server)

[0892] The server uses artificial intelligence models to generate thumbnail images related to the new information collected. For example, for an article related to "biology," the AI ​​might generate a diagram of DNA or an image of a cell. The generated thumbnail images are then associated with the new information and stored.

[0893] Incorporating emotion recognition (server)

[0894] The server uses an emotion engine to recognize emotions based on user input data (e.g., feedback and voice input). The emotion engine inputs the user's ratings and feedback into an emotion analysis algorithm to analyze the user's emotional state (e.g., happy, interesting, bored, etc.). This emotion data is added to the user's profile.

[0895] Provide information and thumbnails (device)

[0896] The terminal displays new information and thumbnail images sent from the server to the user. Through the user interface, the user can access new information and view related thumbnail images, which attracts the user's interest and encourages access to the information.

[0897] Collecting user feedback (device and server)

[0898] Users rate the new information and thumbnails provided and provide feedback, including interest, relevance, visual opinion, and emotional state (e.g., enjoyable, interesting, boring, etc.).

[0899] The device sends feedback data from the user to the server, which adds the collected feedback data to a profile for each user.

[0900] The server analyzes the feedback data and updates the user's profile, reflecting the user's new trends and interests and optimizing the selection process for the next offering. The analysis results are used for future "reverse" filtering and information gathering processes.

[0901] Specific examples

[0902] If User A's areas of interest are "technology," "artificial intelligence," and "programming," the user's interest data is collected and key keywords are extracted using a natural language processing model. New keywords such as "biology," "art," and "psychology" are selected using "reverse" filtering, and new information is collected based on these keywords. Visual thumbnail images of this new information are generated by AI and provided to User A.

[0903] User A browses the new information and thumbnail images and provides feedback on whether or not they are interested. The emotion engine then analyzes User A's emotions based on their feedback and voice input to understand their emotional state. The server uses this feedback and emotion data to update User A's profile to provide more appropriate information next time.

[0904] This allows users to be exposed to new perspectives and knowledge, and also provides them with the most appropriate information based on their emotions.

[0905] The processing flow will be explained below.

[0906] Step 1:

[0907] The server collects user access history, search history, bookmark data, and other interest data, including the websites the user has visited, the keywords they have searched for, the articles they have viewed, the pages they have bookmarked, etc. The collected data is stored in a database for each individual user.

[0908] Step 2:

[0909] The server uses natural language processing models to extract key keywords from the collected user data. For example, if a user frequently views articles about "technology," "artificial intelligence," and "programming," these keywords will be extracted. The extracted keywords are then mapped based on their relevance to generate a keyword map for each user.

[0910] Step 3:

[0911] The server analyzes the generated keyword map and eliminates central keywords, including the user's primary interests: "technology," "artificial intelligence," and "programming." It then extracts peripheral keywords (e.g., "biology," "art," and "psychology"). This "reverse" filtering generates a new keyword list.

[0912] Step 4:

[0913] The server searches and collects relevant information from the Internet based on the new keywords extracted by "reverse" filtering. It searches for articles, news, research papers, etc. related to the new keywords to collect appropriate information. The collected information is stored in a database.

[0914] Step 5:

[0915] The server uses artificial intelligence models to generate thumbnail images related to the new information collected. For example, for an article related to "biology," the AI ​​might generate a diagram of DNA or an image of a cell. The generated thumbnail images are then associated with the new information and stored.

[0916] Step 6:

[0917] The server uses an emotion engine to recognize emotions based on user input data (e.g., feedback and voice input). The emotion engine inputs the user's ratings and feedback into an emotion analysis algorithm to analyze the user's emotional state (e.g., happy, interesting, bored, etc.). This emotion data is added to the user's profile.

[0918] Step 7:

[0919] The terminal displays new information and thumbnail images sent from the server to the user. Through the user interface, the user can access new information and view related thumbnail images, which attracts the user's interest and encourages access to the information.

[0920] Step 8:

[0921] Users rate the new information and thumbnails provided and provide feedback, including interest and relevance of the information, visual opinion, and emotional state (e.g., enjoyable, interesting, boring, etc.).

[0922] Step 9:

[0923] The device sends feedback data from the user to the server, which adds the collected feedback data to a profile for each user.

[0924] Step 10:

[0925] The server analyzes the feedback data and updates the user's profile, reflecting the user's new trends and interests and optimizing the selection process for the next offering. The analysis results are used for future "reverse" filtering and information gathering processes.

[0926] Specific examples

[0927] As a specific example, we will explain the case where User A's areas of interest are "technology," "artificial intelligence," and "programming."

[0928] Step 1:

[0929] The server collects user A's past access history (e.g., technews.com), search history (e.g., "latest AI algorithm"), bookmarks (e.g., programmingblog.com), etc. and stores them in a database.

[0930] Step 2:

[0931] The server uses a natural language processing model to extract key keywords such as "technology," "artificial intelligence," and "programming" from the collected data, and generates a keyword map based on this.

[0932] Step 3:

[0933] The server removes the central keywords "technology," "artificial intelligence," and "programming" from the generated keyword map and extracts peripheral keywords such as "biology," "art," and "psychology."

[0934] Step 4:

[0935] The server searches the Internet for relevant articles and news based on the extracted new keywords, collects the necessary information, and stores it in a database.

[0936] Step 5:

[0937] The server uses artificial intelligence models to generate thumbnail images related to the new information collected. For example, for an article related to "biology," the AI ​​model generates a diagram of DNA or an image of a cell, which is then stored along with the new information.

[0938] Step 6:

[0939] The server analyzes emotions using an emotion engine based on user feedback and voice input. User A's emotions when viewing information (fun, interesting, bored, etc.) are input into the emotion analysis algorithm, and the analysis results are added to the user profile.

[0940] Step 7:

[0941] The terminal displays the new information and thumbnail image sent from the server and provides them to User A. User A views the new information and thumbnail image.

[0942] Step 8:

[0943] User A evaluates the new information and thumbnails provided and enters feedback (interesting / uninteresting, useful / unhelpful, etc.).

[0944] Step 9:

[0945] The device sends feedback data from user A to the server, which adds this feedback data to the profile for each user A.

[0946] Step 10:

[0947] The server analyzes the feedback data and updates the profile of User A. Based on the feedback and sentiment data, the server optimizes the selection process for the next information to be provided and utilizes it in the selection of new keywords and information gathering process.

[0948] As a result, User A is exposed to new perspectives and knowledge, and is provided with information that is optimally tailored to their emotions.

[0949] Example 2

[0950] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0951] Modern information systems mainly present information based on the user's existing interests, limiting opportunities for users to discover new interests and knowledge. Furthermore, information is often provided in a uniform manner without considering the user's emotional state. This makes it difficult to capture the user's attention, resulting in low information receptivity.

[0952] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting user interest data, means for generating a keyword map based on the interest data, a "reverse" filtering means for excluding keywords located at the center of the keyword map and extracting new keywords from peripheral keywords, means for searching and collecting new information based on the new keywords, means for generating visual content related to the new information, means for providing the new information and the visual content to the user, means for collecting feedback and emotional state from the user, and means for updating a user profile based on the feedback and emotional data. This allows the user to discover new information that they have not encountered before and provides optimal information according to their emotional state.

[0953] "User interest data" is a general term for behavioral history that indicates a user's interests and concerns, such as websites visited by the user, keywords searched for, articles viewed, and pages bookmarked.

[0954] A "keyword map" is a visual or digital mapping of key keywords extracted from user interest data and their relationships.

[0955] "Inverse filtering" is the process of eliminating the main keywords located in the center of the keyword map and extracting new keywords located around them.

[0956] "New information" is related information collected from the Internet and databases based on new keywords extracted by reverse filtering.

[0957] "Visual content" refers to images and visual materials related to new information generated using artificial intelligence models, etc.

[0958] "Feedback" refers to the evaluations and impressions users make of new information and visual content.

[0959] "Emotional state" refers to the user's emotional response or state, as analyzed from user feedback, voice input, etc.

[0960] The embodiments of the present invention will be described in detail. In this invention, new information that the user would not normally come across is provided based on the user's interest data. Also, a system is constructed that recognizes the user's emotions and optimizes feedback.

[0961] The server first collects interest data, such as the websites visited by the user, keywords searched, articles viewed, pages bookmarked, etc. This data is collected using technologies such as cookies and local storage data from the browser, and the collected data is stored in a database for each individual user.

[0962] The server then uses the collected user data to extract key keywords using a natural language processing (NLP) model, such as BERT or GPT, and maps the extracted keywords based on their relevance to generate a keyword map for each user.

[0963] In the process of analyzing the generated keyword map, the server will eliminate the main keywords located in the center and extract new keywords located in the periphery. This "reverse filtering" process will reveal keywords in areas that users are not usually interested in.

[0964] The server then searches and collects relevant information from the Internet based on the new keywords extracted by reverse filtering, using Google Search API and News API to retrieve articles and research papers related to the keywords and store them in a database.

[0965] The server then uses artificial intelligence models such as DALL-E and MidJourney to generate visual content related to the collected information: for example, articles related to "biology" will generate models of DNA and microscopic images of cells.

[0966] The terminal displays new information and visual content sent from the server to the user, and through a web browser or application interface, the user can access the new information and view the associated visual content.

[0967] Users provide feedback on the new information and visual content provided, including ratings of interest and relevance of the information. The device sends the feedback data to a server, where it is added to each user's profile.

[0968] The server uses an emotion engine to analyze emotions from user feedback and voice input. The feedback text is fed into the emotion analysis algorithm to determine the user's emotional state. This emotion data is also added to the user profile.

[0969] Finally, the server analyzes the feedback and sentiment data and updates the user's profile to reflect new trends and interests and optimize the next round of information delivery.

[0970] Specific examples

[0971] For example, if User A's areas of interest are "technology," "artificial intelligence," and "programming," these interest data are collected and key keywords are extracted using the NLP model. New keywords such as "biology," "art," and "psychology" are selected through reverse filtering, and new information is collected based on these keywords. Visual thumbnail images of this new information are generated by the AI ​​model and provided to User A.

[0972] Prompt Sentence Examples

[0973] "Suggest relevant articles and visual content based on the user's emerging areas of interest."

[0974] "Generate thumbnail images of new information based on reverse-filtered keywords."

[0975] "Use user feedback and sentiment data to suggest ways to optimize your next offering."

[0976] This allows users to access new perspectives and knowledge, and provides optimal information tailored to their emotions.

[0977] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0978] Step 1: Collecting User Data (Server)

[0979] The server collects behavioral data from the user's browser, such as the websites visited, keywords searched, articles viewed, and pages bookmarked, and uses the browser's cookies and local storage to obtain the data and store it in a database for each individual user.

[0980] Input: User behavior history data (website URL, search keywords, viewed articles, bookmark data)

[0981] Output: Database entries per user

[0982] Step 2: Generate Keyword Mapping (Server)

[0983] The server extracts key keywords from the collected user data using natural language processing (NLP) models such as BERT and GPT, which analyze and extract important keywords from text data, and then generates a keyword map based on the relevance of the keywords.

[0984] Input: User behavior history data

[0985] Output: Keyword map per user

[0986] Step 3: Performing "reverse" filtering (server)

[0987] The server analyzes the generated keyword map, eliminates the main keywords located in the center, and extracts new keywords located around these eliminated keywords, thereby obtaining keywords in areas that users are not usually interested in.

[0988] Input: User-specific keyword map

[0989] Output: New keyword list

[0990] Step 4: Selecting new information (server)

[0991] The server searches and collects relevant information from the Internet based on the new keywords extracted by reverse filtering, using Google Search API and News API to retrieve articles, research papers, etc. related to the keywords.

[0992] Input: New keyword list

[0993] Output: New information collected (articles, news, research papers, etc.)

[0994] Step 5: Generate visual content (server)

[0995] The server uses artificial intelligence models to generate visual content related to the collected information, for example, image generation AI such as DALL-E or MidJourney to generate visual content related to keywords.

[0996] Input: New information (articles, news, research papers, etc.)

[0997] Output: Relevant visual content (e.g. thumbnail image)

[0998] Step 6: Providing information and visual content (device)

[0999] The device displays new information and visual content sent from the server to the user. Through a web browser or application interface, the user can access new information and view related visual content (such as thumbnail images).

[1000] Input: New information and visual content sent from the server

[1001] Output: Information and visual content displayed in the user interface

[1002] Step 7: Collecting User Feedback (Devices and Servers)

[1003] The user inputs ratings and feedback for the new information and visual content provided, including interest in the information, relevance, visual opinion, and emotional state. The device then transmits the feedback data to the server.

[1004] Input: User feedback data

[1005] Output: Feedback data sent to the server

[1006] Step 8: Integrating Emotion Recognition (Server)

[1007] The server uses an emotion engine to analyze emotions from the user's feedback and voice input, and uses an emotion analysis algorithm to determine the user's emotional state from the feedback text.

[1008] Input: User feedback and voice input data

[1009] Output: Parsed user sentiment data

[1010] Step 9: Update Profile (Server)

[1011] The server analyzes the feedback and sentiment data and updates the user profile, so that new trends and interests are reflected in the profile and the next information delivery is optimized.

[1012] Input: User feedback data, sentiment data

[1013] Output: Updated user profile

[1014] The above are the specific processing steps of this system.

[1015] (Application example 2)

[1016] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1017] Conventional information delivery systems tend to only provide information that users are already interested in, limiting opportunities to encounter new perspectives and knowledge. Furthermore, it is difficult to provide optimal information based on the user's interests and emotions, resulting in a lack of improvement in the user experience. Furthermore, the lack of feedback optimization incorporating emotion recognition can reduce the accuracy and relevance of information provided.

[1018] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1019] In this invention, the server includes means for collecting user interest data, means for generating a keyword map based on the interest data, a "reverse" filtering means for excluding keywords located at the center of the keyword map and extracting new keywords from peripheral keywords, means for collecting new information based on the new keywords, means for generating thumbnail images related to the new information, means for providing the new information and the thumbnail images to the user, means for collecting feedback from the user, emotion recognition means for analyzing the user's emotions based on the feedback, and means for optimizing the next information to be provided based on the analyzed emotions. This allows the user to be exposed to new perspectives and knowledge and to be provided with optimal information according to their emotions.

[1020] "User interest data" refers to data that reflects a user's interests, such as user access history, search history, and bookmark data.

[1021] A "keyword map" is a map showing key keyword relationships generated using natural language processing models based on user interest data.

[1022] "Inverse filtering" is a technique that eliminates the main keywords located at the center of the keyword map and extracts new keywords from the periphery.

[1023] "New information" is information related to new perspectives and knowledge that is collected based on new keywords extracted by reverse filtering.

[1024] A "thumbnail image" is a small visual image associated with the new information collected.

[1025] "User feedback" refers to users' ratings and impressions of the provided information and thumbnail images.

[1026] An "emotion recognition means" is an engine or algorithm that analyzes emotions based on user feedback and understands their emotional state.

[1027] "Information optimization means" refers to the means of adjusting and optimizing the content of the next information provided based on the analyzed user sentiment.

[1028] A "natural language processing model" is an algorithm or machine learning model for extracting and analyzing meaning and keywords from text data.

[1029] An "artificial intelligence model" is an algorithm or system that learns from large amounts of data and automates specific tasks.

[1030] The present invention provides a technology for constructing a system that provides new information that a user would not normally come across, and further recognizes the user's emotions and optimizes feedback. Specific embodiments of the present invention will be described below.

[1031] System configuration

[1032] This system consists of a server that collects user interest data, a server that generates a keyword map based on the interest data, a server that performs "reverse" filtering to eliminate major keywords from the generated keyword map and extract new keywords, a server that collects new information and generates thumbnail images, a terminal that provides information and thumbnail images to users, a server that collects user feedback and performs emotion recognition, and a server that optimizes the information to be provided next time.

[1033] Collection and Creation

[1034] The server collects interest data such as user access history, search history, and bookmark data, including the websites the user has visited, the keywords they have searched for, the articles they have viewed, the pages they have bookmarked, etc. The collected data is stored in a database for each user.

[1035] The server then uses natural language processing models to extract key keywords from the collected user data. For example, if a user frequently views articles about "technology," "artificial intelligence," and "programming," these keywords will be extracted. The extracted keywords are then mapped based on their relevance to generate a keyword map for each user.

[1036] Reverse filtering and new information gathering

[1037] The generated keyword map is analyzed and the main keywords located in the center are eliminated. This elimination includes the user's main interests, such as "technology," "artificial intelligence," and "programming." New keywords located in the periphery (e.g., "biology," "art," and "psychology") are then extracted. This "reverse" filtering generates a new keyword list.

[1038] The server searches and collects relevant information from the Internet or specific information sources based on the new keywords extracted by "reverse" filtering. It selects articles, news, research papers, etc. related to the new keywords from the search results and collects appropriate information. The collected information is stored in a database.

[1039] Thumbnail generation and serving

[1040] The server uses artificial intelligence models to generate thumbnail images related to the new information collected. For example, for an article related to "biology," the generated thumbnail image might be a diagram of DNA or an image of a cell. The generated thumbnail image is associated with the new information and stored.

[1041] The terminal displays the new information and thumbnail images sent from the server to the user. Through the user interface, the user can access the new information and view the related thumbnail images.

[1042] Feedback collection and emotion recognition

[1043] The user evaluates the provided new information and thumbnails and enters feedback, including interest, relevance, visual evaluation, and emotional state (e.g., fun, interesting, boring, etc.). The device then transmits this feedback data to the server.

[1044] The server uses an emotion engine to recognize emotions based on user input data. The emotion engine inputs user ratings and feedback into an emotion analysis algorithm to analyze the user's emotional state. This emotion data is added to the user's profile.

[1045] Optimizing next advertisement

[1046] The server analyzes the feedback data and sentiment data and updates the user's profile to reflect the user's new trends and interests and optimize the next information offering.

[1047] Examples of specific examples and prompts

[1048] As a concrete example, assume that User A is interested in "technology," "artificial intelligence," and "programming." This user's interest data is collected, and key keywords are extracted using a natural language processing model. New keywords such as "biology," "art," and "psychology" are selected using a "reverse" filtering method, and new information is collected based on these keywords. Visual thumbnail images of this new information are generated by AI and provided to User A. User A views the new information and thumbnail images and provides feedback on whether or not they are interested. Furthermore, an emotion engine analyzes emotions based on User A's feedback and voice input, and grasps the user's emotional state. The server uses this feedback and emotion data to update User A's profile to provide more appropriate information next time.

[1049] An example prompt might be, "The user is interested in technology, artificial intelligence, and programming. After reverse filtering, the keywords are biology, art, and psychology. Based on these keywords, please find and provide relevant, up-to-date articles and information."

[1050] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1051] Step 1:

[1052] Server collection of user interest data

[1053] The server collects interest data such as user access history, search history, and bookmark data, including the websites the user visits, keywords searched, articles viewed, and pages bookmarked. The collected data is stored in a database as individual user profiles.

[1054] Input: Access history, search history, bookmark data

[1055] Output: Save interest data in user profile

[1056] Step 2:

[1057] Server-generated keyword mappings

[1058] The server extracts key keywords from the collected interest data using natural language processing models (e.g., TF-IDF-based vectorization and clustering). For example, if a user frequently views articles related to "technology," "artificial intelligence," and "programming," these keywords will be extracted. The extracted keywords are then mapped to a keyword map based on their relevance.

[1059] Input: Interest data (user profile)

[1060] Output: Keyword map (including a list of primary keywords)

[1061] Step 3:

[1062] Server performs "reverse" filtering

[1063] The server analyzes the generated keyword map and eliminates central keywords, including the user's main interests: "technology," "artificial intelligence," and "programming." It then extracts peripheral new keywords (e.g., "biology," "art," and "psychology").

[1064] Input: Keyword Map

[1065] Output: New "reverse" filtered keyword list

[1066] Step 4:

[1067] Server selection and collection of new information

[1068] The server searches and collects relevant information from the Internet and specific information sources based on the new keywords extracted through "reverse" filtering, and stores the information, such as articles, news, and research papers, related to the new keywords in a database.

[1069] Input: New "reverse" filtered keyword list

[1070] Output: New information (articles, news, research papers, etc.)

[1071] Step 5:

[1072] Server-generated thumbnail images

[1073] The server uses artificial intelligence models (e.g., image generation AI) to generate thumbnail images related to the new information collected. For example, for an article related to "biology," the generated thumbnail images might be diagrams of DNA or images of cells.

[1074] Input: New information

[1075] Output: Thumbnail image

[1076] Step 6:

[1077] Device provides information and thumbnail images

[1078] The terminal displays the new information and thumbnail images sent from the server to the user, and through the user interface, the user can access the new information and visually check the related thumbnail images.

[1079] Input: New information, thumbnail image

[1080] Output: User interface display (new information and thumbnail image)

[1081] Step 7:

[1082] Collecting and entering user feedback

[1083] Users input feedback about the new information and thumbnail images provided, such as interest, relevance, rating, visual opinion, and emotional state, which is then sent to the server via the device.

[1084] Input: User feedback on new information and thumbnail images

[1085] Output: Feedback data (ratings, emotional state, etc.)

[1086] Step 8:

[1087] Emotion recognition and analysis by the server

[1088] The server uses an emotion recognition engine to analyze the user's emotional state based on the feedback data received from the user, and the analysis results are added to the user's profile.

[1089] Input: Feedback data

[1090] Output: Emotion data (fun, interesting, bored, etc.)

[1091] Step 9:

[1092] Server optimization of next information provision

[1093] The server analyzes the feedback and sentiment data and updates the user's profile to reflect new trends and interests and optimize the content of the next information provided.

[1094] Input: Feedback data, emotion data

[1095] Output: Updated user profile, optimization plan for next information provided

[1096] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1097] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1098] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1099] [Fourth embodiment]

[1100] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1101] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1102] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1103] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1104] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1105] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1106] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1107] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1108] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1109] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1110] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1111] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1112] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1113] The embodiments for carrying out the present invention will be described in detail below.

[1114] User data collection (server)

[1115] The server collects users' access history, search history, bookmark data, etc. This includes the websites they have visited, the keywords they have searched for, the articles they have viewed, the pages they have bookmarked, etc. The collected data is stored in a database for each individual user and used for later analysis.

[1116] Generate Keyword Mapping (Server)

[1117] The server uses a natural language processing model to extract key keywords from the collected user data. For example, if a user frequently views articles related to "technology," "artificial intelligence," and "programming," these keywords are extracted. The extracted keywords are then mapped based on their relevance to generate a keyword map for each user. This keyword map has a graph structure that displays keywords near the center as the user's main interests and places less relevant keywords on the periphery.

[1118] "Reverse" filtering implementation (server)

[1119] The server analyzes the generated keyword map and eliminates central keywords. For example, if the user's primary interests are "technology," "artificial intelligence," and "programming," these keywords are excluded. It then extracts more peripheral keywords (e.g., "biology," "art," and "psychology"). This "reverse" filtering extracts keywords that the user would not normally be interested in, but which may provide relevant new information.

[1120] Selection of new information (server)

[1121] The server searches and collects relevant information from the Internet based on the new keywords extracted through "reverse" filtering, such as articles, news, and research papers related to the new keywords. The collected information is stored in a database and later provided to users.

[1122] Thumbnail generation (server)

[1123] The server uses artificial intelligence models to generate thumbnail images related to new information collected. For example, for articles related to "biology," the AI ​​model might generate diagrams of DNA or images of cells. The generated thumbnail images are then associated with the new information, increasing visual engagement.

[1124] Provide information and thumbnails (device)

[1125] The terminal displays new information and thumbnail images sent from the server to the user. Through the user interface, the user can access new information and view related thumbnail images, which can attract the user's interest and encourage them to access the information.

[1126] Collecting user feedback (device and server)

[1127] The user evaluates the new information and thumbnails provided and enters feedback. The device collects the user's feedback data (e.g., interesting / uninteresting, useful / unhelpful, etc.) and sends it to the server. The server analyzes this feedback and updates the user's profile. Based on the user's feedback, the selection process for the next information offering is optimized.

[1128] Specific examples

[1129] If User A's areas of interest are "technology," "artificial intelligence," and "programming," the user's interest data is collected and key keywords are extracted using a natural language processing model. New keywords such as "biology," "art," and "psychology" are selected using "reverse" filtering, and new information is collected based on these keywords. Visual thumbnail images of this new information are generated by AI and provided to User A. User A views the new information and thumbnail images and provides feedback on whether or not they are interested. The server uses this feedback to provide more appropriate information next time.

[1130] The processing flow will be explained below.

[1131] Step 1:

[1132] The server collects information about users' interests, such as their access history, search history, and bookmark data, including the websites they visit, the keywords they search for, the articles they read, and the pages they bookmark. The collected data is stored in a database for each individual user.

[1133] Step 2:

[1134] The server uses natural language processing models to extract key keywords from the collected user data. For example, if a user frequently views articles related to "technology," "artificial intelligence," and "programming," these keywords will be extracted. The extracted keywords are then mapped based on their relevance, generating a keyword map for each user.

[1135] Step 3:

[1136] The server analyzes the generated keyword map and eliminates central keywords, including the user's primary interests: "technology," "artificial intelligence," and "programming." It then extracts peripheral keywords (e.g., "biology," "art," and "psychology"). This "reverse" filtering generates a new keyword list.

[1137] Step 4:

[1138] The server searches and collects relevant information from the Internet based on the new keywords extracted by "reverse" filtering. It selects articles, news, research papers, etc. related to the new keywords from the search results and collects appropriate information. The collected information is stored in a database.

[1139] Step 5:

[1140] The server uses artificial intelligence models to generate thumbnail images related to the new information collected. For example, for an article related to "biology," the AI ​​might generate a diagram of DNA or an image of a cell. The generated thumbnail images are then associated with the new information and stored.

[1141] Step 6:

[1142] The terminal displays new information and thumbnail images sent from the server to the user. Through the user interface, the user can access new information and view related thumbnail images, which can attract the user's interest and encourage them to access the information.

[1143] Step 7:

[1144] Users rate the new information and thumbnails provided and provide feedback, including interest, relevance, and visual opinion about the information.

[1145] Step 8:

[1146] The device sends feedback data from the user to the server, which adds the collected feedback data to a profile for each user.

[1147] Step 9:

[1148] The server analyzes the feedback data and updates the user's profile, reflecting the user's new trends and interests and optimizing the selection process for the next offering. The analysis results are used for future "reverse" filtering and information gathering processes.

[1149] Example 1

[1150] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1151] Conventional information provision systems provide information tailored to users' interests, but it has been difficult to provide new information that expands users' interests. Furthermore, there is a lack of methods to attract users' attention by using visually appealing thumbnail images. Furthermore, there are also challenges in effectively collecting user feedback and optimizing the information provided based on that feedback.

[1152] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1153] In this invention, the server includes a means for collecting user information, a means for generating a keyword map based on the user information, a means for inverse filtering that excludes keywords at the center of the keyword map and extracts new keywords from peripheral keywords, a means for collecting new information based on the new keywords, a means for generating images related to the new information, a means for providing the new information and the images to the user, and a means for collecting user ratings. This allows the server to efficiently provide new information that is not of interest to the user and generate visually appealing thumbnail images related to that information, thereby attracting the user's interest. Furthermore, by collecting and analyzing user feedback, the server can provide optimized information the next time it provides information.

[1154] "User information" refers to data such as a user's access history, search history, articles viewed, and bookmarked pages.

[1155] A "keyword map" is a graph structure that arranges key keywords extracted based on user interests in a relevance manner.

[1156] "Inverse filtering" is a technique that removes the main keywords in the center of a keyword map and extracts new, relevant keywords that are located on the periphery.

[1157] "New information" refers to content such as articles, news, and research papers that are collected based on new keywords obtained through reverse filtering.

[1158] "Image" refers to a visual thumbnail image generated in association with new information.

[1159] A "generative artificial intelligence model" is an artificial intelligence model that generates new images or text based on input data provided by humans.

[1160] "Evaluation" refers to feedback data that reflects whether the user is interested in the information or images provided, whether it is useful, etc.

[1161] The present invention provides an information provision system based on a user's interest, which is implemented through the following steps.

[1162] The server first collects user information, including the history of websites the user has visited, keywords they have searched for, articles they have viewed, and pages they have bookmarked. This data is collected using, for example, Google Analytics or a dedicated tracking code. The collected data is associated with each user and stored in a database (for example, MySQL or PostgreSQL).

[1163] The server then analyzes the collected user information and extracts key keywords using a natural language processing model (e.g., Hugging Face's BERT model). This analysis may also utilize Term Frequency-Inverse Document Frequency (TF-IDF). For example, if a user frequently views articles related to "programming," "machine learning," and "data science," these keywords will be extracted as their key interests.

[1164] These extracted key keywords are then mapped based on their relevance to generate a keyword map, which has a graph structure with the key keyword at the center and related keywords around it.

[1165] Additionally, the server performs reverse filtering, which removes key keywords (e.g., "programming," "machine learning," etc.) from the center of the keyword map and extracts peripheral keywords (e.g., "psychology," "biology," etc.). This reverse filtering allows for the extraction of new, relevant keywords that are outside the user's usual interests.

[1166] The server collects new information from the Internet based on the new keywords obtained through reverse filtering. Specifically, it uses the Google News API, PubMed API, etc. to collect related articles, news, research papers, etc. This collected information is stored in a database and managed as candidates for information to be provided to users.

[1167] As new information is collected, the server uses generative AI models (such as DALL-E 2 or StyleGAN) to generate thumbnail images related to that information. For example, for articles related to "biology," it might generate diagrams of DNA or images of cells. The generated thumbnail images are then associated with the new information and presented to the user in a visually appealing format.

[1168] The device displays new information and thumbnail images sent from the server to the user. The front-end user interface uses React or Vue.js, and the user is visually captivated by the information and thumbnail images.

[1169] Finally, the user rates the provided information and thumbnail images and enters their feedback. The device collects this feedback data and sends it to the server. For example, it can be in the form of a simple rating such as "useful" or "not interesting." The server analyzes the feedback and updates the user's profile, enabling it to provide more optimized information the next time information is provided.

[1170] Specific examples

[1171] For example, if User A is interested in "technology," "artificial intelligence," and "programming," the server collects related data and uses the BERT model to extract key keywords. It then performs reverse filtering to identify new keywords, such as "biology," "art," and "psychology." Based on these keywords, related information is collected using the Google News API and PubMed API. "Thumbnail images related to biology" generated by DALL-E 2 are provided to User A as visual aids. User A provides feedback on the information provided, and the server uses that feedback to optimize the information provided.

[1172] Prompt Sentence Examples

[1173] "Extract key keywords based on user data, then use reverse filtering to gather new information related to the resulting keywords. Then generate thumbnail images related to the new information."

[1174] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1175] Step 1: Collecting User Data (Server)

[1176] The server collects data such as the history of websites visited by the user, keywords searched, articles viewed, and pages bookmarked. This data is collected using Google Analytics or a dedicated tracking code. The server receives each user's action data and uses this data to create a detailed history database for each user. The collected data is associated with each user and stored in a database (e.g., MySQL, PostgreSQL).

[1177] Step 2: Generate Keyword Mapping (Server)

[1178] The server receives the collected user information as input and uses a natural language processing model (e.g., Hugging Face's BERT model) to extract key keywords. This analysis also uses Term Frequency-Inverse Document Frequency (TF-IDF) scoring of the text data. The output is a list of keywords that indicate the user's main interests. For example, "programming," "machine learning," and "data science" may be extracted.

[1179] Step 3: Generate a Keyword Map (Server)

[1180] The server receives the extracted keywords as input and generates a keyword map based on their relevance. This keyword map has a graph structure with the main keyword at the center and related keywords around it. The output is a graph-structured keyword map.

[1181] Step 4: Performing Reverse Filtering (Server)

[1182] The server receives the generated keyword map as input, eliminates the main keywords located in the center, and extracts peripheral keywords. This process is called inverse filtering. The output is a list of new keywords that are outside the user's usual interests. For example, "psychology," "biology," and "philosophy" are extracted.

[1183] Step 5: Selecting new information (server)

[1184] The server receives the new keywords obtained through reverse filtering as input and collects new information from the Internet based on them, using APIs such as Google News and PubMed. The output is a dataset of articles, news articles, and research papers related to the new keywords. This information is then stored in a database.

[1185] Step 6: Thumbnail generation (server)

[1186] The server takes the collected new information as input and uses a generative AI model (e.g., DALL-E 2 or StyleGAN) to generate relevant thumbnail images. The output is a visually appealing thumbnail image that corresponds to the new information. For example, an article on "biology" might generate a diagram of DNA or an image of a cell.

[1187] Step 7: Provide information and thumbnail (device)

[1188] The terminal receives new information and thumbnail images provided by the server as input and displays them to the user. The front-end user interface uses React or Vue.js. The output is a visually appealing display of information provided to the user. The user can browse these displays.

[1189] Step 8: Collecting User Feedback (Devices and Servers)

[1190] Users rate the provided information and thumbnail images and enter their feedback. The input includes ratings such as "useful" or "not interesting." The device collects this feedback data and sends it to the server. The server outputs the analyzed feedback data, which is used to update the user's profile. This optimizes the next information provided.

[1191] (Application example 1)

[1192] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1193] Conventional information provision systems only provide information based on users' existing interests, making it difficult to stimulate new interests. Furthermore, the methods for providing information based on user data are limited, preventing effective provision of information that will interest users in local stores. Furthermore, there is a lack of information provision based on the user's location information, making it insufficient to improve the user experience in physical stores.

[1194] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1195] In this invention, the server includes means for collecting user interest data, means for generating a keyword map based on the interest data, means for "reverse" filtering to exclude keywords located at the center of the keyword map and extract new keywords from peripheral keywords, means for collecting new information based on the new keywords, means for generating thumbnail images related to the new information, means for providing the new information and the thumbnail images to the user, means for collecting feedback from the user, means for recognizing the user within the terminal, and means for the terminal to provide information based on the user's location, thereby enabling the server to elicit new interests beyond the user's existing interests and improve the in-store experience.

[1196] "User Interest Data" is data about a user's interests, such as the user's website visit history, search history, and bookmarks.

[1197] A "keyword map" is a graph structure that associates key keywords based on user interest data, with the main interest at the center and less relevant keywords at the periphery.

[1198] "Inverse filtering" is a technique that eliminates the main keywords located at the center of the keyword map and extracts new keywords from the remaining peripheral keywords.

[1199] A "thumbnail image" is a small visual image generated in association with new information collected and used to attract user attention.

[1200] "Device" refers to an electronic device used by a user to receive information, including smartphones, tablets, and personal computers.

[1201] "Means of recognizing the user" refers to technology that enables the device to detect and identify the user's presence and location within the store, and uses beacons, Wi-Fi, etc.

[1202] User Data Collection

[1203] The server collects user interest data, including user website visit history, search history, and bookmark data, which is stored in a database for each user and used for later analysis.

[1204] Generate Keyword Mapping

[1205] The server uses a natural language processing model to extract key keywords from the collected user data. For example, if a user frequently views articles related to "technology," "artificial intelligence," and "programming," these keywords are extracted. The extracted keywords are then mapped based on their relevance to generate a keyword map for each user. This keyword map has a graph structure that displays keywords near the center as the user's main interests and places less relevant keywords on the periphery.

[1206] Implementing "reverse" filtering

[1207] The server analyzes the generated keyword map and eliminates central keywords. For example, if the user's primary interests are "technology," "artificial intelligence," and "programming," these keywords are excluded. It then extracts more peripheral keywords (e.g., "biology," "art," and "psychology"). This "reverse" filtering extracts keywords that the user would not normally be interested in, but which may provide relevant new information.

[1208] Selection of new information

[1209] The server searches and collects relevant information from the Internet based on the new keywords extracted through "reverse" filtering, such as articles, news, and research papers related to the new keywords. The collected information is stored in a database and later provided to users.

[1210] Thumbnail Generation

[1211] The server uses artificial intelligence models to generate thumbnail images related to new information collected. For example, for articles related to "biology," the AI ​​model might generate diagrams of DNA or images of cells. The generated thumbnail images are then associated with the new information, increasing visual engagement.

[1212] Providing information and thumbnails

[1213] The device displays new information and thumbnail images sent from the server to the user. Through the user interface, the user can access new information and view related thumbnail images, which can attract the user's interest and encourage access to the information. It also recognizes the user within the device and provides more relevant information based on the user's location.

[1214] Collecting user feedback

[1215] The user evaluates the new information and thumbnails provided and enters feedback. The device collects the user's feedback data (e.g., interesting / uninteresting, useful / unhelpful, etc.) and sends it to the server. The server analyzes this feedback and updates the user's profile. Based on the user's feedback, the selection process for the next information offering is optimized.

[1216] Specific examples

[1217] If User A's areas of interest are "technology," "artificial intelligence," and "programming," the user's interest data is collected and key keywords are extracted using a natural language processing model. New keywords such as "biology," "art," and "psychology" are selected using "reverse" filtering, and new information is collected based on these keywords. Visual thumbnail images of this new information are generated by AI and provided to User A. User A views the new information and thumbnail images and provides feedback on whether or not they are interested. The server uses this feedback to provide more appropriate information next time.

[1218] Prompt Sentence Examples

[1219] For users with an interest in "Technology," suggest recent research articles in "Biology" or "Psychology" that are relevant but may spark new interest. Also generate visual thumbnails associated with each article.

[1220] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1221] Step 1:

[1222] The server collects user website visit history, search history, and bookmark data. This data is stored in a database for each user and converted into a format for analysis. The input is the user's online activity history, and the output is formatted user data.

[1223] Step 2:

[1224] The server uses a natural language processing model to extract key keywords from the formatted user data. The input is the user data organized in the previous step, and the output is a list of key keywords. Specifically, the data is input into the natural language processing algorithm to extract frequently occurring keywords.

[1225] Step 3:

[1226] The server visualizes the extracted keywords as a keyword map based on their relevance. The input is a list of key keywords, and the output is a graph-structured map of the keywords. Specifically, it calculates the relevance between keywords and generates the map using a graph library.

[1227] Step 4:

[1228] The server analyzes the generated keyword map, eliminates the main keywords located in the center, and extracts new keywords from the periphery using a "reverse" filtering method. The input is the keyword map, and the output is a list of new keywords. Specifically, the server reduces the weight of the central node and selects new keywords from the periphery.

[1229] Step 5:

[1230] The server searches and collects relevant information from the Internet based on the new keywords extracted by "reverse" filtering. It takes a list of new keywords as input and a list of new information as output. Specifically, it uses a search API to query related information on the web and organizes the results.

[1231] Step 6:

[1232] The server uses an artificial intelligence model to generate thumbnail images related to the new information collected. The input is a list of new information, and the output is a list of thumbnail images. Specifically, the content of the information is input into the AI ​​model, which generates related visual thumbnails.

[1233] Step 7:

[1234] The terminal displays the new information and thumbnail image sent from the server to the user. The input is the new information and thumbnail image, and the output is the information and image displayed on the user's display. The specific operation is to arrange the information and image in the terminal's user interface.

[1235] Step 8:

[1236] Users rate the provided new information and thumbnails and enter their feedback. The input is the user's rating data, and the output is the feedback data. The specific operation is for users to enter their interest and rating through the interface.

[1237] Step 9:

[1238] The terminal collects feedback data from the user and sends it to the server. The input is the user's feedback data, and the output is the feedback data sent to the server. Specifically, the terminal uses the sending function to upload the feedback data to the server.

[1239] Step 10:

[1240] The server analyzes the feedback data from the user and updates the user profile. The input is the feedback data sent from the device, and the output is the updated user profile. Specifically, the server reevaluates the user's interest trends based on the feedback data and optimizes the next information provision.

[1241] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1242] The following describes in detail the mode for carrying out the present invention. In the present invention, a system is constructed that provides new information that the user would not normally come across based on the user's interest data, and further recognizes the user's emotions to optimize the feedback.

[1243] User data collection (server)

[1244] The server collects user access history, search history, bookmark data, and other interest data, including the websites the user has visited, the keywords they have searched for, the articles they have viewed, the pages they have bookmarked, etc. The collected data is stored in a database for each individual user.

[1245] Generate Keyword Mapping (Server)

[1246] The server uses natural language processing models to extract key keywords from the collected user data. For example, if a user frequently views articles about "technology," "artificial intelligence," and "programming," these keywords will be extracted. The extracted keywords are then mapped based on their relevance to generate a keyword map for each user.

[1247] "Reverse" filtering implementation (server)

[1248] The server analyzes the generated keyword map and eliminates central keywords, including the user's primary interests: "technology," "artificial intelligence," and "programming." It then extracts peripheral keywords (e.g., "biology," "art," and "psychology"). This "reverse" filtering generates a new keyword list.

[1249] Selection of new information (server)

[1250] The server searches and collects relevant information from the Internet based on the new keywords extracted by "reverse" filtering. It selects articles, news, research papers, etc. related to the new keywords from the search results and collects appropriate information. The collected information is stored in a database.

[1251] Thumbnail generation (server)

[1252] The server uses artificial intelligence models to generate thumbnail images related to the new information collected. For example, for an article related to "biology," the AI ​​might generate a diagram of DNA or an image of a cell. The generated thumbnail images are then associated with the new information and stored.

[1253] Incorporating emotion recognition (server)

[1254] The server uses an emotion engine to recognize emotions based on user input data (e.g., feedback and voice input). The emotion engine inputs the user's ratings and feedback into an emotion analysis algorithm to analyze the user's emotional state (e.g., happy, interesting, bored, etc.). This emotion data is added to the user's profile.

[1255] Provide information and thumbnails (device)

[1256] The terminal displays new information and thumbnail images sent from the server to the user. Through the user interface, the user can access new information and view related thumbnail images, which can attract the user's interest and encourage them to access the information.

[1257] Collecting user feedback (device and server)

[1258] Users rate the new information and thumbnails provided and provide feedback, including interest, relevance, visual opinion, and emotional state (e.g., enjoyable, interesting, boring, etc.).

[1259] The device sends feedback data from the user to the server, which adds the collected feedback data to a profile for each user.

[1260] The server analyzes the feedback data and updates the user's profile, reflecting the user's new trends and interests and optimizing the selection process for the next offering. The analysis results are used for future "reverse" filtering and information gathering processes.

[1261] Specific examples

[1262] If User A's areas of interest are "technology," "artificial intelligence," and "programming," the user's interest data is collected and key keywords are extracted using a natural language processing model. New keywords such as "biology," "art," and "psychology" are selected using "reverse" filtering, and new information is collected based on these keywords. Visual thumbnail images of this new information are generated by AI and provided to User A.

[1263] User A browses the new information and thumbnail images and provides feedback on whether or not they are interested. The emotion engine then analyzes User A's emotions based on their feedback and voice input to understand their emotional state. The server uses this feedback and emotion data to update User A's profile to provide more appropriate information next time.

[1264] This allows users to be exposed to new perspectives and knowledge, and also provides them with the most appropriate information based on their emotions.

[1265] The processing flow will be explained below.

[1266] Step 1:

[1267] The server collects user access history, search history, bookmark data, and other interest data, including the websites the user has visited, the keywords they have searched for, the articles they have viewed, the pages they have bookmarked, etc. The collected data is stored in a database for each individual user.

[1268] Step 2:

[1269] The server uses natural language processing models to extract key keywords from the collected user data. For example, if a user frequently views articles about "technology," "artificial intelligence," and "programming," these keywords will be extracted. The extracted keywords are then mapped based on their relevance to generate a keyword map for each user.

[1270] Step 3:

[1271] The server analyzes the generated keyword map and eliminates central keywords, including the user's primary interests: "technology," "artificial intelligence," and "programming." It then extracts peripheral keywords (e.g., "biology," "art," and "psychology"). This "reverse" filtering generates a new keyword list.

[1272] Step 4:

[1273] The server searches and collects relevant information from the Internet based on the new keywords extracted by "reverse" filtering. It searches for articles, news, research papers, etc. related to the new keywords to collect appropriate information. The collected information is stored in a database.

[1274] Step 5:

[1275] The server uses artificial intelligence models to generate thumbnail images related to the new information collected. For example, for an article related to "biology," the AI ​​might generate a diagram of DNA or an image of a cell. The generated thumbnail images are then associated with the new information and stored.

[1276] Step 6:

[1277] The server uses an emotion engine to recognize emotions based on user input data (e.g., feedback and voice input). The emotion engine inputs the user's ratings and feedback into an emotion analysis algorithm to analyze the user's emotional state (e.g., happy, interesting, bored, etc.). This emotion data is added to the user's profile.

[1278] Step 7:

[1279] The terminal displays new information and thumbnail images sent from the server to the user. Through the user interface, the user can access new information and view related thumbnail images, which can attract the user's interest and encourage them to access the information.

[1280] Step 8:

[1281] Users rate the new information and thumbnails provided and provide feedback, including interest and relevance of the information, visual opinion, and emotional state (e.g., enjoyable, interesting, boring, etc.).

[1282] Step 9:

[1283] The device sends feedback data from the user to the server, which adds the collected feedback data to a profile for each user.

[1284] Step 10:

[1285] The server analyzes the feedback data and updates the user's profile, reflecting the user's new trends and interests and optimizing the selection process for the next offering. The analysis results are used for future "reverse" filtering and information gathering processes.

[1286] Specific examples

[1287] As a specific example, we will explain the case where User A's areas of interest are "technology," "artificial intelligence," and "programming."

[1288] Step 1:

[1289] The server collects user A's past access history (e.g., technews.com), search history (e.g., "latest AI algorithm"), bookmarks (e.g., programmingblog.com), etc. and stores them in a database.

[1290] Step 2:

[1291] The server uses a natural language processing model to extract key keywords such as "technology," "artificial intelligence," and "programming" from the collected data, and generates a keyword map based on this.

[1292] Step 3:

[1293] The server removes the central keywords "technology," "artificial intelligence," and "programming" from the generated keyword map and extracts peripheral keywords such as "biology," "art," and "psychology."

[1294] Step 4:

[1295] The server searches the Internet for relevant articles and news based on the extracted new keywords, collects the necessary information, and stores it in a database.

[1296] Step 5:

[1297] The server uses artificial intelligence models to generate thumbnail images related to the new information collected. For example, for an article related to "biology," the AI ​​model generates a diagram of DNA or an image of a cell, which is then stored along with the new information.

[1298] Step 6:

[1299] The server analyzes emotions using an emotion engine based on user feedback and voice input. User A's emotions when viewing information (fun, interesting, bored, etc.) are input into the emotion analysis algorithm, and the analysis results are added to the user profile.

[1300] Step 7:

[1301] The terminal displays the new information and thumbnail image sent from the server and provides them to User A. User A views the new information and thumbnail image.

[1302] Step 8:

[1303] User A evaluates the new information and thumbnails provided and enters feedback (interesting / uninteresting, useful / unhelpful, etc.).

[1304] Step 9:

[1305] The device sends feedback data from user A to the server, which adds this feedback data to the profile for each user A.

[1306] Step 10:

[1307] The server analyzes the feedback data and updates the profile of User A. Based on the feedback and sentiment data, the server optimizes the selection process for the next information to be provided and utilizes it in the selection of new keywords and information gathering process.

[1308] As a result, User A is exposed to new perspectives and knowledge, and is provided with information that is optimally tailored to their emotions.

[1309] Example 2

[1310] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1311] Modern information systems mainly present information based on the user's existing interests, limiting opportunities for users to discover new interests and knowledge. Furthermore, information is often provided in a uniform manner without considering the user's emotional state. This makes it difficult to capture the user's attention, resulting in low information receptivity.

[1312] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting user interest data, means for generating a keyword map based on the interest data, a "reverse" filtering means for excluding keywords located at the center of the keyword map and extracting new keywords from peripheral keywords, means for searching and collecting new information based on the new keywords, means for generating visual content related to the new information, means for providing the new information and the visual content to the user, means for collecting feedback and emotional state from the user, and means for updating a user profile based on the feedback and emotional data. This allows the user to discover new information that they have not encountered before and provides optimal information according to their emotional state.

[1313] "User interest data" is a general term for behavioral history that indicates a user's interests and concerns, such as websites visited by the user, keywords searched for, articles viewed, and pages bookmarked.

[1314] A "keyword map" is a visual or digital mapping of key keywords extracted from user interest data and their relationships.

[1315] "Inverse filtering" is the process of eliminating the main keywords located in the center of the keyword map and extracting new keywords located around them.

[1316] "New information" is related information collected from the Internet and databases based on new keywords extracted by reverse filtering.

[1317] "Visual content" refers to images and visual materials related to new information generated using artificial intelligence models, etc.

[1318] "Feedback" refers to the evaluations and impressions users make of new information and visual content.

[1319] "Emotional state" refers to the user's emotional response or state, as analyzed from user feedback, voice input, etc.

[1320] The embodiments of the present invention will be described in detail. In this invention, new information that the user would not normally come across is provided based on the user's interest data. Also, a system is constructed that recognizes the user's emotions and optimizes feedback.

[1321] The server first collects interest data, such as the websites visited by the user, keywords searched, articles viewed, pages bookmarked, etc. This data is collected using technologies such as cookies and local storage data from the browser, and the collected data is stored in a database for each individual user.

[1322] The server then uses the collected user data to extract key keywords using a natural language processing (NLP) model, such as BERT or GPT, and maps the extracted keywords based on their relevance to generate a keyword map for each user.

[1323] In the process of analyzing the generated keyword map, the server will eliminate the main keywords located in the center and extract new keywords located in the periphery. This "reverse filtering" process will reveal keywords in areas that users are not usually interested in.

[1324] The server then searches and collects relevant information from the Internet based on the new keywords extracted by reverse filtering, using Google Search API and News API to retrieve articles and research papers related to the keywords and store them in a database.

[1325] The server then uses artificial intelligence models such as DALL-E and MidJourney to generate visual content related to the collected information: for example, articles related to "biology" will generate models of DNA and microscopic images of cells.

[1326] The terminal displays new information and visual content sent from the server to the user, and through a web browser or application interface, the user can access the new information and view the associated visual content.

[1327] Users provide feedback on the new information and visual content provided, including ratings of interest and relevance of the information. The device sends the feedback data to a server, where it is added to each user's profile.

[1328] The server uses an emotion engine to analyze emotions from user feedback and voice input. The feedback text is fed into the emotion analysis algorithm to determine the user's emotional state. This emotion data is also added to the user profile.

[1329] Finally, the server analyzes the feedback and sentiment data and updates the user's profile to reflect new trends and interests and optimize the next round of information delivery.

[1330] Specific examples

[1331] For example, if User A's areas of interest are "technology," "artificial intelligence," and "programming," these interest data are collected and key keywords are extracted using the NLP model. New keywords such as "biology," "art," and "psychology" are selected through reverse filtering, and new information is collected based on these keywords. Visual thumbnail images of this new information are generated by the AI ​​model and provided to User A.

[1332] Prompt Sentence Examples

[1333] "Suggest relevant articles and visual content based on the user's emerging areas of interest."

[1334] "Generate thumbnail images of new information based on reverse-filtered keywords."

[1335] "Use user feedback and sentiment data to suggest ways to optimize your next offering."

[1336] This allows users to access new perspectives and knowledge, and provides optimal information tailored to their emotions.

[1337] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1338] Step 1: Collecting User Data (Server)

[1339] The server collects behavioral history data from the user's browser, such as the websites visited, keywords searched, articles viewed, and pages bookmarked, and uses the browser's cookies and local storage to obtain the data and store it in a database for each individual user.

[1340] Input: User behavior history data (website URL, search keywords, viewed articles, bookmark data)

[1341] Output: Database entries per user

[1342] Step 2: Generate Keyword Mapping (Server)

[1343] The server extracts key keywords from the collected user data using a natural language processing (NLP) model, such as BERT or GPT, to analyze and extract important keywords from the text data. It then generates a keyword map based on the relevance of the keywords.

[1344] Input: User behavior history data

[1345] Output: Keyword map per user

[1346] Step 3: Performing "reverse" filtering (server)

[1347] The server analyzes the generated keyword map, eliminates the main keywords located in the center, and extracts new keywords located around these eliminated keywords, thereby obtaining keywords in areas that users are not usually interested in.

[1348] Input: User-specific keyword map

[1349] Output: New keyword list

[1350] Step 4: Selecting new information (server)

[1351] The server searches and collects relevant information from the Internet based on the new keywords extracted by reverse filtering, using Google Search API and News API to retrieve articles, research papers, etc. related to the keywords.

[1352] Input: New keyword list

[1353] Output: New information collected (articles, news, research papers, etc.)

[1354] Step 5: Generate visual content (server)

[1355] The server uses artificial intelligence models to generate visual content related to the collected information, for example, image generation AI such as DALL-E or MidJourney to generate visual content related to keywords.

[1356] Input: New information (articles, news, research papers, etc.)

[1357] Output: Relevant visual content (e.g. thumbnail image)

[1358] Step 6: Providing information and visual content (device)

[1359] The device displays new information and visual content sent from the server to the user. Through a web browser or application interface, the user can access new information and view related visual content (such as thumbnail images).

[1360] Input: New information and visual content sent from the server

[1361] Output: Information and visual content displayed in the user interface

[1362] Step 7: Collecting User Feedback (Devices and Servers)

[1363] The user inputs ratings and feedback for the new information and visual content provided, including interest in the information, relevance, visual opinion, and emotional state. The device then transmits the feedback data to the server.

[1364] Input: User feedback data

[1365] Output: Feedback data sent to the server

[1366] Step 8: Integrating Emotion Recognition (Server)

[1367] The server uses an emotion engine to analyze emotions from the user's feedback and voice input, and uses an emotion analysis algorithm to determine the user's emotional state from the feedback text.

[1368] Input: User feedback and voice input data

[1369] Output: Parsed user sentiment data

[1370] Step 9: Update Profile (Server)

[1371] The server analyzes the feedback and sentiment data and updates the user profile, so that new trends and interests are reflected in the profile and the next information delivery is optimized.

[1372] Input: User feedback data, sentiment data

[1373] Output: Updated user profile

[1374] The above are the specific processing steps of this system.

[1375] (Application example 2)

[1376] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1377] Conventional information delivery systems tend to only provide information that users are already interested in, limiting opportunities to encounter new perspectives and knowledge. Furthermore, it is difficult to provide optimal information based on the user's interests and emotions, resulting in a lack of improvement in the user experience. Furthermore, the lack of feedback optimization incorporating emotion recognition can reduce the accuracy and relevance of information provided.

[1378] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1379] In this invention, the server includes means for collecting user interest data, means for generating a keyword map based on the interest data, a "reverse" filtering means for excluding keywords located at the center of the keyword map and extracting new keywords from peripheral keywords, means for collecting new information based on the new keywords, means for generating thumbnail images related to the new information, means for providing the new information and the thumbnail images to the user, means for collecting feedback from the user, emotion recognition means for analyzing the user's emotions based on the feedback, and means for optimizing the next information to be provided based on the analyzed emotions. This allows the user to be exposed to new perspectives and knowledge and to be provided with optimal information according to their emotions.

[1380] "User interest data" refers to data that reflects a user's interests, such as user access history, search history, and bookmark data.

[1381] A "keyword map" is a map showing key keyword relationships generated using natural language processing models based on user interest data.

[1382] "Inverse filtering" is a technique that eliminates the main keywords located at the center of the keyword map and extracts new keywords from the periphery.

[1383] "New information" is information related to new perspectives and knowledge that is collected based on new keywords extracted by reverse filtering.

[1384] A "thumbnail image" is a small visual image associated with the new information collected.

[1385] "User feedback" refers to users' ratings and impressions of the provided information and thumbnail images.

[1386] An "emotion recognition means" is an engine or algorithm that analyzes emotions based on user feedback and understands their emotional state.

[1387] "Information optimization means" refers to the means of adjusting and optimizing the content of the next information provided based on the analyzed user sentiment.

[1388] A "natural language processing model" is an algorithm or machine learning model for extracting and analyzing meaning and keywords from text data.

[1389] An "artificial intelligence model" is an algorithm or system that learns from large amounts of data and automates specific tasks.

[1390] The present invention provides a technology for constructing a system that provides new information that a user would not normally come across, and further recognizes the user's emotions and optimizes feedback. Specific embodiments of the present invention will be described below.

[1391] System configuration

[1392] This system consists of a server that collects user interest data, a server that generates a keyword map based on the interest data, a server that performs "reverse" filtering to eliminate major keywords from the generated keyword map and extract new keywords, a server that collects new information and generates thumbnail images, a terminal that provides information and thumbnail images to users, a server that collects user feedback and performs emotion recognition, and a server that optimizes the information to be provided next time.

[1393] Collection and Creation

[1394] The server collects interest data such as user access history, search history, and bookmark data, including the websites the user has visited, the keywords they have searched for, the articles they have viewed, the pages they have bookmarked, etc. The collected data is stored in a database for each user.

[1395] The server then uses natural language processing models to extract key keywords from the collected user data. For example, if a user frequently views articles about "technology," "artificial intelligence," and "programming," these keywords will be extracted. The extracted keywords are then mapped based on their relevance to generate a keyword map for each user.

[1396] Reverse filtering and new information gathering

[1397] The generated keyword map is analyzed and the main keywords located in the center are eliminated. This elimination includes the user's main interests, such as "technology," "artificial intelligence," and "programming." New keywords located in the periphery (e.g., "biology," "art," and "psychology") are then extracted. This "reverse" filtering generates a new keyword list.

[1398] The server searches and collects relevant information from the Internet or specific information sources based on the new keywords extracted by "reverse" filtering. It selects articles, news, research papers, etc. related to the new keywords from the search results and collects appropriate information. The collected information is stored in a database.

[1399] Thumbnail generation and serving

[1400] The server uses artificial intelligence models to generate thumbnail images related to the new information collected. For example, for an article related to "biology," the generated thumbnail image might be a diagram of DNA or an image of a cell. The generated thumbnail image is associated with the new information and stored.

[1401] The terminal displays the new information and thumbnail images sent from the server to the user. Through the user interface, the user can access the new information and view the related thumbnail images.

[1402] Feedback collection and emotion recognition

[1403] The user evaluates the provided new information and thumbnails and enters feedback, including interest, relevance, visual evaluation, and emotional state (e.g., fun, interesting, boring, etc.). The device then transmits this feedback data to the server.

[1404] The server uses an emotion engine to recognize emotions based on user input data. The emotion engine inputs user ratings and feedback into an emotion analysis algorithm to analyze the user's emotional state. This emotion data is added to the user's profile.

[1405] Optimizing next advertisement

[1406] The server analyzes the feedback data and sentiment data and updates the user's profile to reflect the user's new trends and interests and optimize the next information offering.

[1407] Examples of specific examples and prompts

[1408] As a concrete example, assume that User A is interested in "technology," "artificial intelligence," and "programming." This user's interest data is collected, and key keywords are extracted using a natural language processing model. New keywords such as "biology," "art," and "psychology" are selected using a "reverse" filtering method, and new information is collected based on these keywords. Visual thumbnail images of this new information are generated by AI and provided to User A. User A views the new information and thumbnail images and provides feedback on whether or not they are interested. Furthermore, an emotion engine analyzes emotions based on User A's feedback and voice input, and grasps the user's emotional state. The server uses this feedback and emotion data to update User A's profile to provide more appropriate information next time.

[1409] An example prompt might be, "The user is interested in technology, artificial intelligence, and programming. After reverse filtering, the keywords are biology, art, and psychology. Based on these keywords, please find and provide relevant, up-to-date articles and information."

[1410] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1411] Step 1:

[1412] Server collection of user interest data

[1413] The server collects interest data such as user access history, search history, and bookmark data, including the websites the user visits, keywords searched, articles viewed, and pages bookmarked. The collected data is stored in a database as individual user profiles.

[1414] Input: Access history, search history, bookmark data

[1415] Output: Save interest data in user profile

[1416] Step 2:

[1417] Server-generated keyword mappings

[1418] The server extracts key keywords from the collected interest data using natural language processing models (e.g., TF-IDF-based vectorization and clustering). For example, if a user frequently views articles related to "technology," "artificial intelligence," and "programming," these keywords will be extracted. The extracted keywords are then mapped to a keyword map based on their relevance.

[1419] Input: Interest data (user profile)

[1420] Output: Keyword map (including a list of primary keywords)

[1421] Step 3:

[1422] Server performs "reverse" filtering

[1423] The server analyzes the generated keyword map and eliminates centrally located key keywords, including the user's primary interests: "technology," "artificial intelligence," and "programming." It then extracts peripheral new keywords (e.g., "biology," "art," and "psychology").

[1424] Input: Keyword Map

[1425] Output: New "reverse" filtered keyword list

[1426] Step 4:

[1427] Server selection and collection of new information

[1428] The server searches and collects relevant information from the Internet and specific information sources based on the new keywords extracted through "reverse" filtering, and stores the information, such as articles, news, and research papers, related to the new keywords in a database.

[1429] Input: New "reverse" filtered keyword list

[1430] Output: New information (articles, news, research papers, etc.)

[1431] Step 5:

[1432] Server-generated thumbnail images

[1433] The server uses artificial intelligence models (e.g., image generation AI) to generate thumbnail images related to the new information collected. For example, for an article related to "biology," the generated thumbnail images might be diagrams of DNA or images of cells.

[1434] Input: New information

[1435] Output: Thumbnail image

[1436] Step 6:

[1437] Device provides information and thumbnail images

[1438] The terminal displays the new information and thumbnail images sent from the server to the user, and through the user interface, the user can access the new information and visually check the related thumbnail images.

[1439] Input: New information, thumbnail image

[1440] Output: User interface display (new information and thumbnail image)

[1441] Step 7:

[1442] Collecting and entering user feedback

[1443] Users input feedback about the new information and thumbnail images provided, such as interest, relevance, rating, visual opinion, and emotional state, which is then sent to the server via the device.

[1444] Input: User feedback on new information and thumbnail images

[1445] Output: Feedback data (ratings, emotional state, etc.)

[1446] Step 8:

[1447] Emotion recognition and analysis by the server

[1448] The server uses an emotion recognition engine to analyze the user's emotional state based on the feedback data received from the user, and the analysis results are added to the user's profile.

[1449] Input: Feedback data

[1450] Output: Emotion data (fun, interesting, bored, etc.)

[1451] Step 9:

[1452] Server optimization of next information provision

[1453] The server analyzes the feedback and sentiment data and updates the user's profile to reflect new trends and interests and optimize the content of the next information provided.

[1454] Input: Feedback data, emotion data

[1455] Output: Updated user profile, optimization plan for next information provided

[1456] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1457] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1458] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1459] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1460] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1461] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1462] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1463] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1464] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1465] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1466] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1467] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1468] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1469] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1470] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1471] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1472] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1473] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1474] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1475] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1476] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1477] The following is further disclosed regarding the above embodiment.

[1478] (Claim 1)

[1479] means of collecting user interest data;

[1480] means for generating a keyword map based on the interest data;

[1481] a "reverse" filtering means for excluding keywords located at the center of the keyword map and extracting new keywords from peripheral keywords;

[1482] means for collecting new information based on the new keywords;

[1483] means for generating a thumbnail image associated with the new information;

[1484] means for providing said new information and said thumbnail image to a user;

[1485] means for collecting feedback from said users;

[1486] A system including:

[1487] (Claim 2)

[1488] 10. The system of claim 1, wherein the means for generating the keyword map uses a natural language processing model.

[1489] (Claim 3)

[1490] 10. The system of claim 1, wherein the means for generating the thumbnail images uses an artificial intelligence model.

[1491] "Example 1"

[1492] (Claim 1)

[1493] a means for collecting user information;

[1494] means for generating a keyword map based on the user information;

[1495] an inverse filtering means for excluding keywords located at the center of the keyword map and extracting new keywords from peripheral keywords;

[1496] means for collecting new information based on the new keywords;

[1497] means for generating an image related to the new information;

[1498] means for providing said new information and said image to a user;

[1499] means for collecting ratings from said users;

[1500] A system including:

[1501] (Claim 2)

[1502] 10. The system of claim 1, wherein the means for generating the keyword map uses a natural language processing model.

[1503] (Claim 3)

[1504] 10. The system of claim 1, wherein the means for generating the image uses a generative artificial intelligence model.

[1505] "Application Example 1"

[1506] (Claim 1)

[1507] means of collecting user interest data;

[1508] means for generating a keyword map based on the interest data;

[1509] a "reverse" filtering means for excluding keywords located at the center of the keyword map and extracting new keywords from peripheral keywords;

[1510] means for collecting new information based on the new keywords;

[1511] means for generating a thumbnail image associated with the new information;

[1512] means for providing said new information and said thumbnail image to a user;

[1513] means for collecting feedback from said users;

[1514] A means of recognizing the user within the device;

[1515] a means for the device to provide information based on the user's location;

[1516] A system including:

[1517] (Claim 2)

[1518] 10. The system of claim 1, wherein the means for generating the keyword map uses a natural language processing model.

[1519] (Claim 3)

[1520] 10. The system of claim 1, wherein the means for generating the thumbnail images uses an artificial intelligence model.

[1521] "Example 2: Combining Emotion Engines"

[1522] (Claim 1)

[1523] means of collecting user interest data;

[1524] means for generating a keyword map based on the interest data;

[1525] a "reverse" filtering means for excluding keywords located at the center of the keyword map and extracting new keywords from peripheral keywords;

[1526] A means for searching and collecting new information based on the new keywords;

[1527] means for generating visual content related to the new information;

[1528] means for providing said new information and said visual content to a user;

[1529] means for collecting feedback and emotional state from said user;

[1530] means for updating a user profile based on said feedback and emotion data;

[1531] A system including:

[1532] (Claim 2)

[1533] 10. The system of claim 1, wherein the means for generating the keyword map uses a natural language processing model.

[1534] (Claim 3)

[1535] 10. The system of claim 1, wherein the means for generating visual content uses an artificial intelligence model.

[1536] "Application example 2 when combining emotion engines"

[1537] (Claim 1)

[1538] means of collecting user interest data;

[1539] means for generating a keyword map based on the interest data;

[1540] a "reverse" filtering means for excluding keywords located at the center of the keyword map and extracting new keywords from peripheral keywords;

[1541] means for collecting new information based on the new keywords;

[1542] means for generating a thumbnail image associated with the new information;

[1543] means for providing said new information and said thumbnail image to a user;

[1544] means for collecting feedback from said users;

[1545] emotion recognition means for analyzing the user's emotion based on the feedback;

[1546] A means for optimizing next information to be provided based on the analyzed emotion;

[1547] A system including:

[1548] (Claim 2)

[1549] 10. The system of claim 1, wherein the means for generating the keyword map uses a natural language processing model.

[1550] (Claim 3)

[1551] 10. The system of claim 1, wherein the means for generating the thumbnail images uses an artificial intelligence model. [Explanation of symbols]

[1552] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means of collecting user interest data; means for generating a keyword map based on the interest data; a "reverse" filtering means for excluding keywords located at the center of the keyword map and extracting new keywords from peripheral keywords; means for collecting new information based on the new keywords; means for generating a thumbnail image associated with the new information; means for providing said new information and said thumbnail image to a user; means for collecting feedback from said users; A system including:

2. 10. The system of claim 1, wherein the means for generating the keyword map uses a natural language processing model.

3. 10. The system of claim 1, wherein the means for generating the thumbnail images uses an artificial intelligence model.

Images