system

Abstract

Provide a system. [Solution] means for selecting a news article and saving the URL of the selected news article; a means for periodically retrieving news articles based on the URLs of the stored news articles; means for summarizing the content of the retrieved news articles; a means for storing the summarized news article content in a database; and means for displaying updated information on summarized news articles to a user.

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] In news, only the initial impactful information is often widely reported, while subsequent updates and completed content tend to be overlooked. This makes it difficult to keep track of a particular news article until the end. Users want news articles they have added to their favorites to be updated with the latest information, but there is no system that efficiently provides this. Therefore, there is a need to provide a system that allows users to always keep track of summaries and the latest information of their favorite news articles. [Means for solving the problem]

[0005] The present invention provides a system including a means for selecting a news article and saving the URL of the selected news article, a means for periodically retrieving news articles based on the saved URL of the news article, a means for summarizing the content of the retrieved news article, a means for saving the content of the summarized news article in a database, and a means for displaying the latest information of the summarized news article to a user. This allows a user to easily summarize the content of favorited news articles and obtain the latest information in real time. The system allows a user to view constantly updated news summaries.

[0006] A "news article" is the textual content of information published by a news provider.

[0007] "URL" is an abbreviation for Uniform Resource Locator, and is an address that indicates the location of a resource on the Internet.

[0008] "User" means an individual or organization that uses the system.

[0009] "Storage means" refers to a method or device for recording the URL of a news article or its summarized content in a storage device such as a database.

[0010] "Means for acquiring" refers to a method or device for collecting the contents of news articles from the Internet or the like based on the URLs of the stored news articles.

[0011] The "summarizing means" is a method or device that shortens the content of the acquired news article and extracts and summarizes only the important information.

[0012] A "displaying means" is a method or device that makes the summarized news article content viewable on a user's terminal.

[0013] "Periodic" means repeated at predetermined time intervals.

[0014] A "database" is a structured collection of data and a system for efficiently storing, retrieving, and managing information.

[0015] "Latest News" is the content or summary of the most recent news article as of the current time.

[0016] "Real-time" means processing information instantly and providing it to users without delay. [Brief explanation of the drawings]

[0017] [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 illustrating 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

[0018] 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.

[0019] First, the terms used in the following description will be explained.

[0020] 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).

[0021] 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.

[0022] 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.

[0023] 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.

[0024] 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."

[0025] [First embodiment]

[0026] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0027] 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.

[0028] 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).

[0029] 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.

[0030] 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.

[0031] 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.

[0032] 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.

[0033] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0034] 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.

[0035] 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.

[0036] 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.

[0037] 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."

[0038] This invention provides a system that allows users to easily track news articles and obtain the latest information. By following the steps below, users can always check the latest summaries of their favorite news articles.

[0039] First, when a user finds an article of interest on a news site, they select the article and click the "Favorite" button. The device (browser) then sends the URL of the news article selected by the user and their user ID to the server.

[0040] The server stores the received URL and user ID in its database, which adds the selected news article to the user's favorites and saves it.

[0041] Next, the server periodically accesses news sites based on the URLs of the news articles stored in the database to check for new information. This is done using web scraping technology. The server sends an HTTP request and receives HTML data as a response. The received HTML data is then analyzed to extract the article content (title, body, date, etc.).

[0042] The content of the retrieved news articles is summarized using natural language processing (NLP) technology. The server inputs the retrieved article text into an NLP model to generate a summary. The generated summary is stored in a database. The stored information includes the news article URL, summary content, and the date and time the summary was generated.

[0043] When a user accesses a dedicated web page, the device sends the user ID to the server and requests the latest summaries of their favorite articles. The server retrieves summaries of the latest news articles related to the user ID from the database and sends them back to the device. The device then displays the summaries on the user's screen.

[0044] As a concrete example, consider the case where a user adds an article about earthquakes to their favorites. When the user clicks the "Favorite" button, the URL (e.g., https: / / example.com / earthquake-news) is sent to the server and stored in a database. The server periodically accesses this URL and scrapes new information. For example, if new information is added to the article, such as "The magnitude of the earthquake was 6.8, and the damage was minor," the server inputs that information into an NLP model and generates a summary.

[0045] When a user accesses a dedicated web page, the server retrieves the latest news summary and sends it to the user's device, such as "The magnitude of the earthquake is 6.8, and the damage is minor." The device displays this information, allowing the user to easily check the latest news summary.

[0046] In this way, the present invention provides a system that efficiently provides users with up-to-date information on their favorite news articles.

[0047] The processing flow will be explained below.

[0048] Step 1:

[0049] The user selects a news article and clicks the "Favorite" button. The device (browser) sends the news article URL and user ID to the server.

[0050] Step 2:

[0051] The server receives the request and stores the news article URL and the user ID in a database, which adds the selected news article to the user's favorites.

[0052] Step 3:

[0053] The server sets up a scheduler to periodically retrieve the list of favorite article URLs from the database. For example, the scheduler is set to perform scraping processing every hour.

[0054] Step 4:

[0055] At the scheduled time, the server retrieves the list of URLs from the database, sends an HTTP request to each news site, and receives the HTML response.

[0056] Step 5:

[0057] The server parses the received HTML data and extracts the article content, such as the news article title, body, date, etc. For this, it uses a scraping library (e.g. BeautifulSoup).

[0058] Step 6:

[0059] The server inputs the extracted news article content into a natural language processing (NLP) model, such as a text summarization model like BERT or GPT.

[0060] Step 7:

[0061] The server generates a summary based on the results of the NLP model and stores it in a database. The stored information includes the URL of the news article, the summary content, and the date and time the summary was generated.

[0062] Step 8:

[0063] A user accesses a dedicated web page. The device (browser) sends an HTTP request including the user ID to the server, requesting the latest summaries of favorite articles.

[0064] Step 9:

[0065] The server retrieves summaries of the latest news articles associated with the user ID from the database and returns the retrieved summaries to the terminal as an HTTP response.

[0066] Step 10:

[0067] The terminal (browser) displays the received summary on the user's viewing screen. For example, a summary such as "The magnitude of the earthquake was 6.8, and the damage was minor" may be displayed.

[0068] In this way, users can efficiently track and stay up to date on their favorite news articles in real time.

[0069] Example 1

[0070] 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."

[0071] It is difficult to efficiently track articles that interest users from the wide variety of news articles available on the Internet and to obtain the latest information in a timely manner. Especially in today's information-intensive world, it takes time and effort for users to manually check articles regularly and for new information. To solve this problem, a system is needed that can automatically retrieve the latest information for news articles that users have shown interest in, and then summarize and provide them.

[0072] 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.

[0073] In this invention, the server includes means for saving URLs of news articles selected by the user, means for periodically retrieving news articles based on the URLs of the saved news articles, means for summarizing the content of the retrieved news articles using a natural language processing model, means for saving the content of the summarized news articles in a database, and means for displaying the latest information on the summarized news articles to the user, thereby enabling the user to automatically and efficiently obtain the latest information on news articles in which they are interested.

[0074] A "news article" is an information article published on a news site on the Internet.

[0075] A "URL" is an Internet address that uniquely identifies a news article or web page.

[0076] "Means for preservation" refers to a data storage function for recording the URL of a news article and its summarized content.

[0077] "Means of acquisition" refers to the technology used to collect news article data via the Internet.

[0078] A "natural language processing model" is an artificial intelligence technology that automatically summarizes and analyzes text data.

[0079] "Summarization methods" are techniques for shortening the content of news articles and extracting the main points.

[0080] A "database" is a system for efficiently storing, managing, and retrieving organized data.

[0081] "Display means" refers to a technique for visually presenting information to a user's terminal.

[0082] The "Internet" is a communications infrastructure that connects computer networks around the world.

[0083] A "user" is any person or entity that utilizes the invention to track news stories and obtain up-to-date information.

[0084] A "server" is a computer system that processes requests from clients (user terminals) and provides information.

[0085] The present invention provides a system that allows users to easily track and update news articles, allowing users to always see the latest summaries of their favorite news articles.

[0086] First, when a user finds an article of interest on a news site, they select the article and click the "Favorite" button. The device (browser) then sends the URL of the news article selected by the user and their user ID to the server.

[0087] The server stores the received URL and user ID in a database, which adds the selected news article to the user's favorites. The database also stores metadata associated with the article (e.g., title, date).

[0088] Next, the server periodically accesses news sites based on the URLs of the news articles stored in the database to check for new information. This is done using web scraping techniques (e.g., BeautifulSoup, Scrapy). The server sends an HTTP request and receives HTML data as a response. The received HTML data is then parsed to extract the article content (title, body, date, etc.).

[0089] The content of the retrieved news articles is summarized using natural language processing (NLP) technology (e.g., OpenAI's GPT-4®). The server inputs the retrieved article text into an NLP model to generate a summary. The generated summary is stored in a database. The stored information includes the news article URL, summary content, and the date and time the summary was generated.

[0090] When a user accesses a dedicated web page, the device sends the user ID to the server and requests the latest summaries of their favorite articles. The server retrieves summaries of the latest news articles related to the user ID from the database and sends them back to the device. The device then displays the summaries on the user's screen.

[0091] As a concrete example, consider the case where a user adds an article about earthquakes to their favorites. When the user clicks the "Favorite" button, the URL (e.g., https: / / example.com / earthquake-news) is sent to the server and stored in a database. The server periodically accesses this URL and scrapes new information. For example, if new information is added to the article, such as "The magnitude of the earthquake was 6.8, and the damage was minor," the server inputs that information into an NLP model and generates a summary.

[0092] When a user accesses a dedicated web page, the server retrieves the latest news summary and sends it to the user's device, such as "The magnitude of the earthquake is 6.8, and the damage is minor." The device displays this information, allowing the user to easily check the latest news summary.

[0093] Example prompt sentence:

[0094] Write a summary of the following news article:

[0095] Title: Earthquake Alert

[0096] Body text: At 10:00 AM on April 1, 2023, a magnitude 6.8 earthquake occurred in a certain area. Damage was minor, and details are currently under investigation.

[0097] In this way, the present invention provides a system that efficiently provides users with up-to-date information on their favorite news articles.

[0098] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0099] Program processing steps

[0100] Step 1:

[0101] Input: A user browses a news site and clicks the "favorite" button on an article that interests them.

[0102] How it works: When a user clicks the "Favorite" button, the device (browser) generates the URL of the selected news article and the user ID in JSON format.

[0103] Output: The generated JSON data is sent to the server as an HTTP POST request.

[0104] Step 2:

[0105] Input: HTTP POST request sent from the device (news article URL and user ID).

[0106] How it works: The server parses the incoming HTTP POST request and extracts the news article URL and the user ID. The server then stores this information in a database. For example, using a SQL database, this would execute an SQL INSERT statement.

[0107] Output: The news article URL and the user ID are saved in the database.

[0108] Step 3:

[0109] Input: A list of URLs for news articles stored on the server.

[0110] How it works: The server runs a periodically scheduled job (e.g., a cron job) that retrieves a list of news article URLs from a database, sends an HTTP GET request to each URL, and parses the resulting HTML.

[0111] Output: The retrieved HTML data is converted into the content of the news article (title, body, date, etc.).

[0112] Step 4:

[0113] Input: The content of the retrieved news article (title, body, date, etc.).

[0114] How it works: The server inputs the text data of the retrieved news article into a natural language processing (NLP) model (e.g., OpenAI's GPT-4). Specifically, it passes the text of the news article to the NLP model as a prompt.

[0115] Output: Receive the summary text generated by the NLP model and obtain it as the summary content.

[0116] Step 5:

[0117] Input: The generated summary text.

[0118] Operation: The server saves the generated summary text to a database. The information to be saved includes the news article URL, summary content, and the date and time the summary was generated. If an SQL database is used, it executes an SQL INSERT or UPDATE statement.

[0119] Output: The summary text and related information are stored in a database.

[0120] Step 6:

[0121] Input: The user accesses a dedicated web page and logs in.

[0122] How it works: When a user logs into a dedicated web page, the device sends an HTTP GET request to the server, including the user ID, to retrieve the latest summaries of the user's favorite news articles.

[0123] Output: A GET request containing the user ID is sent to the server.

[0124] Step 7:

[0125] Input: An HTTP GET request containing a user ID.

[0126] What it does: The server queries the database based on the received user ID to get the latest summaries of relevant news articles, or if using an SQL database, executes a SELECT statement.

[0127] Output: The relevant latest summary text is retrieved from the server.

[0128] Step 8:

[0129] Input: The latest summary text retrieved from the server.

[0130] Operation: The server returns the retrieved summary text to the terminal, which embeds it into HTML for display on the user's screen.

[0131] Output: The summary text is displayed on the user's device, allowing the user to check the latest news information.

[0132] (Application example 1)

[0133] 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."

[0134] It is difficult for users to easily obtain the latest summary information from the vast number of news articles on the Internet. Existing news distribution systems lack the functionality to enable users to quickly and efficiently obtain the information they need, resulting in problems such as missing the latest news and information overload. In particular, tracking the latest information for a specific article is done manually, placing a heavy burden on users. The object of this invention is to build a system that efficiently provides summary information for news articles individually selected by the user, thereby automating the acquisition of the latest information.

[0135] 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.

[0136] In this invention, the server includes means for selecting a news article and saving the URL of the selected news article, means for periodically retrieving news articles based on the URL of the saved news article, means for summarizing the content of the retrieved news article, means for saving the content of the summarized news article in a database, means for displaying the latest information on the summarized news article to a user, means for summarizing the content of the retrieved article using a natural language processing model for generating the summary, and means for notifying the user terminal of the summary information, thereby enabling automatic tracking of the latest information and provision of a summary for a specific news article selected by a user.

[0137] A "news article" is written information about current events provided on the Internet or through other electronic media.

[0138] A "user" is an individual or organization that uses this system to obtain news article information.

[0139] A "URL" is a string of characters that indicates the address on the Internet where a news article is located.

[0140] A "means" is a way or mechanism for performing a particular function or process.

[0141] "Periodic" means to repeat at regular intervals.

[0142] "Storing" means keeping data in a fixed location.

[0143] "Acquire" means to obtain information or data from an external source.

[0144] "To summarize" means to briefly summarize the contents of a news article.

[0145] A "database" is an information system for systematically storing and managing data.

[0146] "Displaying" means providing information visually to a user.

[0147] A "natural language processing model" is a machine learning algorithm for understanding and processing human language.

[0148] "Notify" means to inform a user of specific information.

[0149] To implement this invention, a user first selects a news article and sends the URL of the selected news article from their terminal to a server. The server stores this URL and the user ID in a database. Periodically, the server retrieves news articles via the Internet based on the news article URLs stored in the database. The content of the retrieved news articles is summarized on the server using a natural language processing model. For example, a machine learning algorithm such as spaCy is used as this NLP model. The summarized content of the news article is then stored again in the database.

[0150] When a user accesses the server from a device with a dedicated application installed, the server retrieves the latest summary associated with the user ID from the database and notifies the user's device of that summary information. This allows users to easily access the latest news summaries at any time. Specific hardware components include internet-enabled devices (such as smartphones or smart glasses) and a cloud server. Software components include a database management system (such as SQLite), web scraping technologies (requests, BeautifulSoup), an NLP model (spaCy), and a job scheduling library (apscheduler).

[0151] As a concrete example, consider the case where a user adds a news article about an earthquake to their favorites. When the user enters the URL of the news article and clicks the "Favorite" button, the URL (e.g., https: / / example.com / earthquake-news) is sent to the server and stored in the database. The server periodically accesses this URL and scrapes new information. For example, if new information is added to the news article stating that "the magnitude of the earthquake was 6.8, and the damage was minor," the server inputs this content into a natural language processing model and generates a summary. The summary generated is "The magnitude of the earthquake was 6.8, and the damage was minor."

[0152] When a user accesses the server from a dedicated application, the server retrieves the latest summary and notifies the user. The user can then view the latest summary information on their device screen, giving them a concise understanding of the article's content. An example of a prompt sentence generated using a generative AI model is as follows:

[0153] Prompt: "Generate a summary for the following news article text: 'The earthquake had a magnitude of 6.8 and caused minor damage. This earthquake...'"

[0154] In this way, the present invention provides a system that allows users to keep up to date with specific news articles and easily obtain summary information.

[0155] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0156] Step 1:

[0157] The user selects a news article and clicks the "Favorite" button.

[0158] Input: The URL of the news article selected by the user

[0159] Output: News article URL and user ID

[0160] Specific operation: The device sends the URL of the news article selected by the user and the user ID to the server.

[0161] Step 2:

[0162] The server stores the received URL and user ID in a database.

[0163] Input: Obtained URL and user ID

[0164] Output: URL and user ID stored in the database

[0165] Specific behavior: The server receives the URL and user ID and stores it in the database.

[0166] Step 3:

[0167] The server periodically retrieves news articles based on the URLs stored in the database.

[0168] Input: URL stored in the database

[0169] Output: HTML data of retrieved news articles

[0170] What happens: The server periodically retrieves the URL from the database and sends an HTTP request to retrieve the HTML data of the news article.

[0171] Step 4:

[0172] The server analyzes the HTML data of the retrieved news article and extracts the content of the article.

[0173] Input: Retrieved HTML data

[0174] Output: Extracted article content (text)

[0175] What happens: The server uses the BeautifulSoup library to parse the HTML data and extract the text content of the article.

[0176] Step 5:

[0177] The server inputs the extracted article content into a natural language processing model to generate a summary.

[0178] Input: Extracted article text content

[0179] Output: Summary text

[0180] What it does: The server summarizes the article text using a natural language processing model like spaCy.

[0181] Step 6:

[0182] The server stores the generated summaries in a database.

[0183] Input: Generated summary text

[0184] Output: Summary text stored in the database

[0185] Specific operation: The server stores the generated summary text in a database.

[0186] Step 7:

[0187] A user accesses the server from a terminal on which a dedicated application is installed.

[0188] Input: User request

[0189] Output: A summary list of a particular user's favorite articles

[0190] Specific operation: The terminal sends the user ID to the server and requests related summary information.

[0191] Step 8:

[0192] The server retrieves the latest summary associated with the user ID from the database and sends it to the terminal.

[0193] Input: User ID

[0194] Output: Summarized news article updates

[0195] Specific operation: The server retrieves the relevant summary text from the database and sends it to the user's terminal.

[0196] Step 9:

[0197] The user terminal displays the received summary information on the screen.

[0198] Input: Received summary text

[0199] Output: Summary text displayed on the user's screen

[0200] Specific operation: The terminal displays the received summary text on the user's screen, allowing the user to check the latest news summary.

[0201] 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.

[0202] This invention provides a system that periodically acquires the latest news article information and summarizes it by allowing a user to select a news article and save the URL of the selected news article, as well as a system that recognizes the user's emotions and optimizes the summaries and display order of news articles based on those emotions. By following the format described below, a user can always check the latest news articles that best suit their emotions.

[0203] First, when a user finds an article of interest on a news site, they select the article and click the "Favorite" button. The device (browser) then sends the URL of the news article selected by the user and their user ID to the server.

[0204] The server stores the received URL and user ID in its database, which adds the selected news article to the user's favorites and saves it.

[0205] The server periodically accesses news sites based on the URLs of news articles stored in the database to check for new information. This is done using web scraping technology. The server sends an HTTP request and receives HTML data as a response. The received HTML data is then analyzed to extract the article content (title, body, date, etc.).

[0206] The content of the retrieved news articles is summarized using natural language processing (NLP) technology. The server inputs the retrieved article text into an NLP model to generate a summary. The generated summary is stored in a database. The stored information includes the news article URL, summary content, and the date and time the summary was generated.

[0207] Furthermore, the present invention recognizes the user's emotions by combining an emotion engine. When a user accesses a dedicated web page and browses an article, the emotion engine analyzes the user's emotions based on the user's interactions and access history. This allows the user's emotional state (e.g., excitement, joy, sadness, etc.) to be recognized when reading the article.

[0208] The user's emotion data obtained by the emotion engine is used to adjust the summary content of the news article. For example, if the user's emotion is recognized as sad, the emotion engine can adjust the summary of the news article to emphasize the positive aspects.

[0209] Furthermore, the emotion engine optimizes the display order of news articles based on the user's emotional data. For example, if the user is feeling stressed, the emotion engine will prioritize articles with relaxing content.

[0210] As a concrete example, consider the case where a user adds an article about earthquakes to their favorites. When the user clicks the "Favorite" button, the URL (e.g., https: / / example.com / earthquake-news) is sent to the server and stored in a database. The server periodically accesses this URL and scrapes new information. For example, if new information is added to the article, such as "The magnitude of the earthquake was 6.8, and the damage was minor," the server inputs that information into an NLP model and generates a summary.

[0211] When a user accesses a dedicated web page, the emotion engine recognizes the user's emotions and provides summaries that emphasize positive news and news articles in an optimized viewing order according to the user's emotional state, allowing users to always view the most appropriate news summary that matches their emotions.

[0212] In this way, the present invention builds a system that not only efficiently provides the latest news information, but also realizes a personalized news browsing experience based on the user's emotions.

[0213] The processing flow will be explained below.

[0214] Step 1:

[0215] The user selects a news article and clicks the "Favorite" button. The device (browser) sends the news article URL and user ID to the server.

[0216] Step 2:

[0217] The server receives the request and stores the news article URL and the user ID in a database, which adds the selected news article to the user's favorites.

[0218] Step 3:

[0219] Set up a scheduler so that the server periodically retrieves the list of favorite article URLs from the database. For example, the scheduler can be set to perform scraping every hour.

[0220] Step 4:

[0221] At the scheduled time, the server retrieves the list of URLs from the database, sends an HTTP request to each news site, and receives the HTML response.

[0222] Step 5:

[0223] The server parses the received HTML data and extracts the article content, such as the news article title, body, date, etc. For this, it uses a scraping library (e.g. BeautifulSoup).

[0224] Step 6:

[0225] The server inputs the extracted news article content into a natural language processing (NLP) model, such as a text summarization model like BERT or GPT.

[0226] Step 7:

[0227] The server generates a summary based on the results of the NLP model and stores it in a database. The stored information includes the URL of the news article, the summary content, and the date and time the summary was generated.

[0228] Step 8:

[0229] A user accesses a dedicated web page. The device (browser) sends an HTTP request including the user ID to the server, requesting the latest summaries of favorite articles.

[0230] Step 9:

[0231] The server retrieves summaries of the latest news articles associated with the user ID from the database and returns the retrieved summaries to the terminal as an HTTP response.

[0232] Step 10:

[0233] The terminal (browser) displays the received summary on the user's viewing screen. For example, a summary such as "The magnitude of the earthquake was 6.8, and the damage was minor" may be displayed.

[0234] Step 11:

[0235] While a user is browsing a news article, the device collects the user's emotional data (facial expressions, click patterns, browsing time, etc.).

[0236] Step 12:

[0237] The device sends the collected emotion data to a server where it is analyzed by an emotion engine.

[0238] Step 13:

[0239] The server uses an emotion engine to infer the user's emotional state (e.g., happy, sad, excited, etc.).

[0240] Step 14:

[0241] The server tailors the summary of a news article based on the user's emotional state: for example, if the user is feeling sad, it generates a summary that emphasizes the positive aspects.

[0242] Step 15:

[0243] The server optimizes the display order of news articles based on the user's emotional data, for example, prioritizing articles with relaxing content.

[0244] Step 16:

[0245] When the user accesses the web page again, the server sends the adjusted summary and the optimized display order to the terminal, which then displays it on the user's viewing screen.

[0246] Through the above steps, users can efficiently track the latest updates of their favorite news articles and also check the best news summaries that match their emotions in real time.

[0247] Example 2

[0248] 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."

[0249] Conventional news article delivery systems do not adequately personalize news articles based on the user's emotions, making it difficult to provide optimal news articles based on the user's current emotional state. Furthermore, news article summarization and display order are not optimized, resulting in a poor user experience. There is a need to address these issues and provide personalized news articles based on the user's emotions.

[0250] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0251] In this invention, the server includes means for selecting a news article and saving the URL of the selected news article, means for periodically retrieving news articles based on the URL of the saved news article, means for summarizing the content of the retrieved news article, means for saving the summarized news article content in a database, means for recognizing a user's emotion and adjusting the content of the news article summary based on the emotion, means for optimizing the display order of the news articles based on the user's emotional state, and means for displaying the latest information on the summarized news articles to the user, thereby enabling personalized news article summaries according to the user's emotion and an optimal browsing experience.

[0252] "News article" refers to a description of an event or piece of information published by online or offline media.

[0253] "User" refers to an entity that uses the system to view news articles and obtain information.

[0254] "URL" is an abbreviation for Uniform Resource Locator, and is an address that indicates the location of a resource on the Internet.

[0255] A "server" refers to a computer system that provides data and services to multiple clients over a network.

[0256] A "database" refers to a structured collection of data that allows for efficient storage, retrieval, and management of large amounts of data.

[0257] "Storage methods" refers to the technologies and processes used to store data in a form that can be used for a long period of time.

[0258] "Means of acquisition" refers to the technology and processes used to search and extract stored data or information based on specific criteria.

[0259] "Summarization methods" refers to techniques or processes that extract the main points or essence from long text data and present them in a shortened form.

[0260] "Means for displaying" refers to a display device or interface for visually presenting data or information to a user.

[0261] "Emotion recognition means" refers to techniques and algorithms for detecting and classifying a user's emotional state.

[0262] "Adjustment" refers to a technique or process for optimizing some output based on input data.

[0263] "Optimization means" refers to techniques or processes that maximize the performance or efficiency of a system to achieve a certain goal.

[0264] MODE FOR CARRYING OUT THE INVENTION

[0265] This invention is a system that allows a user to select a news article and save the URL of the selected news article, thereby periodically obtaining the latest news article information, summarizing it, and providing it. Furthermore, the system has the function of recognizing the user's emotions and optimizing the summaries and display order of news articles based on those emotions. A detailed description of an embodiment of this system is provided below.

[0266] User operations

[0267] When a user finds an article of interest on a news site, they select the article and click the "Favorites" button. For example, if a user is interested in an article about earthquakes, they add the URL to their "Favorites." This causes the URL of the article to be sent to the server by the device (browser).

[0268] Terminal (browser) processing

[0269] The terminal (browser) sends the URL of the news article selected by the user and the user ID to the server using an HTTP request. The HTTP protocol is the most commonly used technology at this stage.

[0270] Example of data sent:

[0271] {

[0272] "userID": "user123",

[0273] "articleURL": "https: / / example.com / earthquake-news"

[0274] }

[0275] Server Processing

[0276] The server analyzes the received URL and user ID and stores them in a database. For example, a relational database such as MySQL (registered trademark) is commonly used. The format of the stored data is as follows:

[0277] UserID: user123

[0278] ArticleURL: https: / / example.com / earthquake-news

[0279] Check for news article updates

[0280] At regular intervals (for example, every hour), the server accesses the URLs of each news article stored in the database to check for new information. This is done using web scraping techniques, specifically software libraries such as Python's BeautifulSoup and Scrapy.

[0281] Data extraction and summary generation

[0282] The server analyzes the HTML data obtained by web scraping and extracts the article content (title, body text, date, etc.). This text data is summarized using natural language processing (NLP) techniques. Specifically, NLP models such as the BERT model and GPT-3 (registered trademark) are used. The summarized article data is then stored back in the database.

[0283] Example of retrieved news article summary:

[0284] An earthquake occurred, with a magnitude of 6.8, but the damage was minor and no injuries were reported.

[0285] Emotion recognition

[0286] When a user accesses a dedicated web page and browses news articles, the emotion engine analyzes the user's interactions and access history. This allows the system to recognize the user's emotional state (e.g., excitement, joy, sadness, etc.). Sentiment analysis algorithms include Python NLP libraries "TextBlob" and "VADER."

[0287] Example of recognized user emotion data:

[0288] {

[0289] "userID": "user123",

[0290] "emotion": "sadness",

[0291] "timestamp": "2023-10-01T12:34:56Z"

[0292] }

[0293] Optimizing news article display

[0294] The emotion engine adjusts the summary content and display order of news articles based on the user's emotion data. For example, if the user feels "sadness," the emotion engine adjusts the summary to emphasize positive summaries and prioritizes relaxing articles.

[0295] Display order example:

[0296] 1. Positive News Summary

[0297] 2. Articles related to user interests

[0298] 3. Other news

[0299] Prompt Sentence Examples

[0300] Here is an example prompt for generating a news article summary using an NLP model:

[0301] Prompt statement:

[0302] "Generate a summary of a news article. The content of the news article is as follows:

[0303] Article content: An earthquake occurred, with shaking of magnitude 6.8 observed. Damage was minor, and no casualties were reported.

[0304] Summary generated: An earthquake occurred, with shaking measured at magnitude 6.8, but the damage was minor and no casualties were reported.

[0305] In this way, the present invention can build a system that not only efficiently provides the latest news articles but also realizes a personalized news browsing experience based on the user's emotions, allowing the user to always see the most appropriate news summary that suits their emotions.

[0306] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0307] Program processing steps

[0308] Step 1:

[0309] A user adds a news article to their "favorites."

[0310] When a user browses a news site, they find an article they are interested in and click its "Favorite" button. The input is the user's click action, and the output is the click event of the "Favorite" button. This operation obtains the article's URL and the user ID.

[0311] Step 2:

[0312] The terminal (browser) sends the URL and user ID to the server.

[0313] The terminal (browser) receives the click event and sends the news article URL and user ID as JSON format data to the server via an HTTP request. The input is the URL and user ID, and the output is the HTTP request.

[0314] Step 3:

[0315] The server stores the URL and user ID in a database.

[0316] The server parses the received data and stores the news article URL and user ID in a database. Specifically, the data is inserted into a relational database (e.g. MySQL). The input is the received data, and the output is a new record in the database.

[0317] Step 4:

[0318] The server periodically checks for updates to news articles.

[0319] The server accesses the URLs of news articles stored in the database according to a scheduled task to check the latest information. This is done using web scraping techniques (e.g. BeautifulSoup, Scrapy). The input is the list of URLs in the database, and the output is the received HTML data.

[0320] Step 5:

[0321] The server parses the HTML data and extracts the article content.

[0322] The server analyzes the HTML data it receives and extracts information such as the article title, body, and date. Specifically, HTML parsing is performed. The input is HTML data, and the output is text data (title, body, date, etc.).

[0323] Step 6:

[0324] The server inputs the article content into a natural language processing (NLP) model to generate a summary.

[0325] The server inputs the extracted article content into an NLP model (e.g., BERT, GPT-3) to generate a summary. The model used is a generative AI model. The input is the extracted text data, and the output is the summary text.

[0326] Step 7:

[0327] The server stores the summaries in a database.

[0328] The server saves the generated summary text, the original URL, and the generation date and time in a database. The input is the summary text and metadata, and the output is a new record in the database.

[0329] Step 8:

[0330] Recognize emotions when users browse news articles.

[0331] While a user accesses a dedicated web page and browses news articles, the emotion engine analyzes the user's interactions and access history. Specifically, it performs emotion analysis using machine learning algorithms (e.g., TextBlob, VADER). The input is user behavior data, and the output is emotion data.

[0332] Step 9:

[0333] The server adjusts the summary content of the news article based on the user's emotional data.

[0334] The emotion engine can use the user's emotion data to adjust the summary of a news article and emphasize its positive aspects. The input is emotion data, and the output is the adjusted summary text.

[0335] Step 10:

[0336] The server optimizes the display order of news articles based on sentiment.

[0337] Optimize the display order of news articles based on emotion data. For example, if sadness is recognized, prioritize positive news articles. The input is emotion data, and the output is an optimized list of news articles.

[0338] Step 11:

[0339] The server displays summarized news article updates to the user.

[0340] When a user accesses a dedicated web page to view a news article, the server retrieves the latest information from the database and displays it on the user's terminal. The input is the user's request, and the output is the information displayed on the terminal.

[0341] (Application example 2)

[0342] 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."

[0343] Conventional news article summarization systems are limited to acquiring and summarizing news articles and do not optimize based on the user's emotions. As a result, they are unable to provide personalized news that adapts to the user's emotional state, resulting in an insufficient improvement in the user experience. Furthermore, even in physical stores, there is a lack of functionality to optimize in-store guidance and recommended products based on customer emotions, limiting the means for increasing customer satisfaction. The present invention aims to solve these problems and provide a system that optimizes news and store guidance based on the user's emotions.

[0344] 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.

[0345] In this invention, the server includes means for selecting a news article and saving the URL of the selected news article, means for periodically retrieving news articles based on the URL of the saved news article, means for summarizing the content of the retrieved news article, means for saving the content of the summarized news article in a database, means for displaying the latest information on the summarized news article to the user, means for recognizing the user's emotion, means for optimizing the summaries and display order of the news articles based on the recognized emotion, and means for recognizing the emotion of a customer in a physical store and optimizing in-store guidance and recommended products based on the emotion. This not only ensures that news articles are always provided in a form that is adapted to the user's emotion, but also enables optimal guidance and recommendations to be made in the physical store in accordance with the customer's emotion, thereby improving the overall user experience and increasing customer satisfaction.

[0346] A "news article" is a report of an event or information published online or in any other media.

[0347] "URL" stands for Uniform Resource Locator and is an address that identifies a specific web page or resource.

[0348] A "summary" is a document that condenses the content of the original news article, extracting only the main information and summarizing it concisely.

[0349] A "database" is a system or software for efficiently storing, retrieving, and managing large amounts of data.

[0350] "User" refers to a person who uses the system to read news articles and receive summarized information.

[0351] "Emotion" refers to the psychological reaction or state that a user expresses when reading an article.

[0352] A "brick and mortar store" refers to a retail store that has a physical presence and where customers can visit and purchase products.

[0353] "Guidance" refers to the information provided to customers in the store and the method of presenting it.

[0354] "Recommended products" are products that a store has specifically selected to encourage customers to purchase.

[0355] "Means of recognizing emotions" refers to technology that determines the psychological state of customers or users from data such as facial expressions and voice.

[0356] "Means of acquiring news articles" refers to the process and technology of acquiring articles from news sites etc. via the Internet.

[0357] "Means for optimizing the display order" refers to techniques and methods for adjusting the order of information provided based on the emotional state of the user or customer.

[0358] This invention is a system that selects news articles, saves the URLs of the selected news articles, periodically obtains the latest information, summarizes them, and provides them to users. It recognizes the user's emotions and optimizes the summaries and display order of news articles based on those emotions, and also optimizes in-store guidance and recommended products based on customer emotions in physical stores. Implementing this invention requires the following steps.

[0359] First, when a user finds an article of interest on a news site, they select the article and click the "Favorites" button. The device sends the URL of the news article selected by the user and their user ID to the server. The server saves the received URL and user ID in a database. This adds the selected news article to the user's favorites and saves it.

[0360] The server then periodically accesses news sites based on the URLs of the news articles stored in the database to check for new information. This process uses web scraping technology. The server sends an HTTP request and receives HTML data as a response. The received HTML data is then parsed to extract the article content (title, body, date, etc.).

[0361] The content of the retrieved news articles is summarized using natural language processing (NLP) technology. The server inputs the retrieved article text into an NLP model to generate a summary. The generated summary is stored in a database. The stored information includes the news article URL, summary content, and the date and time the summary was generated. Python (registered trademark)-based libraries (e.g., BeautifulSoup, spaCy, and HuggingFace Transformers) are used to retrieve the news articles and generate the summaries.

[0362] Furthermore, the present invention recognizes user emotions by incorporating an emotion engine. When a user accesses a dedicated web page and browses articles, the emotion engine analyzes the user's emotions based on the user's interactions and access history. This uses image processing and machine learning models (e.g., OpenCV, Dlib, TENSORFLOW (registered trademark)). This allows the user's emotional state (e.g., excitement, joy, sadness, etc.) to be recognized as they read the articles.

[0363] The user's emotion data acquired by the emotion engine is used to adjust the summary content and display order of news articles. For example, if the user's emotion of sadness is recognized, the emotion engine can adjust the summary of the news article to emphasize the positive aspects. Also, if the user is feeling stressed, articles with content that will help them relax will be displayed preferentially.

[0364] This system can also be applied in physical stores. Using cameras installed in the store and the camera on the customer's smartphone, it recognizes emotions in real time from the customer's facial expressions and voice. The emotion recognition engine determines the customer's emotions and optimizes in-store guidance and recommended products according to those emotions. For example, if a customer is feeling stressed, relaxation products will be recommended first. The selection of recommended products and the order in which guidance information is displayed are also adjusted appropriately.

[0365] An example of an input prompt for a generative AI model is:

[0366] "If the customer's current emotional state is identified as stressed, generate information to recommend relaxation products. Also include information about the latest sales."

[0367] Based on these prompts, the generative AI model can generate stress-relieving products, their features, applicable discount information, and other information, which can be delivered to a display or smart glasses in real time.

[0368] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0369] Step 1:

[0370] When a user finds an article of interest on a news site, the device clicks the "Favorites" button.

[0371] Input: URL of the news article selected by the user, user ID

[0372] Specific operation: When the "Favorites" button is clicked, the device sends the URL of the selected news article and the user ID to the server.

[0373] Output: The URL and user ID are sent to the server.

[0374] Step 2:

[0375] The server stores the received URL and user ID in a database.

[0376] Input: URL of news article, user ID

[0377] Specific operation: The server stores the received URL and user ID in a database, and the selected news article is added to the user's favorites.

[0378] Output: URL and user ID stored in the database

[0379] Step 3:

[0380] The server periodically accesses news sites based on the URLs of news articles stored in the database to check for new information.

[0381] Input: URL stored in the database

[0382] Specific operation: The server sends an HTTP request and receives HTML data as a response. The received HTML data is analyzed to extract the article content (title, body, date, etc.).

[0383] Output: Extracted article content

[0384] Step 4:

[0385] The server inputs the retrieved news article text into a natural language processing (NLP) model to generate a summary.

[0386] Input: Article content (text data)

[0387] What it does: The server uses NLP techniques (e.g., spaCy or HuggingFace Transformers) to summarize the article text.

[0388] Output: Generated summary

[0389] Step 5:

[0390] The server stores the generated summaries in a database.

[0391] Input: Generated summary, news article URL, summary generation date and time

[0392] Specific operation: The server stores the generated summary, the URL of the news article, and the date and time the summary was generated in a database.

[0393] Output: Summary information stored in a database

[0394] Step 6:

[0395] When a user accesses a dedicated web page and reads a news article, the server uses an emotion recognition engine to recognize the user's emotion.

[0396] Input: Real-time facial and voice data of the user

[0397] Specific operation: The server uses image processing and machine learning models (e.g., OpenCV, Dlib, TensorFlow) to determine the user's emotions.

[0398] Output: Recognized emotion data

[0399] Step 7:

[0400] The server adjusts the summary content and display order of news articles based on the recognized emotion data.

[0401] Input: News article summary information, sentiment data

[0402] What it does: The emotion engine uses emotion data to adjust the summary content, highlighting positive aspects and changing the display order.

[0403] Output: Adjusted news article display information

[0404] Step 8:

[0405] The server transmits the adjusted news article display information to the user's terminal, and displays the latest news articles.

[0406] Input: Adjusted news article display information

[0407] Specific operation: Based on the received information, the device provides the user with a summary of the news articles and a display order that reflects the adjusted information.

[0408] Output: News article displayed on user's device

[0409] Step 9:

[0410] In a physical store, the device captures the customer's facial expressions and voice and sends them to an emotion recognition engine.

[0411] Input: Real-time facial and voice data

[0412] Specific operation: Cameras and smartphones in physical stores capture customers' facial expressions and voices and send them to a server.

[0413] Output: Facial expression and voice data sent to the server

[0414] Step 10:

[0415] The server optimizes in-store guidance and recommended products based on the customer's emotional state.

[0416] Input: Customer emotion data, in-store product information

[0417] Specific operation: The emotion recognition engine selects appropriate guidance and recommended products based on emotional data and displays them on the display or smart device.

[0418] Output: Optimized navigation information and recommended products

[0419] Step 11:

[0420] Smart displays in physical stores present emotion-based information.

[0421] Input: Optimized guidance information and recommended product information

[0422] Specific operation: Displays in physical stores display optimized guidance and recommended products to customers based on information received from the server.

[0423] Output: Optimization information displayed on the screen

[0424] Step 12:

[0425] It uses input prompt sentences to the generative AI model to generate that information.

[0426] Input: Prompt (e.g., "If the customer's current emotional state is identified as stressed, generate information to recommend relaxation products. Also include information about the latest sales.")

[0427] What it does: Based on this prompt, the generative AI model generates and delivers appropriate information in real time.

[0428] Output: Generated relaxation product information and special sale information

[0429] 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.

[0430] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (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.

[0431] 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.

[0432] [Second embodiment]

[0433] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0434] 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.

[0435] 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).

[0436] 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.

[0437] 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.

[0438] 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).

[0439] 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.

[0440] 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.

[0441] 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.

[0442] 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.

[0443] In the smart glasses 214, the 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.

[0444] 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."

[0445] This invention provides a system that allows users to easily track news articles and obtain the latest information. By following the steps below, users can always check the latest summaries of their favorite news articles.

[0446] First, when a user finds an article of interest on a news site, they select the article and click the "Favorite" button. The device (browser) then sends the URL of the news article selected by the user and their user ID to the server.

[0447] The server stores the received URL and user ID in its database, which adds the selected news article to the user's favorites and saves it.

[0448] Next, the server periodically accesses news sites based on the URLs of the news articles stored in the database to check for new information. This is done using web scraping technology. The server sends an HTTP request and receives HTML data as a response. The received HTML data is then analyzed to extract the article content (title, body, date, etc.).

[0449] The content of the retrieved news articles is summarized using natural language processing (NLP) technology. The server inputs the retrieved article text into an NLP model to generate a summary. The generated summary is stored in a database. The stored information includes the news article URL, summary content, and the date and time the summary was generated.

[0450] When a user accesses a dedicated web page, the device sends the user ID to the server and requests the latest summaries of their favorite articles. The server retrieves summaries of the latest news articles related to the user ID from the database and sends them back to the device. The device then displays the summaries on the user's screen.

[0451] As a concrete example, consider the case where a user adds an article about earthquakes to their favorites. When the user clicks the "Favorite" button, the URL (e.g., https: / / example.com / earthquake-news) is sent to the server and stored in a database. The server periodically accesses this URL and scrapes new information. For example, if new information is added to the article, such as "The magnitude of the earthquake was 6.8, and the damage was minor," the server inputs that information into an NLP model and generates a summary.

[0452] When a user accesses a dedicated web page, the server retrieves the latest news summary and sends it to the user's device, such as "The magnitude of the earthquake is 6.8, and the damage is minor." The device displays this information, allowing the user to easily check the latest news summary.

[0453] In this way, the present invention provides a system that efficiently provides users with up-to-date information on their favorite news articles.

[0454] The processing flow will be explained below.

[0455] Step 1:

[0456] The user selects a news article and clicks the "Favorite" button. The device (browser) sends the news article URL and user ID to the server.

[0457] Step 2:

[0458] The server receives the request and stores the news article URL and the user ID in a database, which adds the selected news article to the user's favorites.

[0459] Step 3:

[0460] The server sets up a scheduler to periodically retrieve the list of favorite article URLs from the database. For example, the scheduler is set to perform scraping processing every hour.

[0461] Step 4:

[0462] At the scheduled time, the server retrieves the list of URLs from the database, sends an HTTP request to each news site, and receives the HTML response.

[0463] Step 5:

[0464] The server parses the received HTML data and extracts the article content, such as the news article title, body, date, etc. For this, it uses a scraping library (e.g. BeautifulSoup).

[0465] Step 6:

[0466] The server inputs the extracted news article content into a natural language processing (NLP) model, such as a text summarization model like BERT or GPT.

[0467] Step 7:

[0468] The server generates a summary based on the results of the NLP model and stores it in a database. The stored information includes the URL of the news article, the summary content, and the date and time the summary was generated.

[0469] Step 8:

[0470] A user accesses a dedicated web page. The device (browser) sends an HTTP request including the user ID to the server, requesting the latest summaries of favorite articles.

[0471] Step 9:

[0472] The server retrieves summaries of the latest news articles associated with the user ID from the database and returns the retrieved summaries to the terminal as an HTTP response.

[0473] Step 10:

[0474] The terminal (browser) displays the received summary on the user's viewing screen. For example, a summary such as "The magnitude of the earthquake was 6.8, and the damage was minor" may be displayed.

[0475] In this way, users can efficiently track and stay up to date on their favorite news articles in real time.

[0476] Example 1

[0477] 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."

[0478] It is difficult to efficiently track articles that interest users from the wide variety of news articles available on the Internet and to obtain the latest information in a timely manner. Especially in today's information-intensive world, it takes time and effort for users to manually check articles regularly and for new information. To solve this problem, a system is needed that can automatically retrieve the latest information for news articles that users have shown interest in, and then summarize and provide them.

[0479] 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.

[0480] In this invention, the server includes means for saving URLs of news articles selected by the user, means for periodically retrieving news articles based on the URLs of the saved news articles, means for summarizing the content of the retrieved news articles using a natural language processing model, means for saving the content of the summarized news articles in a database, and means for displaying the latest information on the summarized news articles to the user, thereby enabling the user to automatically and efficiently obtain the latest information on news articles in which they are interested.

[0481] A "news article" is an information article published on a news site on the Internet.

[0482] A "URL" is an Internet address that uniquely identifies a news article or web page.

[0483] "Means for preservation" refers to a data storage function for recording the URL of a news article and its summarized content.

[0484] "Means of acquisition" refers to the technology used to collect news article data via the Internet.

[0485] A "natural language processing model" is an artificial intelligence technology that automatically summarizes and analyzes text data.

[0486] "Summarization methods" are techniques for shortening the content of news articles and extracting the main points.

[0487] A "database" is a system for efficiently storing, managing, and retrieving organized data.

[0488] "Display means" refers to a technique for visually presenting information to a user's terminal.

[0489] The "Internet" is a communications infrastructure that connects computer networks around the world.

[0490] A "user" is any person or entity that utilizes the invention to track news stories and obtain up-to-date information.

[0491] A "server" is a computer system that processes requests from clients (user terminals) and provides information.

[0492] The present invention provides a system that allows users to easily track and update news articles, allowing users to always see the latest summaries of their favorite news articles.

[0493] First, when a user finds an article of interest on a news site, they select the article and click the "Favorite" button. The device (browser) then sends the URL of the news article selected by the user and their user ID to the server.

[0494] The server stores the received URL and user ID in a database, which adds the selected news article to the user's favorites. The database also stores metadata associated with the article (e.g., title, date).

[0495] Next, the server periodically accesses news sites based on the URLs of the news articles stored in the database to check for new information. This is done using web scraping techniques (e.g., BeautifulSoup, Scrapy). The server sends an HTTP request and receives HTML data as a response. The received HTML data is then parsed to extract the article content (title, body, date, etc.).

[0496] The content of the retrieved news articles is summarized using natural language processing (NLP) technology (e.g., OpenAI's GPT-4). The server inputs the retrieved article text into an NLP model to generate a summary. The generated summary is stored in a database. The stored information includes the news article URL, summary content, and the date and time the summary was generated.

[0497] When a user accesses a dedicated web page, the device sends the user ID to the server and requests the latest summaries of their favorite articles. The server retrieves summaries of the latest news articles related to the user ID from the database and sends them back to the device. The device then displays the summaries on the user's screen.

[0498] As a concrete example, consider the case where a user adds an article about earthquakes to their favorites. When the user clicks the "Favorite" button, the URL (e.g., https: / / example.com / earthquake-news) is sent to the server and stored in a database. The server periodically accesses this URL and scrapes new information. For example, if new information is added to the article, such as "The magnitude of the earthquake was 6.8, and the damage was minor," the server inputs that information into an NLP model and generates a summary.

[0499] When a user accesses a dedicated web page, the server retrieves the latest news summary and sends it to the user's device, such as "The magnitude of the earthquake is 6.8, and the damage is minor." The device displays this information, allowing the user to easily check the latest news summary.

[0500] Example prompt sentence:

[0501] Write a summary of the following news article:

[0502] Title: Earthquake Alert

[0503] Body text: At 10:00 AM on April 1, 2023, a magnitude 6.8 earthquake occurred in a certain area. Damage was minor, and details are currently under investigation.

[0504] In this way, the present invention provides a system that efficiently provides users with up-to-date information on their favorite news articles.

[0505] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0506] Program processing steps

[0507] Step 1:

[0508] Input: A user browses a news site and clicks the "favorite" button on an article that interests them.

[0509] How it works: When a user clicks the "Favorite" button, the device (browser) generates the URL of the selected news article and the user ID in JSON format.

[0510] Output: The generated JSON data is sent to the server as an HTTP POST request.

[0511] Step 2:

[0512] Input: HTTP POST request sent from the device (news article URL and user ID).

[0513] How it works: The server parses the incoming HTTP POST request and extracts the news article URL and the user ID. The server then stores this information in a database. For example, using a SQL database, this would execute an SQL INSERT statement.

[0514] Output: The news article URL and the user ID are saved in the database.

[0515] Step 3:

[0516] Input: A list of URLs for news articles stored on the server.

[0517] How it works: The server runs a periodically scheduled job (e.g., a cron job) that retrieves a list of news article URLs from a database, sends an HTTP GET request to each URL, and parses the resulting HTML.

[0518] Output: The retrieved HTML data is converted into the content of the news article (title, body, date, etc.).

[0519] Step 4:

[0520] Input: The content of the retrieved news article (title, body, date, etc.).

[0521] How it works: The server inputs the text data of the retrieved news article into a natural language processing (NLP) model (e.g., OpenAI's GPT-4). Specifically, it passes the text of the news article to the NLP model as a prompt.

[0522] Output: Receive the summary text generated by the NLP model and obtain it as the summary content.

[0523] Step 5:

[0524] Input: The generated summary text.

[0525] Operation: The server saves the generated summary text to a database. The information to be saved includes the news article URL, summary content, and the date and time the summary was generated. If an SQL database is used, it executes an SQL INSERT or UPDATE statement.

[0526] Output: The summary text and related information are stored in a database.

[0527] Step 6:

[0528] Input: The user accesses a dedicated web page and logs in.

[0529] How it works: When a user logs into a dedicated web page, the device sends an HTTP GET request to the server, including the user ID, to retrieve the latest summaries of the user's favorite news articles.

[0530] Output: A GET request containing the user ID is sent to the server.

[0531] Step 7:

[0532] Input: An HTTP GET request containing a user ID.

[0533] What it does: The server queries the database based on the received user ID to get the latest summaries of relevant news articles, or if using an SQL database, executes a SELECT statement.

[0534] Output: The relevant latest summary text is retrieved from the server.

[0535] Step 8:

[0536] Input: The latest summary text retrieved from the server.

[0537] Operation: The server returns the retrieved summary text to the terminal, which embeds it into HTML for display on the user's screen.

[0538] Output: The summary text is displayed on the user's device, allowing the user to check the latest news information.

[0539] (Application example 1)

[0540] 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."

[0541] It is difficult for users to easily obtain the latest summary information from the vast number of news articles on the Internet. Existing news distribution systems lack the functionality to enable users to quickly and efficiently obtain the information they need, resulting in problems such as missing the latest news and information overload. In particular, tracking the latest information for a specific article is done manually, placing a heavy burden on users. The object of this invention is to build a system that efficiently provides summary information for news articles individually selected by the user, thereby automating the acquisition of the latest information.

[0542] 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.

[0543] In this invention, the server includes means for selecting a news article and saving the URL of the selected news article, means for periodically retrieving news articles based on the URL of the saved news article, means for summarizing the content of the retrieved news article, means for saving the content of the summarized news article in a database, means for displaying the latest information on the summarized news article to a user, means for summarizing the content of the retrieved article using a natural language processing model for generating the summary, and means for notifying the user terminal of the summary information, thereby enabling automatic tracking of the latest information and provision of a summary for a specific news article selected by a user.

[0544] A "news article" is written information about current events provided on the Internet or through other electronic media.

[0545] A "user" is an individual or organization that uses this system to obtain news article information.

[0546] A "URL" is a string of characters that indicates the address on the Internet where a news article is located.

[0547] A "means" is a way or mechanism for performing a particular function or process.

[0548] "Periodic" means to repeat at regular intervals.

[0549] "Storing" means keeping data in a fixed location.

[0550] "Acquire" means to obtain information or data from an external source.

[0551] "To summarize" means to briefly summarize the contents of a news article.

[0552] A "database" is an information system for systematically storing and managing data.

[0553] "Displaying" means providing information visually to a user.

[0554] A "natural language processing model" is a machine learning algorithm for understanding and processing human language.

[0555] "Notify" means to inform a user of specific information.

[0556] To implement this invention, a user first selects a news article and sends the URL of the selected news article from their terminal to a server. The server stores this URL and the user ID in a database. Periodically, the server retrieves news articles via the Internet based on the news article URLs stored in the database. The content of the retrieved news articles is summarized on the server using a natural language processing model. For example, a machine learning algorithm such as spaCy is used as this NLP model. The summarized content of the news article is then stored again in the database.

[0557] When a user accesses the server from a device with a dedicated application installed, the server retrieves the latest summary associated with the user ID from the database and notifies the user's device of that summary information. This allows users to easily access the latest news summaries at any time. Specific hardware components include internet-enabled devices (such as smartphones or smart glasses) and a cloud server. Software components include a database management system (such as SQLite), web scraping technologies (requests, BeautifulSoup), an NLP model (spaCy), and a job scheduling library (apscheduler).

[0558] As a concrete example, consider the case where a user adds a news article about an earthquake to their favorites. When the user enters the URL of the news article and clicks the "Favorite" button, the URL (e.g., https: / / example.com / earthquake-news) is sent to the server and stored in the database. The server periodically accesses this URL and scrapes new information. For example, if new information is added to the news article stating that "the magnitude of the earthquake was 6.8, and the damage was minor," the server inputs this content into a natural language processing model and generates a summary. The summary generated is "The magnitude of the earthquake was 6.8, and the damage was minor."

[0559] When a user accesses the server from a dedicated application, the server retrieves the latest summary and notifies the user. The user can then view the latest summary information on their device screen, giving them a concise understanding of the article's content. An example of a prompt sentence generated using a generative AI model is as follows:

[0560] Prompt: "Generate a summary for the following news article text: 'The earthquake had a magnitude of 6.8 and caused minor damage. This earthquake...'"

[0561] In this way, the present invention provides a system that allows users to keep up to date with specific news articles and easily obtain summary information.

[0562] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0563] Step 1:

[0564] The user selects a news article and clicks the "Favorite" button.

[0565] Input: The URL of the news article selected by the user

[0566] Output: News article URL and user ID

[0567] Specific operation: The device sends the URL of the news article selected by the user and the user ID to the server.

[0568] Step 2:

[0569] The server stores the received URL and user ID in a database.

[0570] Input: Obtained URL and user ID

[0571] Output: URL and user ID stored in the database

[0572] Specific behavior: The server receives the URL and user ID and stores it in the database.

[0573] Step 3:

[0574] The server periodically retrieves news articles based on the URLs stored in the database.

[0575] Input: URL stored in the database

[0576] Output: HTML data of retrieved news articles

[0577] What happens: The server periodically retrieves the URL from the database and sends an HTTP request to retrieve the HTML data of the news article.

[0578] Step 4:

[0579] The server analyzes the HTML data of the retrieved news article and extracts the content of the article.

[0580] Input: Retrieved HTML data

[0581] Output: Extracted article content (text)

[0582] What happens: The server uses the BeautifulSoup library to parse the HTML data and extract the text content of the article.

[0583] Step 5:

[0584] The server inputs the extracted article content into a natural language processing model to generate a summary.

[0585] Input: Extracted article text content

[0586] Output: Summary text

[0587] What it does: The server summarizes the article text using a natural language processing model like spaCy.

[0588] Step 6:

[0589] The server stores the generated summaries in a database.

[0590] Input: Generated summary text

[0591] Output: Summary text stored in the database

[0592] Specific operation: The server stores the generated summary text in a database.

[0593] Step 7:

[0594] A user accesses the server from a terminal on which a dedicated application is installed.

[0595] Input: User request

[0596] Output: A summary list of a particular user's favorite articles

[0597] Specific operation: The terminal sends the user ID to the server and requests related summary information.

[0598] Step 8:

[0599] The server retrieves the latest summary associated with the user ID from the database and sends it to the terminal.

[0600] Input: User ID

[0601] Output: Summarized news article updates

[0602] Specific operation: The server retrieves the relevant summary text from the database and sends it to the user's terminal.

[0603] Step 9:

[0604] The user terminal displays the received summary information on the screen.

[0605] Input: Received summary text

[0606] Output: Summary text displayed on the user's screen

[0607] Specific operation: The terminal displays the received summary text on the user's screen, allowing the user to check the latest news summary.

[0608] 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.

[0609] This invention provides a system that periodically acquires the latest news article information and summarizes it by allowing a user to select a news article and save the URL of the selected news article, as well as a system that recognizes the user's emotions and optimizes the summaries and display order of news articles based on those emotions. By following the format described below, a user can always check the latest news articles that best suit their emotions.

[0610] First, when a user finds an article of interest on a news site, they select the article and click the "Favorite" button. The device (browser) then sends the URL of the news article selected by the user and their user ID to the server.

[0611] The server stores the received URL and user ID in its database, which adds the selected news article to the user's favorites and saves it.

[0612] The server periodically accesses news sites based on the URLs of news articles stored in the database to check for new information. This is done using web scraping technology. The server sends an HTTP request and receives HTML data as a response. The received HTML data is then analyzed to extract the article content (title, body, date, etc.).

[0613] The content of the retrieved news articles is summarized using natural language processing (NLP) technology. The server inputs the retrieved article text into an NLP model to generate a summary. The generated summary is stored in a database. The stored information includes the news article URL, summary content, and the date and time the summary was generated.

[0614] Furthermore, the present invention recognizes the user's emotions by combining an emotion engine. When a user accesses a dedicated web page and browses an article, the emotion engine analyzes the user's emotions based on the user's interactions and access history. This allows the user's emotional state (e.g., excitement, joy, sadness, etc.) to be recognized when reading the article.

[0615] The user's emotion data obtained by the emotion engine is used to adjust the summary content of the news article. For example, if the user's emotion is recognized as sad, the emotion engine can adjust the summary of the news article to emphasize the positive aspects.

[0616] Furthermore, the emotion engine optimizes the display order of news articles based on the user's emotional data. For example, if the user is feeling stressed, the emotion engine will prioritize articles with relaxing content.

[0617] As a concrete example, consider the case where a user adds an article about earthquakes to their favorites. When the user clicks the "Favorite" button, the URL (e.g., https: / / example.com / earthquake-news) is sent to the server and stored in a database. The server periodically accesses this URL and scrapes new information. For example, if new information is added to the article, such as "The magnitude of the earthquake was 6.8, and the damage was minor," the server inputs that information into an NLP model and generates a summary.

[0618] When a user accesses a dedicated web page, the emotion engine recognizes the user's emotions and provides summaries that emphasize positive news and news articles in an optimized viewing order according to the user's emotional state, allowing users to always view the most appropriate news summary that matches their emotions.

[0619] In this way, the present invention builds a system that not only efficiently provides the latest news information, but also realizes a personalized news browsing experience based on the user's emotions.

[0620] The processing flow will be explained below.

[0621] Step 1:

[0622] The user selects a news article and clicks the "Favorite" button. The device (browser) sends the news article URL and user ID to the server.

[0623] Step 2:

[0624] The server receives the request and stores the news article URL and the user ID in a database, which adds the selected news article to the user's favorites.

[0625] Step 3:

[0626] Set up a scheduler so that the server periodically retrieves the list of favorite article URLs from the database. For example, the scheduler can be set to perform scraping every hour.

[0627] Step 4:

[0628] At the scheduled time, the server retrieves the list of URLs from the database, sends an HTTP request to each news site, and receives the HTML response.

[0629] Step 5:

[0630] The server parses the received HTML data and extracts the article content, such as the news article title, body, date, etc. For this, it uses a scraping library (e.g. BeautifulSoup).

[0631] Step 6:

[0632] The server inputs the extracted news article content into a natural language processing (NLP) model, such as a text summarization model like BERT or GPT.

[0633] Step 7:

[0634] The server generates a summary based on the results of the NLP model and stores it in a database. The stored information includes the URL of the news article, the summary content, and the date and time the summary was generated.

[0635] Step 8:

[0636] A user accesses a dedicated web page. The device (browser) sends an HTTP request including the user ID to the server, requesting the latest summaries of favorite articles.

[0637] Step 9:

[0638] The server retrieves summaries of the latest news articles associated with the user ID from the database and returns the retrieved summaries to the terminal as an HTTP response.

[0639] Step 10:

[0640] The terminal (browser) displays the received summary on the user's viewing screen. For example, a summary such as "The magnitude of the earthquake was 6.8, and the damage was minor" may be displayed.

[0641] Step 11:

[0642] While a user is browsing a news article, the device collects the user's emotional data (facial expressions, click patterns, browsing time, etc.).

[0643] Step 12:

[0644] The device sends the collected emotion data to a server where it is analyzed by an emotion engine.

[0645] Step 13:

[0646] The server uses an emotion engine to infer the user's emotional state (e.g., happy, sad, excited, etc.).

[0647] Step 14:

[0648] The server tailors the summary of a news article based on the user's emotional state: for example, if the user is feeling sad, it generates a summary that emphasizes the positive aspects.

[0649] Step 15:

[0650] The server optimizes the display order of news articles based on the user's emotional data, for example, prioritizing articles with relaxing content.

[0651] Step 16:

[0652] When the user accesses the web page again, the server sends the adjusted summary and the optimized display order to the terminal, which then displays it on the user's viewing screen.

[0653] Through the above steps, users can efficiently track the latest updates of their favorite news articles and also check the best news summaries that match their emotions in real time.

[0654] Example 2

[0655] 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."

[0656] Conventional news article delivery systems do not adequately personalize news articles based on the user's emotions, making it difficult to provide optimal news articles based on the user's current emotional state. Furthermore, news article summarization and display order are not optimized, resulting in a poor user experience. There is a need to address these issues and provide personalized news articles based on the user's emotions.

[0657] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0658] In this invention, the server includes means for selecting a news article and saving the URL of the selected news article, means for periodically retrieving news articles based on the URL of the saved news article, means for summarizing the content of the retrieved news article, means for saving the summarized news article content in a database, means for recognizing a user's emotion and adjusting the content of the news article summary based on the emotion, means for optimizing the display order of the news articles based on the user's emotional state, and means for displaying the latest information on the summarized news articles to the user, thereby enabling personalized news article summaries according to the user's emotion and an optimal browsing experience.

[0659] "News article" refers to a description of an event or piece of information published by online or offline media.

[0660] "User" refers to an entity that uses the system to view news articles and obtain information.

[0661] "URL" is an abbreviation for Uniform Resource Locator, and is an address that indicates the location of a resource on the Internet.

[0662] A "server" refers to a computer system that provides data and services to multiple clients over a network.

[0663] A "database" refers to a structured collection of data that allows for efficient storage, retrieval, and management of large amounts of data.

[0664] "Storage methods" refers to the technologies and processes used to store data in a form that can be used for a long period of time.

[0665] "Means of acquisition" refers to the technology and processes used to search and extract stored data or information based on specific criteria.

[0666] "Summarization methods" refers to techniques or processes that extract the main points or essence from long text data and present them in a shortened form.

[0667] "Means for displaying" refers to a display device or interface for visually presenting data or information to a user.

[0668] "Emotion recognition means" refers to techniques and algorithms for detecting and classifying a user's emotional state.

[0669] "Adjustment" refers to a technique or process for optimizing some output based on input data.

[0670] "Optimization means" refers to techniques or processes that maximize the performance or efficiency of a system to achieve a certain goal.

[0671] MODE FOR CARRYING OUT THE INVENTION

[0672] This invention is a system that allows a user to select a news article and save the URL of the selected news article, thereby periodically obtaining the latest news article information, summarizing it, and providing it. Furthermore, the system has the function of recognizing the user's emotions and optimizing the summaries and display order of news articles based on those emotions. A detailed description of an embodiment of this system is provided below.

[0673] User operations

[0674] When a user finds an article of interest on a news site, they select the article and click the "Favorites" button. For example, if a user is interested in an article about earthquakes, they add the URL to their "Favorites." This causes the URL of the article to be sent to the server by the device (browser).

[0675] Terminal (browser) processing

[0676] The terminal (browser) sends the URL of the news article selected by the user and the user ID to the server using an HTTP request. The HTTP protocol is the most commonly used technology at this stage.

[0677] Example of data sent:

[0678] {

[0679] "userID": "user123",

[0680] "articleURL": "https: / / example.com / earthquake-news"

[0681] }

[0682] Server Processing

[0683] The server analyzes the received URL and user ID and stores it in a database. For example, a relational database such as MySQL is commonly used. The format of the stored data is as follows:

[0684] UserID: user123

[0685] ArticleURL: https: / / example.com / earthquake-news

[0686] Check for news article updates

[0687] At regular intervals (for example, every hour), the server accesses the URLs of each news article stored in the database to check for new information. This is done using web scraping techniques, specifically software libraries such as Python's BeautifulSoup and Scrapy.

[0688] Data extraction and summary generation

[0689] The server analyzes the HTML data obtained by web scraping and extracts the article content (title, body text, date, etc.). This text data is summarized using natural language processing (NLP) techniques. Specifically, NLP models such as the BERT model and GPT-3 are used. The summarized article data is then stored back in the database.

[0690] Example of retrieved news article summary:

[0691] An earthquake occurred, with a magnitude of 6.8, but the damage was minor and no injuries were reported.

[0692] Emotion recognition

[0693] When a user accesses a dedicated web page and browses news articles, the emotion engine analyzes the user's interactions and access history. This allows the system to recognize the user's emotional state (e.g., excitement, joy, sadness, etc.). Sentiment analysis algorithms include Python NLP libraries "TextBlob" and "VADER."

[0694] Example of recognized user emotion data:

[0695] {

[0696] "userID": "user123",

[0697] "emotion": "sadness",

[0698] "timestamp": "2023-10-01T12:34:56Z"

[0699] }

[0700] Optimizing news article display

[0701] The emotion engine adjusts the summary content and display order of news articles based on the user's emotion data. For example, if the user feels "sadness," the emotion engine adjusts the summary to emphasize positive summaries and prioritizes relaxing articles.

[0702] Display order example:

[0703] 1. Positive News Summary

[0704] 2. Articles related to user interests

[0705] 3. Other news

[0706] Prompt Sentence Examples

[0707] Here is an example prompt for generating a news article summary using an NLP model:

[0708] Prompt statement:

[0709] "Generate a summary of a news article. The content of the news article is as follows:

[0710] Article content: An earthquake occurred, with shaking of magnitude 6.8 observed. Damage was minor, and no casualties were reported.

[0711] Summary generated: An earthquake occurred, with shaking measured at magnitude 6.8, but the damage was minor and no casualties were reported.

[0712] In this way, the present invention can build a system that not only efficiently provides the latest news articles but also realizes a personalized news browsing experience based on the user's emotions, allowing the user to always see the most appropriate news summary that suits their emotions.

[0713] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0714] Program processing steps

[0715] Step 1:

[0716] A user adds a news article to their "favorites."

[0717] When a user browses a news site, they find an article they are interested in and click its "Favorite" button. The input is the user's click action, and the output is the click event of the "Favorite" button. This operation obtains the article's URL and the user ID.

[0718] Step 2:

[0719] The terminal (browser) sends the URL and user ID to the server.

[0720] The terminal (browser) receives the click event and sends the news article URL and user ID as JSON format data to the server via an HTTP request. The input is the URL and user ID, and the output is the HTTP request.

[0721] Step 3:

[0722] The server stores the URL and user ID in a database.

[0723] The server parses the received data and stores the news article URL and user ID in a database. Specifically, the data is inserted into a relational database (e.g. MySQL). The input is the received data, and the output is a new record in the database.

[0724] Step 4:

[0725] The server periodically checks for updates to news articles.

[0726] The server accesses the URLs of news articles stored in the database according to a scheduled task to check the latest information. This is done using web scraping techniques (e.g. BeautifulSoup, Scrapy). The input is the list of URLs in the database, and the output is the received HTML data.

[0727] Step 5:

[0728] The server parses the HTML data and extracts the article content.

[0729] The server analyzes the HTML data it receives and extracts information such as the article title, body, and date. Specifically, HTML parsing is performed. The input is HTML data, and the output is text data (title, body, date, etc.).

[0730] Step 6:

[0731] The server inputs the article content into a natural language processing (NLP) model to generate a summary.

[0732] The server inputs the extracted article content into an NLP model (e.g., BERT, GPT-3) to generate a summary. The model used is a generative AI model. The input is the extracted text data, and the output is the summary text.

[0733] Step 7:

[0734] The server stores the summaries in a database.

[0735] The server saves the generated summary text, the original URL, and the generation date and time in a database. The input is the summary text and metadata, and the output is a new record in the database.

[0736] Step 8:

[0737] Recognize emotions when users browse news articles.

[0738] While a user accesses a dedicated web page and browses news articles, the emotion engine analyzes the user's interactions and access history. Specifically, it performs emotion analysis using machine learning algorithms (e.g., TextBlob, VADER). The input is user behavior data, and the output is emotion data.

[0739] Step 9:

[0740] The server adjusts the summary content of the news article based on the user's emotional data.

[0741] The emotion engine can use the user's emotion data to adjust the summary of a news article and emphasize its positive aspects. The input is emotion data, and the output is the adjusted summary text.

[0742] Step 10:

[0743] The server optimizes the display order of news articles based on sentiment.

[0744] Optimize the display order of news articles based on emotion data. For example, if sadness is recognized, prioritize positive news articles. The input is emotion data, and the output is an optimized list of news articles.

[0745] Step 11:

[0746] The server displays summarized news article updates to the user.

[0747] When a user accesses a dedicated web page to view a news article, the server retrieves the latest information from the database and displays it on the user's terminal. The input is the user's request, and the output is the information displayed on the terminal.

[0748] (Application example 2)

[0749] 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."

[0750] Conventional news article summarization systems are limited to acquiring and summarizing news articles and do not optimize based on the user's emotions. As a result, they are unable to provide personalized news that adapts to the user's emotional state, resulting in an insufficient improvement in the user experience. Furthermore, even in physical stores, there is a lack of functionality to optimize in-store guidance and recommended products based on customer emotions, limiting the means for increasing customer satisfaction. The present invention aims to solve these problems and provide a system that optimizes news and store guidance based on the user's emotions.

[0751] 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.

[0752] In this invention, the server includes means for selecting a news article and saving the URL of the selected news article, means for periodically retrieving news articles based on the URL of the saved news article, means for summarizing the content of the retrieved news article, means for saving the content of the summarized news article in a database, means for displaying the latest information on the summarized news article to the user, means for recognizing the user's emotion, means for optimizing the summaries and display order of the news articles based on the recognized emotion, and means for recognizing the emotion of a customer in a physical store and optimizing in-store guidance and recommended products based on the emotion. This not only ensures that news articles are always provided in a form that is adapted to the user's emotion, but also enables optimal guidance and recommendations to be made in the physical store in accordance with the customer's emotion, thereby improving the overall user experience and increasing customer satisfaction.

[0753] A "news article" is a report of an event or information published online or in any other media.

[0754] "URL" stands for Uniform Resource Locator and is an address that identifies a specific web page or resource.

[0755] A "summary" is a document that condenses the content of the original news article, extracting only the main information and summarizing it concisely.

[0756] A "database" is a system or software for efficiently storing, retrieving, and managing large amounts of data.

[0757] "User" refers to a person who uses the system to read news articles and receive summarized information.

[0758] "Emotion" refers to the psychological reaction or state that a user expresses when reading an article.

[0759] A "brick and mortar store" refers to a retail store that has a physical presence and where customers can visit and purchase products.

[0760] "Guidance" refers to the information provided to customers in the store and the method of presenting it.

[0761] "Recommended products" are products that a store has specifically selected to encourage customers to purchase.

[0762] "Means of recognizing emotions" refers to technology that determines the psychological state of customers or users from data such as facial expressions and voice.

[0763] "Means of acquiring news articles" refers to the process and technology of acquiring articles from news sites etc. via the Internet.

[0764] "Means for optimizing the display order" refers to techniques and methods for adjusting the order of information provided based on the emotional state of the user or customer.

[0765] This invention is a system that selects news articles, saves the URLs of the selected news articles, periodically obtains the latest information, summarizes them, and provides them to users. It recognizes the user's emotions and optimizes the summaries and display order of news articles based on those emotions, and also optimizes in-store guidance and recommended products based on customer emotions in physical stores. Implementing this invention requires the following steps.

[0766] First, when a user finds an article of interest on a news site, they select the article and click the "Favorites" button. The device sends the URL of the news article selected by the user and their user ID to the server. The server saves the received URL and user ID in a database. This adds the selected news article to the user's favorites and saves it.

[0767] The server then periodically accesses news sites based on the URLs of the news articles stored in the database to check for new information. This process uses web scraping technology. The server sends an HTTP request and receives HTML data as a response. The received HTML data is then parsed to extract the article content (title, body, date, etc.).

[0768] The content of the retrieved news articles is summarized using natural language processing (NLP) technology. The server inputs the retrieved article text into an NLP model to generate a summary. The generated summary is stored in a database. The stored information includes the news article URL, summary content, and the date and time the summary was generated. Python-based libraries (e.g., BeautifulSoup, spaCy, and HuggingFace Transformers) are used to retrieve news articles and generate summaries.

[0769] Furthermore, the present invention recognizes user emotions by incorporating an emotion engine. When a user accesses a dedicated web page and browses articles, the emotion engine analyzes the user's emotions based on the user's interactions and access history. This uses image processing and machine learning models (e.g., OpenCV, Dlib, TensorFlow). This allows the user's emotional state (e.g., excitement, joy, sadness, etc.) to be recognized as they read the articles.

[0770] The user's emotion data acquired by the emotion engine is used to adjust the summary content and display order of news articles. For example, if the user's emotion of sadness is recognized, the emotion engine can adjust the summary of the news article to emphasize the positive aspects. Also, if the user is feeling stressed, articles with content that will help them relax will be displayed preferentially.

[0771] This system can also be applied in physical stores. Using cameras installed in the store and the camera on the customer's smartphone, it recognizes emotions in real time from the customer's facial expressions and voice. The emotion recognition engine determines the customer's emotions and optimizes in-store guidance and recommended products according to those emotions. For example, if a customer is feeling stressed, relaxation products will be recommended first. The selection of recommended products and the order in which guidance information is displayed are also adjusted appropriately.

[0772] An example of an input prompt for a generative AI model is:

[0773] "If the customer's current emotional state is identified as stressed, generate information to recommend relaxation products. Also include information about the latest sales."

[0774] Based on these prompts, the generative AI model can generate stress-relieving products, their features, applicable discount information, and other information, which can be delivered to a display or smart glasses in real time.

[0775] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0776] Step 1:

[0777] When a user finds an article of interest on a news site, the device clicks the "Favorites" button.

[0778] Input: URL of the news article selected by the user, user ID

[0779] Specific operation: When the "Favorites" button is clicked, the device sends the URL of the selected news article and the user ID to the server.

[0780] Output: The URL and user ID are sent to the server.

[0781] Step 2:

[0782] The server stores the received URL and user ID in a database.

[0783] Input: URL of news article, user ID

[0784] Specific operation: The server stores the received URL and user ID in a database, and the selected news article is added to the user's favorites.

[0785] Output: URL and user ID stored in the database

[0786] Step 3:

[0787] The server periodically accesses news sites based on the URLs of news articles stored in the database to check for new information.

[0788] Input: URL stored in the database

[0789] Specific operation: The server sends an HTTP request and receives HTML data as a response. The received HTML data is analyzed to extract the article content (title, body, date, etc.).

[0790] Output: Extracted article content

[0791] Step 4:

[0792] The server inputs the retrieved news article text into a natural language processing (NLP) model to generate a summary.

[0793] Input: Article content (text data)

[0794] What it does: The server uses NLP techniques (e.g., spaCy or HuggingFace Transformers) to summarize the article text.

[0795] Output: Generated summary

[0796] Step 5:

[0797] The server stores the generated summaries in a database.

[0798] Input: Generated summary, news article URL, summary generation date and time

[0799] Specific operation: The server stores the generated summary, the URL of the news article, and the date and time the summary was generated in a database.

[0800] Output: Summary information stored in a database

[0801] Step 6:

[0802] When a user accesses a dedicated web page and reads a news article, the server uses an emotion recognition engine to recognize the user's emotion.

[0803] Input: Real-time facial and voice data of the user

[0804] Specific operation: The server uses image processing and machine learning models (e.g., OpenCV, Dlib, TensorFlow) to determine the user's emotions.

[0805] Output: Recognized emotion data

[0806] Step 7:

[0807] The server adjusts the summary content and display order of news articles based on the recognized emotion data.

[0808] Input: News article summary information, sentiment data

[0809] What it does: The emotion engine uses emotion data to adjust the summary content, highlighting positive aspects and changing the display order.

[0810] Output: Adjusted news article display information

[0811] Step 8:

[0812] The server transmits the adjusted news article display information to the user's terminal, and displays the latest news articles.

[0813] Input: Adjusted news article display information

[0814] Specific operation: Based on the received information, the device provides the user with a summary of the news articles and a display order that reflects the adjusted information.

[0815] Output: News article displayed on user's device

[0816] Step 9:

[0817] In a physical store, the device captures the customer's facial expressions and voice and sends them to an emotion recognition engine.

[0818] Input: Real-time facial and voice data

[0819] Specific operation: Cameras and smartphones in physical stores capture customers' facial expressions and voices and send them to a server.

[0820] Output: Facial expression and voice data sent to the server

[0821] Step 10:

[0822] The server optimizes in-store guidance and recommended products based on the customer's emotional state.

[0823] Input: Customer emotion data, in-store product information

[0824] Specific operation: The emotion recognition engine selects appropriate guidance and recommended products based on emotional data and displays them on the display or smart device.

[0825] Output: Optimized navigation information and recommended products

[0826] Step 11:

[0827] Smart displays in physical stores present emotion-based information.

[0828] Input: Optimized guidance information and recommended product information

[0829] Specific operation: Displays in physical stores display optimized guidance and recommended products to customers based on information received from the server.

[0830] Output: Optimization information displayed on the screen

[0831] Step 12:

[0832] It uses input prompt sentences to the generative AI model to generate that information.

[0833] Input: Prompt (e.g., "If the customer's current emotional state is identified as stressed, generate information to recommend relaxation products. Also include information about the latest sales.")

[0834] What it does: Based on this prompt, the generative AI model generates and delivers appropriate information in real time.

[0835] Output: Generated relaxation product information and special sale information

[0836] 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.

[0837] 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.

[0838] 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.

[0839] [Third embodiment]

[0840] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0841] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0842] 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).

[0843] 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.

[0844] 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.

[0845] 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).

[0846] 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.

[0847] 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.

[0848] 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.

[0849] 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.

[0850] 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.

[0851] 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."

[0852] This invention provides a system that allows users to easily track news articles and obtain the latest information. By following the steps below, users can always check the latest summaries of their favorite news articles.

[0853] First, when a user finds an article of interest on a news site, they select the article and click the "Favorite" button. The device (browser) then sends the URL of the news article selected by the user and their user ID to the server.

[0854] The server stores the received URL and user ID in its database, which adds the selected news article to the user's favorites and saves it.

[0855] Next, the server periodically accesses news sites based on the URLs of the news articles stored in the database to check for new information. This is done using web scraping technology. The server sends an HTTP request and receives HTML data as a response. The received HTML data is then analyzed to extract the article content (title, body, date, etc.).

[0856] The content of the retrieved news articles is summarized using natural language processing (NLP) technology. The server inputs the retrieved article text into an NLP model to generate a summary. The generated summary is stored in a database. The stored information includes the news article URL, summary content, and the date and time the summary was generated.

[0857] When a user accesses a dedicated web page, the device sends the user ID to the server and requests the latest summaries of their favorite articles. The server retrieves summaries of the latest news articles related to the user ID from the database and sends them back to the device. The device then displays the summaries on the user's screen.

[0858] As a concrete example, consider the case where a user adds an article about earthquakes to their favorites. When the user clicks the "Favorite" button, the URL (e.g., https: / / example.com / earthquake-news) is sent to the server and stored in a database. The server periodically accesses this URL and scrapes new information. For example, if new information is added to the article, such as "The magnitude of the earthquake was 6.8, and the damage was minor," the server inputs that information into an NLP model and generates a summary.

[0859] When a user accesses a dedicated web page, the server retrieves the latest news summary and sends it to the user's device, such as "The magnitude of the earthquake is 6.8, and the damage is minor." The device displays this information, allowing the user to easily check the latest news summary.

[0860] In this way, the present invention provides a system that efficiently provides users with up-to-date information on their favorite news articles.

[0861] The processing flow will be explained below.

[0862] Step 1:

[0863] The user selects a news article and clicks the "Favorite" button. The device (browser) sends the news article URL and user ID to the server.

[0864] Step 2:

[0865] The server receives the request and stores the news article URL and the user ID in a database, which adds the selected news article to the user's favorites.

[0866] Step 3:

[0867] The server sets up a scheduler to periodically retrieve the list of favorite article URLs from the database. For example, the scheduler is set to perform scraping processing every hour.

[0868] Step 4:

[0869] At the scheduled time, the server retrieves the list of URLs from the database, sends an HTTP request to each news site, and receives the HTML response.

[0870] Step 5:

[0871] The server parses the received HTML data and extracts the article content, such as the news article title, body, date, etc. For this, it uses a scraping library (e.g. BeautifulSoup).

[0872] Step 6:

[0873] The server inputs the extracted news article content into a natural language processing (NLP) model, such as a text summarization model like BERT or GPT.

[0874] Step 7:

[0875] The server generates a summary based on the results of the NLP model and stores it in a database. The stored information includes the URL of the news article, the summary content, and the date and time the summary was generated.

[0876] Step 8:

[0877] A user accesses a dedicated web page. The device (browser) sends an HTTP request including the user ID to the server, requesting the latest summaries of favorite articles.

[0878] Step 9:

[0879] The server retrieves summaries of the latest news articles associated with the user ID from the database and returns the retrieved summaries to the terminal as an HTTP response.

[0880] Step 10:

[0881] The terminal (browser) displays the received summary on the user's viewing screen. For example, a summary such as "The magnitude of the earthquake was 6.8, and the damage was minor" may be displayed.

[0882] In this way, users can efficiently track and stay up to date on their favorite news articles in real time.

[0883] Example 1

[0884] 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."

[0885] It is difficult to efficiently track articles that interest users from the wide variety of news articles available on the Internet and to obtain the latest information in a timely manner. Especially in today's information-intensive world, it takes time and effort for users to manually check articles regularly and for new information. To solve this problem, a system is needed that can automatically retrieve the latest information for news articles that users have shown interest in, and then summarize and provide them.

[0886] 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.

[0887] In this invention, the server includes means for saving URLs of news articles selected by the user, means for periodically retrieving news articles based on the URLs of the saved news articles, means for summarizing the content of the retrieved news articles using a natural language processing model, means for saving the content of the summarized news articles in a database, and means for displaying the latest information on the summarized news articles to the user, thereby enabling the user to automatically and efficiently obtain the latest information on news articles in which they are interested.

[0888] A "news article" is an information article published on a news site on the Internet.

[0889] A "URL" is an Internet address that uniquely identifies a news article or web page.

[0890] "Means for preservation" refers to a data storage function for recording the URL of a news article and its summarized content.

[0891] "Means of acquisition" refers to the technology used to collect news article data via the Internet.

[0892] A "natural language processing model" is an artificial intelligence technology that automatically summarizes and analyzes text data.

[0893] "Summarization methods" are techniques for shortening the content of news articles and extracting the main points.

[0894] A "database" is a system for efficiently storing, managing, and retrieving organized data.

[0895] "Display means" refers to a technique for visually presenting information to a user's terminal.

[0896] The "Internet" is a communications infrastructure that connects computer networks around the world.

[0897] A "user" is any person or entity that utilizes the invention to track news stories and obtain up-to-date information.

[0898] A "server" is a computer system that processes requests from clients (user terminals) and provides information.

[0899] The present invention provides a system that allows users to easily track and update news articles, allowing users to always see the latest summaries of their favorite news articles.

[0900] First, when a user finds an article of interest on a news site, they select the article and click the "Favorite" button. The device (browser) then sends the URL of the news article selected by the user and their user ID to the server.

[0901] The server stores the received URL and user ID in a database, which adds the selected news article to the user's favorites. The database also stores metadata associated with the article (e.g., title, date).

[0902] Next, the server periodically accesses news sites based on the URLs of the news articles stored in the database to check for new information. This is done using web scraping techniques (e.g., BeautifulSoup, Scrapy). The server sends an HTTP request and receives HTML data as a response. The received HTML data is then parsed to extract the article content (title, body, date, etc.).

[0903] The content of the retrieved news articles is summarized using natural language processing (NLP) technology (e.g., OpenAI's GPT-4). The server inputs the retrieved article text into an NLP model to generate a summary. The generated summary is stored in a database. The stored information includes the news article URL, summary content, and the date and time the summary was generated.

[0904] When a user accesses a dedicated web page, the device sends the user ID to the server and requests the latest summaries of their favorite articles. The server retrieves summaries of the latest news articles related to the user ID from the database and sends them back to the device. The device then displays the summaries on the user's screen.

[0905] As a concrete example, consider the case where a user adds an article about earthquakes to their favorites. When the user clicks the "Favorite" button, the URL (e.g., https: / / example.com / earthquake-news) is sent to the server and stored in a database. The server periodically accesses this URL and scrapes new information. For example, if new information is added to the article, such as "The magnitude of the earthquake was 6.8, and the damage was minor," the server inputs that information into an NLP model and generates a summary.

[0906] When a user accesses a dedicated web page, the server retrieves the latest news summary and sends it to the user's device, such as "The magnitude of the earthquake is 6.8, and the damage is minor." The device displays this information, allowing the user to easily check the latest news summary.

[0907] Example prompt sentence:

[0908] Write a summary of the following news article:

[0909] Title: Earthquake Alert

[0910] Body text: At 10:00 AM on April 1, 2023, a magnitude 6.8 earthquake occurred in a certain area. Damage was minor, and details are currently under investigation.

[0911] In this way, the present invention provides a system that efficiently provides users with up-to-date information on their favorite news articles.

[0912] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0913] Program processing steps

[0914] Step 1:

[0915] Input: A user browses a news site and clicks the "favorite" button on an article that interests them.

[0916] How it works: When a user clicks the "Favorite" button, the device (browser) generates the URL of the selected news article and the user ID in JSON format.

[0917] Output: The generated JSON data is sent to the server as an HTTP POST request.

[0918] Step 2:

[0919] Input: HTTP POST request sent from the device (news article URL and user ID).

[0920] How it works: The server parses the incoming HTTP POST request and extracts the news article URL and the user ID. The server then stores this information in a database. For example, using a SQL database, this would execute an SQL INSERT statement.

[0921] Output: The news article URL and the user ID are saved in the database.

[0922] Step 3:

[0923] Input: A list of URLs for news articles stored on the server.

[0924] How it works: The server runs a periodically scheduled job (e.g., a cron job) that retrieves a list of news article URLs from a database, sends an HTTP GET request to each URL, and parses the resulting HTML.

[0925] Output: The retrieved HTML data is converted into the content of the news article (title, body, date, etc.).

[0926] Step 4:

[0927] Input: The content of the retrieved news article (title, body, date, etc.).

[0928] How it works: The server inputs the text data of the retrieved news article into a natural language processing (NLP) model (e.g., OpenAI's GPT-4). Specifically, it passes the text of the news article to the NLP model as a prompt.

[0929] Output: Receive the summary text generated by the NLP model and obtain it as the summary content.

[0930] Step 5:

[0931] Input: The generated summary text.

[0932] Operation: The server saves the generated summary text to a database. The information to be saved includes the news article URL, summary content, and the date and time the summary was generated. If an SQL database is used, it executes an SQL INSERT or UPDATE statement.

[0933] Output: The summary text and related information are stored in a database.

[0934] Step 6:

[0935] Input: The user accesses a dedicated web page and logs in.

[0936] How it works: When a user logs into a dedicated web page, the device sends an HTTP GET request to the server, including the user ID, to retrieve the latest summaries of the user's favorite news articles.

[0937] Output: A GET request containing the user ID is sent to the server.

[0938] Step 7:

[0939] Input: An HTTP GET request containing a user ID.

[0940] What it does: The server queries the database based on the received user ID to get the latest summaries of relevant news articles, or if using an SQL database, executes a SELECT statement.

[0941] Output: The relevant latest summary text is retrieved from the server.

[0942] Step 8:

[0943] Input: The latest summary text retrieved from the server.

[0944] Operation: The server returns the retrieved summary text to the terminal, which embeds it into HTML for display on the user's screen.

[0945] Output: The summary text is displayed on the user's device, allowing the user to check the latest news information.

[0946] (Application example 1)

[0947] 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."

[0948] It is difficult for users to easily obtain the latest summary information from the vast number of news articles on the Internet. Existing news distribution systems lack the functionality to enable users to quickly and efficiently obtain the information they need, resulting in problems such as missing the latest news and information overload. In particular, tracking the latest information for a specific article is done manually, placing a heavy burden on users. The object of this invention is to build a system that efficiently provides summary information for news articles individually selected by the user, thereby automating the acquisition of the latest information.

[0949] 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.

[0950] In this invention, the server includes means for selecting a news article and saving the URL of the selected news article, means for periodically retrieving news articles based on the URL of the saved news article, means for summarizing the content of the retrieved news article, means for saving the content of the summarized news article in a database, means for displaying the latest information on the summarized news article to a user, means for summarizing the content of the retrieved article using a natural language processing model for generating the summary, and means for notifying the user terminal of the summary information, thereby enabling automatic tracking of the latest information and provision of a summary for a specific news article selected by a user.

[0951] A "news article" is written information about current events provided on the Internet or through other electronic media.

[0952] A "user" is an individual or organization that uses this system to obtain news article information.

[0953] A "URL" is a string of characters that indicates the address on the Internet where a news article is located.

[0954] A "means" is a way or mechanism for performing a particular function or process.

[0955] "Periodic" means to repeat at regular intervals.

[0956] "Storing" means keeping data in a fixed location.

[0957] "Acquire" means to obtain information or data from an external source.

[0958] "To summarize" means to briefly summarize the contents of a news article.

[0959] A "database" is an information system for systematically storing and managing data.

[0960] "Displaying" means providing information visually to a user.

[0961] A "natural language processing model" is a machine learning algorithm for understanding and processing human language.

[0962] "Notify" means to inform a user of specific information.

[0963] To implement this invention, a user first selects a news article and sends the URL of the selected news article from their terminal to a server. The server stores this URL and the user ID in a database. Periodically, the server retrieves news articles via the Internet based on the news article URLs stored in the database. The content of the retrieved news articles is summarized on the server using a natural language processing model. For example, a machine learning algorithm such as spaCy is used as this NLP model. The summarized content of the news article is then stored again in the database.

[0964] When a user accesses the server from a device with a dedicated application installed, the server retrieves the latest summary associated with the user ID from the database and notifies the user's device of that summary information. This allows users to easily access the latest news summaries at any time. Specific hardware components include internet-enabled devices (such as smartphones or smart glasses) and a cloud server. Software components include a database management system (such as SQLite), web scraping technologies (requests, BeautifulSoup), an NLP model (spaCy), and a job scheduling library (apscheduler).

[0965] As a concrete example, consider the case where a user adds a news article about an earthquake to their favorites. When the user enters the URL of the news article and clicks the "Favorite" button, the URL (e.g., https: / / example.com / earthquake-news) is sent to the server and stored in the database. The server periodically accesses this URL and scrapes new information. For example, if new information is added to the news article stating that "the magnitude of the earthquake was 6.8, and the damage was minor," the server inputs this content into a natural language processing model and generates a summary. The summary generated is "The magnitude of the earthquake was 6.8, and the damage was minor."

[0966] When a user accesses the server from a dedicated application, the server retrieves the latest summary and notifies the user. The user can then view the latest summary information on their device screen, giving them a concise understanding of the article's content. An example of a prompt sentence generated using a generative AI model is as follows:

[0967] Prompt: "Generate a summary for the following news article text: 'The earthquake had a magnitude of 6.8 and caused minor damage. This earthquake...'"

[0968] In this way, the present invention provides a system that allows users to keep up to date with specific news articles and easily obtain summary information.

[0969] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0970] Step 1:

[0971] The user selects a news article and clicks the "Favorite" button.

[0972] Input: The URL of the news article selected by the user

[0973] Output: News article URL and user ID

[0974] Specific operation: The device sends the URL of the news article selected by the user and the user ID to the server.

[0975] Step 2:

[0976] The server stores the received URL and user ID in a database.

[0977] Input: Obtained URL and user ID

[0978] Output: URL and user ID stored in the database

[0979] Specific behavior: The server receives the URL and user ID and stores it in the database.

[0980] Step 3:

[0981] The server periodically retrieves news articles based on the URLs stored in the database.

[0982] Input: URL stored in the database

[0983] Output: HTML data of retrieved news articles

[0984] What happens: The server periodically retrieves the URL from the database and sends an HTTP request to retrieve the HTML data of the news article.

[0985] Step 4:

[0986] The server analyzes the HTML data of the retrieved news article and extracts the content of the article.

[0987] Input: Retrieved HTML data

[0988] Output: Extracted article content (text)

[0989] What happens: The server uses the BeautifulSoup library to parse the HTML data and extract the text content of the article.

[0990] Step 5:

[0991] The server inputs the extracted article content into a natural language processing model to generate a summary.

[0992] Input: Extracted article text content

[0993] Output: Summary text

[0994] What it does: The server summarizes the article text using a natural language processing model like spaCy.

[0995] Step 6:

[0996] The server stores the generated summaries in a database.

[0997] Input: Generated summary text

[0998] Output: Summary text stored in the database

[0999] Specific operation: The server stores the generated summary text in a database.

[1000] Step 7:

[1001] A user accesses the server from a terminal on which a dedicated application is installed.

[1002] Input: User request

[1003] Output: A summary list of a particular user's favorite articles

[1004] Specific operation: The terminal sends the user ID to the server and requests related summary information.

[1005] Step 8:

[1006] The server retrieves the latest summary associated with the user ID from the database and sends it to the terminal.

[1007] Input: User ID

[1008] Output: Summarized news article updates

[1009] Specific operation: The server retrieves the relevant summary text from the database and sends it to the user's terminal.

[1010] Step 9:

[1011] The user terminal displays the received summary information on the screen.

[1012] Input: Received summary text

[1013] Output: Summary text displayed on the user's screen

[1014] Specific operation: The terminal displays the received summary text on the user's screen, allowing the user to check the latest news summary.

[1015] 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.

[1016] This invention provides a system that periodically acquires the latest news article information and summarizes it by allowing a user to select a news article and save the URL of the selected news article, as well as a system that recognizes the user's emotions and optimizes the summaries and display order of news articles based on those emotions. By following the format described below, a user can always check the latest news articles that best suit their emotions.

[1017] First, when a user finds an article of interest on a news site, they select the article and click the "Favorite" button. The device (browser) then sends the URL of the news article selected by the user and their user ID to the server.

[1018] The server stores the received URL and user ID in its database, which adds the selected news article to the user's favorites and saves it.

[1019] The server periodically accesses news sites based on the URLs of news articles stored in the database to check for new information. This is done using web scraping technology. The server sends an HTTP request and receives HTML data as a response. The received HTML data is then analyzed to extract the article content (title, body, date, etc.).

[1020] The content of the retrieved news articles is summarized using natural language processing (NLP) technology. The server inputs the retrieved article text into an NLP model to generate a summary. The generated summary is stored in a database. The stored information includes the news article URL, summary content, and the date and time the summary was generated.

[1021] Furthermore, the present invention recognizes the user's emotions by combining an emotion engine. When a user accesses a dedicated web page and browses an article, the emotion engine analyzes the user's emotions based on the user's interactions and access history. This allows the user's emotional state (e.g., excitement, joy, sadness, etc.) to be recognized when reading the article.

[1022] The user's emotion data obtained by the emotion engine is used to adjust the summary content of the news article. For example, if the user's emotion is recognized as sad, the emotion engine can adjust the summary of the news article to emphasize the positive aspects.

[1023] Furthermore, the emotion engine optimizes the display order of news articles based on the user's emotional data. For example, if the user is feeling stressed, the emotion engine will prioritize articles with relaxing content.

[1024] As a concrete example, consider the case where a user adds an article about earthquakes to their favorites. When the user clicks the "Favorite" button, the URL (e.g., https: / / example.com / earthquake-news) is sent to the server and stored in a database. The server periodically accesses this URL and scrapes new information. For example, if new information is added to the article, such as "The magnitude of the earthquake was 6.8, and the damage was minor," the server inputs that information into an NLP model and generates a summary.

[1025] When a user accesses a dedicated web page, the emotion engine recognizes the user's emotions and provides summaries that emphasize positive news and news articles in an optimized viewing order according to the user's emotional state, allowing users to always view the most appropriate news summary that matches their emotions.

[1026] In this way, the present invention builds a system that not only efficiently provides the latest news information, but also realizes a personalized news browsing experience based on the user's emotions.

[1027] The processing flow will be explained below.

[1028] Step 1:

[1029] The user selects a news article and clicks the "Favorite" button. The device (browser) sends the news article URL and user ID to the server.

[1030] Step 2:

[1031] The server receives the request and stores the news article URL and the user ID in a database, which adds the selected news article to the user's favorites.

[1032] Step 3:

[1033] Set up a scheduler so that the server periodically retrieves the list of favorite article URLs from the database. For example, the scheduler can be set to perform scraping every hour.

[1034] Step 4:

[1035] At the scheduled time, the server retrieves the list of URLs from the database, sends an HTTP request to each news site, and receives the HTML response.

[1036] Step 5:

[1037] The server parses the received HTML data and extracts the article content, such as the news article title, body, date, etc. For this, it uses a scraping library (e.g. BeautifulSoup).

[1038] Step 6:

[1039] The server inputs the extracted news article content into a natural language processing (NLP) model, such as a text summarization model like BERT or GPT.

[1040] Step 7:

[1041] The server generates a summary based on the results of the NLP model and stores it in a database. The stored information includes the URL of the news article, the summary content, and the date and time the summary was generated.

[1042] Step 8:

[1043] A user accesses a dedicated web page. The device (browser) sends an HTTP request including the user ID to the server, requesting the latest summaries of favorite articles.

[1044] Step 9:

[1045] The server retrieves summaries of the latest news articles associated with the user ID from the database and returns the retrieved summaries to the terminal as an HTTP response.

[1046] Step 10:

[1047] The terminal (browser) displays the received summary on the user's viewing screen. For example, a summary such as "The magnitude of the earthquake was 6.8, and the damage was minor" may be displayed.

[1048] Step 11:

[1049] While a user is browsing a news article, the device collects the user's emotional data (facial expressions, click patterns, browsing time, etc.).

[1050] Step 12:

[1051] The device sends the collected emotion data to a server where it is analyzed by an emotion engine.

[1052] Step 13:

[1053] The server uses an emotion engine to infer the user's emotional state (e.g., happy, sad, excited, etc.).

[1054] Step 14:

[1055] The server tailors the summary of a news article based on the user's emotional state: for example, if the user is feeling sad, it generates a summary that emphasizes the positive aspects.

[1056] Step 15:

[1057] The server optimizes the display order of news articles based on the user's emotional data, for example, prioritizing articles with relaxing content.

[1058] Step 16:

[1059] When the user accesses the web page again, the server sends the adjusted summary and the optimized display order to the terminal, which then displays it on the user's viewing screen.

[1060] Through the above steps, users can efficiently track the latest updates of their favorite news articles and also check the best news summaries that match their emotions in real time.

[1061] Example 2

[1062] 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."

[1063] Conventional news article delivery systems do not adequately personalize news articles based on the user's emotions, making it difficult to provide optimal news articles based on the user's current emotional state. Furthermore, news article summarization and display order are not optimized, resulting in a poor user experience. There is a need to address these issues and provide personalized news articles based on the user's emotions.

[1064] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1065] In this invention, the server includes means for selecting a news article and saving the URL of the selected news article, means for periodically retrieving news articles based on the URL of the saved news article, means for summarizing the content of the retrieved news article, means for saving the summarized news article content in a database, means for recognizing a user's emotion and adjusting the content of the news article summary based on the emotion, means for optimizing the display order of the news articles based on the user's emotional state, and means for displaying the latest information on the summarized news articles to the user, thereby enabling personalized news article summaries according to the user's emotion and an optimal browsing experience.

[1066] "News article" refers to a description of an event or piece of information published by online or offline media.

[1067] "User" refers to an entity that uses the system to view news articles and obtain information.

[1068] "URL" is an abbreviation for Uniform Resource Locator, and is an address that indicates the location of a resource on the Internet.

[1069] A "server" refers to a computer system that provides data and services to multiple clients over a network.

[1070] A "database" refers to a structured collection of data that allows for efficient storage, retrieval, and management of large amounts of data.

[1071] "Storage methods" refers to the technologies and processes used to store data in a form that can be used for a long period of time.

[1072] "Means of acquisition" refers to the technology and processes used to search and extract stored data or information based on specific criteria.

[1073] "Summarization methods" refers to techniques or processes that extract the main points or essence from long text data and present them in a shortened form.

[1074] "Means for displaying" refers to a display device or interface for visually presenting data or information to a user.

[1075] "Emotion recognition means" refers to techniques and algorithms for detecting and classifying a user's emotional state.

[1076] "Adjustment" refers to a technique or process for optimizing some output based on input data.

[1077] "Optimization means" refers to techniques or processes that maximize the performance or efficiency of a system to achieve a certain goal.

[1078] MODE FOR CARRYING OUT THE INVENTION

[1079] This invention is a system that allows a user to select a news article and save the URL of the selected news article, thereby periodically obtaining the latest news article information, summarizing it, and providing it. Furthermore, the system has the function of recognizing the user's emotions and optimizing the summaries and display order of news articles based on those emotions. A detailed description of an embodiment of this system is provided below.

[1080] User operations

[1081] When a user finds an article of interest on a news site, they select the article and click the "Favorites" button. For example, if a user is interested in an article about earthquakes, they add the URL to their "Favorites." This causes the URL of the article to be sent to the server by the device (browser).

[1082] Terminal (browser) processing

[1083] The terminal (browser) sends the URL of the news article selected by the user and the user ID to the server using an HTTP request. The HTTP protocol is the most commonly used technology at this stage.

[1084] Example of data sent:

[1085] {

[1086] "userID": "user123",

[1087] "articleURL": "https: / / example.com / earthquake-news"

[1088] }

[1089] Server Processing

[1090] The server analyzes the received URL and user ID and stores it in a database. For example, a relational database such as MySQL is commonly used. The format of the stored data is as follows:

[1091] UserID: user123

[1092] ArticleURL: https: / / example.com / earthquake-news

[1093] Check for news article updates

[1094] At regular intervals (for example, every hour), the server accesses the URLs of each news article stored in the database to check for new information. This is done using web scraping techniques, specifically software libraries such as Python's BeautifulSoup and Scrapy.

[1095] Data extraction and summary generation

[1096] The server analyzes the HTML data obtained by web scraping and extracts the article content (title, body text, date, etc.). This text data is summarized using natural language processing (NLP) techniques. Specifically, NLP models such as the BERT model and GPT-3 are used. The summarized article data is then stored back in the database.

[1097] Example of retrieved news article summary:

[1098] An earthquake occurred, with a magnitude of 6.8, but the damage was minor and no injuries were reported.

[1099] Emotion recognition

[1100] When a user accesses a dedicated web page and browses news articles, the emotion engine analyzes the user's interactions and access history. This allows the system to recognize the user's emotional state (e.g., excitement, joy, sadness, etc.). Sentiment analysis algorithms include Python NLP libraries "TextBlob" and "VADER."

[1101] Example of recognized user emotion data:

[1102] {

[1103] "userID": "user123",

[1104] "emotion": "sadness",

[1105] "timestamp": "2023-10-01T12:34:56Z"

[1106] }

[1107] Optimizing news article display

[1108] The emotion engine adjusts the summary content and display order of news articles based on the user's emotion data. For example, if the user feels "sadness," the emotion engine adjusts the summary to emphasize positive summaries and prioritizes relaxing articles.

[1109] Display order example:

[1110] 1. Positive News Summary

[1111] 2. Articles related to user interests

[1112] 3. Other news

[1113] Prompt Sentence Examples

[1114] Here is an example prompt for generating a news article summary using an NLP model:

[1115] Prompt statement:

[1116] "Generate a summary of a news article. The content of the news article is as follows:

[1117] Article content: An earthquake occurred, with shaking of magnitude 6.8 observed. Damage was minor, and no casualties were reported.

[1118] Summary generated: An earthquake occurred, with shaking measured at magnitude 6.8, but the damage was minor and no casualties were reported.

[1119] In this way, the present invention can build a system that not only efficiently provides the latest news articles but also realizes a personalized news browsing experience based on the user's emotions, allowing the user to always see the most appropriate news summary that suits their emotions.

[1120] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1121] Program processing steps

[1122] Step 1:

[1123] A user adds a news article to their "favorites."

[1124] When a user browses a news site, they find an article they are interested in and click its "Favorite" button. The input is the user's click action, and the output is the click event of the "Favorite" button. This operation obtains the article's URL and the user ID.

[1125] Step 2:

[1126] The terminal (browser) sends the URL and user ID to the server.

[1127] The terminal (browser) receives the click event and sends the news article URL and user ID as JSON format data to the server via an HTTP request. The input is the URL and user ID, and the output is the HTTP request.

[1128] Step 3:

[1129] The server stores the URL and user ID in a database.

[1130] The server parses the received data and stores the news article URL and user ID in a database. Specifically, the data is inserted into a relational database (e.g. MySQL). The input is the received data, and the output is a new record in the database.

[1131] Step 4:

[1132] The server periodically checks for updates to news articles.

[1133] The server accesses the URLs of news articles stored in the database according to a scheduled task to check the latest information. This is done using web scraping techniques (e.g. BeautifulSoup, Scrapy). The input is the list of URLs in the database, and the output is the received HTML data.

[1134] Step 5:

[1135] The server parses the HTML data and extracts the article content.

[1136] The server analyzes the HTML data it receives and extracts information such as the article title, body, and date. Specifically, HTML parsing is performed. The input is HTML data, and the output is text data (title, body, date, etc.).

[1137] Step 6:

[1138] The server inputs the article content into a natural language processing (NLP) model to generate a summary.

[1139] The server inputs the extracted article content into an NLP model (e.g., BERT, GPT-3) to generate a summary. The model used is a generative AI model. The input is the extracted text data, and the output is the summary text.

[1140] Step 7:

[1141] The server stores the summaries in a database.

[1142] The server saves the generated summary text, the original URL, and the generation date and time in a database. The input is the summary text and metadata, and the output is a new record in the database.

[1143] Step 8:

[1144] Recognize emotions when users browse news articles.

[1145] While a user accesses a dedicated web page and browses news articles, the emotion engine analyzes the user's interactions and access history. Specifically, it performs emotion analysis using machine learning algorithms (e.g., TextBlob, VADER). The input is user behavior data, and the output is emotion data.

[1146] Step 9:

[1147] The server adjusts the summary content of the news article based on the user's emotional data.

[1148] The emotion engine can use the user's emotion data to adjust the summary of a news article and emphasize its positive aspects. The input is emotion data, and the output is the adjusted summary text.

[1149] Step 10:

[1150] The server optimizes the display order of news articles based on sentiment.

[1151] Optimize the display order of news articles based on emotion data. For example, if sadness is recognized, prioritize positive news articles. The input is emotion data, and the output is an optimized list of news articles.

[1152] Step 11:

[1153] The server displays summarized news article updates to the user.

[1154] When a user accesses a dedicated web page to view a news article, the server retrieves the latest information from the database and displays it on the user's terminal. The input is the user's request, and the output is the information displayed on the terminal.

[1155] (Application example 2)

[1156] 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."

[1157] Conventional news article summarization systems are limited to acquiring and summarizing news articles and do not optimize based on the user's emotions. As a result, they are unable to provide personalized news that adapts to the user's emotional state, resulting in an insufficient improvement in the user experience. Furthermore, even in physical stores, there is a lack of functionality to optimize in-store guidance and recommended products based on customer emotions, limiting the means for increasing customer satisfaction. The present invention aims to solve these problems and provide a system that optimizes news and store guidance based on the user's emotions.

[1158] 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.

[1159] In this invention, the server includes means for selecting a news article and saving the URL of the selected news article, means for periodically retrieving news articles based on the URL of the saved news article, means for summarizing the content of the retrieved news article, means for saving the content of the summarized news article in a database, means for displaying the latest information on the summarized news article to the user, means for recognizing the user's emotion, means for optimizing the summaries and display order of the news articles based on the recognized emotion, and means for recognizing the emotion of a customer in a physical store and optimizing in-store guidance and recommended products based on the emotion. This not only ensures that news articles are always provided in a form that is adapted to the user's emotion, but also enables optimal guidance and recommendations to be made in the physical store in accordance with the customer's emotion, thereby improving the overall user experience and increasing customer satisfaction.

[1160] A "news article" is a report of an event or information published online or in any other media.

[1161] "URL" stands for Uniform Resource Locator and is an address that identifies a specific web page or resource.

[1162] A "summary" is a document that condenses the content of the original news article, extracting only the main information and summarizing it concisely.

[1163] A "database" is a system or software for efficiently storing, retrieving, and managing large amounts of data.

[1164] "User" refers to a person who uses the system to read news articles and receive summarized information.

[1165] "Emotion" refers to the psychological reaction or state that a user expresses when reading an article.

[1166] A "brick and mortar store" refers to a retail store that has a physical presence and where customers can visit and purchase products.

[1167] "Guidance" refers to the information provided to customers in the store and the method of presenting it.

[1168] "Recommended products" are products that a store has specifically selected to encourage customers to purchase.

[1169] "Means of recognizing emotions" refers to technology that determines the psychological state of customers or users from data such as facial expressions and voice.

[1170] "Means of acquiring news articles" refers to the process and technology of acquiring articles from news sites etc. via the Internet.

[1171] "Means for optimizing the display order" refers to techniques and methods for adjusting the order of information provided based on the emotional state of the user or customer.

[1172] This invention is a system that selects news articles, saves the URLs of the selected news articles, periodically obtains the latest information, summarizes them, and provides them to users. It recognizes the user's emotions and optimizes the summaries and display order of news articles based on those emotions, and also optimizes in-store guidance and recommended products based on customer emotions in physical stores. Implementing this invention requires the following steps.

[1173] First, when a user finds an article of interest on a news site, they select the article and click the "Favorites" button. The device sends the URL of the news article selected by the user and their user ID to the server. The server saves the received URL and user ID in a database. This adds the selected news article to the user's favorites and saves it.

[1174] The server then periodically accesses news sites based on the URLs of the news articles stored in the database to check for new information. This process uses web scraping technology. The server sends an HTTP request and receives HTML data as a response. The received HTML data is then parsed to extract the article content (title, body, date, etc.).

[1175] The content of the retrieved news articles is summarized using natural language processing (NLP) technology. The server inputs the retrieved article text into an NLP model to generate a summary. The generated summary is stored in a database. The stored information includes the news article URL, summary content, and the date and time the summary was generated. Python-based libraries (e.g., BeautifulSoup, spaCy, and HuggingFace Transformers) are used to retrieve news articles and generate summaries.

[1176] Furthermore, the present invention recognizes user emotions by incorporating an emotion engine. When a user accesses a dedicated web page and browses articles, the emotion engine analyzes the user's emotions based on the user's interactions and access history. This uses image processing and machine learning models (e.g., OpenCV, Dlib, TensorFlow). This allows the user's emotional state (e.g., excitement, joy, sadness, etc.) to be recognized as they read the articles.

[1177] The user's emotion data acquired by the emotion engine is used to adjust the summary content and display order of news articles. For example, if the user's emotion of sadness is recognized, the emotion engine can adjust the summary of the news article to emphasize the positive aspects. Also, if the user is feeling stressed, articles with content that will help them relax will be displayed preferentially.

[1178] This system can also be applied in physical stores. Using cameras installed in the store and the camera on the customer's smartphone, it recognizes emotions in real time from the customer's facial expressions and voice. The emotion recognition engine determines the customer's emotions and optimizes in-store guidance and recommended products according to those emotions. For example, if a customer is feeling stressed, relaxation products will be recommended first. The selection of recommended products and the order in which guidance information is displayed are also adjusted appropriately.

[1179] An example of an input prompt for a generative AI model is:

[1180] "If the customer's current emotional state is identified as stressed, generate information to recommend relaxation products. Also include information about the latest sales."

[1181] Based on these prompts, the generative AI model can generate stress-relieving products, their features, applicable discount information, and other information, which can be delivered to a display or smart glasses in real time.

[1182] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1183] Step 1:

[1184] When a user finds an article of interest on a news site, the device clicks the "Favorites" button.

[1185] Input: URL of the news article selected by the user, user ID

[1186] Specific operation: When the "Favorites" button is clicked, the device sends the URL of the selected news article and the user ID to the server.

[1187] Output: The URL and user ID are sent to the server.

[1188] Step 2:

[1189] The server stores the received URL and user ID in a database.

[1190] Input: URL of news article, user ID

[1191] Specific operation: The server stores the received URL and user ID in a database, and the selected news article is added to the user's favorites.

[1192] Output: URL and user ID stored in the database

[1193] Step 3:

[1194] The server periodically accesses news sites based on the URLs of news articles stored in the database to check for new information.

[1195] Input: URL stored in the database

[1196] Specific operation: The server sends an HTTP request and receives HTML data as a response. The received HTML data is analyzed to extract the article content (title, body, date, etc.).

[1197] Output: Extracted article content

[1198] Step 4:

[1199] The server inputs the retrieved news article text into a natural language processing (NLP) model to generate a summary.

[1200] Input: Article content (text data)

[1201] What it does: The server uses NLP techniques (e.g., spaCy or HuggingFace Transformers) to summarize the article text.

[1202] Output: Generated summary

[1203] Step 5:

[1204] The server stores the generated summaries in a database.

[1205] Input: Generated summary, news article URL, summary generation date and time

[1206] Specific operation: The server stores the generated summary, the URL of the news article, and the date and time the summary was generated in a database.

[1207] Output: Summary information stored in a database

[1208] Step 6:

[1209] When a user accesses a dedicated web page and reads a news article, the server uses an emotion recognition engine to recognize the user's emotion.

[1210] Input: Real-time facial and voice data of the user

[1211] Specific operation: The server uses image processing and machine learning models (e.g., OpenCV, Dlib, TensorFlow) to determine the user's emotions.

[1212] Output: Recognized emotion data

[1213] Step 7:

[1214] The server adjusts the summary content and display order of news articles based on the recognized emotion data.

[1215] Input: News article summary information, sentiment data

[1216] What it does: The emotion engine uses emotion data to adjust the summary content, highlighting positive aspects and changing the display order.

[1217] Output: Adjusted news article display information

[1218] Step 8:

[1219] The server transmits the adjusted news article display information to the user's terminal, and displays the latest news articles.

[1220] Input: Adjusted news article display information

[1221] Specific operation: Based on the received information, the device provides the user with a summary of the news articles and a display order that reflects the adjusted information.

[1222] Output: News article displayed on user's device

[1223] Step 9:

[1224] In a physical store, the device captures the customer's facial expressions and voice and sends them to an emotion recognition engine.

[1225] Input: Real-time facial and voice data

[1226] Specific operation: Cameras and smartphones in physical stores capture customers' facial expressions and voices and send them to a server.

[1227] Output: Facial expression and voice data sent to the server

[1228] Step 10:

[1229] The server optimizes in-store guidance and recommended products based on the customer's emotional state.

[1230] Input: Customer emotion data, in-store product information

[1231] Specific operation: The emotion recognition engine selects appropriate guidance and recommended products based on emotional data and displays them on the display or smart device.

[1232] Output: Optimized navigation information and recommended products

[1233] Step 11:

[1234] Smart displays in physical stores present emotion-based information.

[1235] Input: Optimized guidance information and recommended product information

[1236] Specific operation: Displays in physical stores display optimized guidance and recommended products to customers based on information received from the server.

[1237] Output: Optimization information displayed on the screen

[1238] Step 12:

[1239] It uses input prompt sentences to the generative AI model to generate that information.

[1240] Input: Prompt (e.g., "If the customer's current emotional state is identified as stressed, generate information to recommend relaxation products. Also include information about the latest sales.")

[1241] What it does: Based on this prompt, the generative AI model generates and delivers appropriate information in real time.

[1242] Output: Generated relaxation product information and special sale information

[1243] 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.

[1244] 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.

[1245] 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.

[1246] [Fourth embodiment]

[1247] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1248] 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.

[1249] 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).

[1250] 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.

[1251] 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.

[1252] 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).

[1253] 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.

[1254] 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.

[1255] 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.

[1256] 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.

[1257] 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.

[1258] 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.

[1259] 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."

[1260] This invention provides a system that allows users to easily track news articles and obtain the latest information. By following the steps below, users can always check the latest summaries of their favorite news articles.

[1261] First, when a user finds an article of interest on a news site, they select the article and click the "Favorite" button. The device (browser) then sends the URL of the news article selected by the user and their user ID to the server.

[1262] The server stores the received URL and user ID in its database, which adds the selected news article to the user's favorites and saves it.

[1263] Next, the server periodically accesses news sites based on the URLs of the news articles stored in the database to check for new information. This is done using web scraping technology. The server sends an HTTP request and receives HTML data as a response. The received HTML data is then analyzed to extract the article content (title, body, date, etc.).

[1264] The content of the retrieved news articles is summarized using natural language processing (NLP) technology. The server inputs the retrieved article text into an NLP model to generate a summary. The generated summary is stored in a database. The stored information includes the news article URL, summary content, and the date and time the summary was generated.

[1265] When a user accesses a dedicated web page, the device sends the user ID to the server and requests the latest summaries of their favorite articles. The server retrieves summaries of the latest news articles related to the user ID from the database and sends them back to the device. The device then displays the summaries on the user's screen.

[1266] As a concrete example, consider the case where a user adds an article about earthquakes to their favorites. When the user clicks the "Favorite" button, the URL (e.g., https: / / example.com / earthquake-news) is sent to the server and stored in a database. The server periodically accesses this URL and scrapes new information. For example, if new information is added to the article, such as "The magnitude of the earthquake was 6.8, and the damage was minor," the server inputs that information into an NLP model and generates a summary.

[1267] When a user accesses a dedicated web page, the server retrieves the latest news summary and sends it to the user's device, such as "The magnitude of the earthquake is 6.8, and the damage is minor." The device displays this information, allowing the user to easily check the latest news summary.

[1268] In this way, the present invention provides a system that efficiently provides users with up-to-date information on their favorite news articles.

[1269] The processing flow will be explained below.

[1270] Step 1:

[1271] The user selects a news article and clicks the "Favorite" button. The device (browser) sends the news article URL and user ID to the server.

[1272] Step 2:

[1273] The server receives the request and stores the news article URL and the user ID in a database, which adds the selected news article to the user's favorites.

[1274] Step 3:

[1275] The server sets up a scheduler to periodically retrieve the list of favorite article URLs from the database. For example, the scheduler is set to perform scraping processing every hour.

[1276] Step 4:

[1277] At the scheduled time, the server retrieves the list of URLs from the database, sends an HTTP request to each news site, and receives the HTML response.

[1278] Step 5:

[1279] The server parses the received HTML data and extracts the article content, such as the news article title, body, date, etc. For this, it uses a scraping library (e.g. BeautifulSoup).

[1280] Step 6:

[1281] The server inputs the extracted news article content into a natural language processing (NLP) model, such as a text summarization model like BERT or GPT.

[1282] Step 7:

[1283] The server generates a summary based on the results of the NLP model and stores it in a database. The stored information includes the URL of the news article, the summary content, and the date and time the summary was generated.

[1284] Step 8:

[1285] A user accesses a dedicated web page. The device (browser) sends an HTTP request including the user ID to the server, requesting the latest summaries of favorite articles.

[1286] Step 9:

[1287] The server retrieves summaries of the latest news articles associated with the user ID from the database and returns the retrieved summaries to the terminal as an HTTP response.

[1288] Step 10:

[1289] The terminal (browser) displays the received summary on the user's viewing screen. For example, a summary such as "The magnitude of the earthquake was 6.8, and the damage was minor" may be displayed.

[1290] In this way, users can efficiently track and stay up to date on their favorite news articles in real time.

[1291] Example 1

[1292] 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."

[1293] It is difficult to efficiently track articles that interest users from the wide variety of news articles available on the Internet and to obtain the latest information in a timely manner. Especially in today's information-intensive world, it takes time and effort for users to manually check articles regularly and for new information. To solve this problem, a system is needed that can automatically retrieve the latest information for news articles that users have shown interest in, and then summarize and provide them.

[1294] 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.

[1295] In this invention, the server includes means for saving URLs of news articles selected by the user, means for periodically retrieving news articles based on the URLs of the saved news articles, means for summarizing the content of the retrieved news articles using a natural language processing model, means for saving the content of the summarized news articles in a database, and means for displaying the latest information on the summarized news articles to the user, thereby enabling the user to automatically and efficiently obtain the latest information on news articles in which they are interested.

[1296] A "news article" is an information article published on a news site on the Internet.

[1297] A "URL" is an Internet address that uniquely identifies a news article or web page.

[1298] "Means for preservation" refers to a data storage function for recording the URL of a news article and its summarized content.

[1299] "Means of acquisition" refers to the technology used to collect news article data via the Internet.

[1300] A "natural language processing model" is an artificial intelligence technology that automatically summarizes and analyzes text data.

[1301] "Summarization methods" are techniques for shortening the content of news articles and extracting the main points.

[1302] A "database" is a system for efficiently storing, managing, and retrieving organized data.

[1303] "Display means" refers to a technique for visually presenting information to a user's terminal.

[1304] The "Internet" is a communications infrastructure that connects computer networks around the world.

[1305] A "user" is any person or entity that utilizes the invention to track news stories and obtain up-to-date information.

[1306] A "server" is a computer system that processes requests from clients (user terminals) and provides information.

[1307] The present invention provides a system that allows users to easily track and update news articles, allowing users to always see the latest summaries of their favorite news articles.

[1308] First, when a user finds an article of interest on a news site, they select the article and click the "Favorite" button. The device (browser) then sends the URL of the news article selected by the user and their user ID to the server.

[1309] The server stores the received URL and user ID in a database, which adds the selected news article to the user's favorites. The database also stores metadata associated with the article (e.g., title, date).

[1310] Next, the server periodically accesses news sites based on the URLs of the news articles stored in the database to check for new information. This is done using web scraping techniques (e.g., BeautifulSoup, Scrapy). The server sends an HTTP request and receives HTML data as a response. The received HTML data is then parsed to extract the article content (title, body, date, etc.).

[1311] The content of the retrieved news articles is summarized using natural language processing (NLP) technology (e.g., OpenAI's GPT-4). The server inputs the retrieved article text into an NLP model to generate a summary. The generated summary is stored in a database. The stored information includes the news article URL, summary content, and the date and time the summary was generated.

[1312] When a user accesses a dedicated web page, the device sends the user ID to the server and requests the latest summaries of their favorite articles. The server retrieves summaries of the latest news articles related to the user ID from the database and sends them back to the device. The device then displays the summaries on the user's screen.

[1313] As a concrete example, consider the case where a user adds an article about earthquakes to their favorites. When the user clicks the "Favorite" button, the URL (e.g., https: / / example.com / earthquake-news) is sent to the server and stored in a database. The server periodically accesses this URL and scrapes new information. For example, if new information is added to the article, such as "The magnitude of the earthquake was 6.8, and the damage was minor," the server inputs that information into an NLP model and generates a summary.

[1314] When a user accesses a dedicated web page, the server retrieves the latest news summary and sends it to the user's device, such as "The magnitude of the earthquake is 6.8, and the damage is minor." The device displays this information, allowing the user to easily check the latest news summary.

[1315] Example prompt sentence:

[1316] Write a summary of the following news article:

[1317] Title: Earthquake Alert

[1318] Body text: At 10:00 AM on April 1, 2023, a magnitude 6.8 earthquake occurred in a certain area. Damage was minor, and details are currently under investigation.

[1319] In this way, the present invention provides a system that efficiently provides users with up-to-date information on their favorite news articles.

[1320] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1321] Program processing steps

[1322] Step 1:

[1323] Input: A user browses a news site and clicks the "favorite" button on an article that interests them.

[1324] How it works: When a user clicks the "Favorite" button, the device (browser) generates the URL of the selected news article and the user ID in JSON format.

[1325] Output: The generated JSON data is sent to the server as an HTTP POST request.

[1326] Step 2:

[1327] Input: HTTP POST request sent from the device (news article URL and user ID).

[1328] How it works: The server parses the incoming HTTP POST request and extracts the news article URL and the user ID. The server then stores this information in a database. For example, using a SQL database, this would execute an SQL INSERT statement.

[1329] Output: The news article URL and the user ID are saved in the database.

[1330] Step 3:

[1331] Input: A list of URLs for news articles stored on the server.

[1332] How it works: The server runs a periodically scheduled job (e.g., a cron job) that retrieves a list of news article URLs from a database, sends an HTTP GET request to each URL, and parses the resulting HTML.

[1333] Output: The retrieved HTML data is converted into the content of the news article (title, body, date, etc.).

[1334] Step 4:

[1335] Input: The content of the retrieved news article (title, body, date, etc.).

[1336] How it works: The server inputs the text data of the retrieved news article into a natural language processing (NLP) model (e.g., OpenAI's GPT-4). Specifically, it passes the text of the news article to the NLP model as a prompt.

[1337] Output: Receive the summary text generated by the NLP model and obtain it as the summary content.

[1338] Step 5:

[1339] Input: The generated summary text.

[1340] Operation: The server saves the generated summary text to a database. The information to be saved includes the news article URL, summary content, and the date and time the summary was generated. If an SQL database is used, it executes an SQL INSERT or UPDATE statement.

[1341] Output: The summary text and related information are stored in a database.

[1342] Step 6:

[1343] Input: The user accesses a dedicated web page and logs in.

[1344] How it works: When a user logs into a dedicated web page, the device sends an HTTP GET request to the server, including the user ID, to retrieve the latest summaries of the user's favorite news articles.

[1345] Output: A GET request containing the user ID is sent to the server.

[1346] Step 7:

[1347] Input: An HTTP GET request containing a user ID.

[1348] What it does: The server queries the database based on the received user ID to get the latest summaries of relevant news articles, or if using an SQL database, executes a SELECT statement.

[1349] Output: The relevant latest summary text is retrieved from the server.

[1350] Step 8:

[1351] Input: The latest summary text retrieved from the server.

[1352] Operation: The server returns the retrieved summary text to the terminal, which embeds it into HTML for display on the user's screen.

[1353] Output: The summary text is displayed on the user's device, allowing the user to check the latest news information.

[1354] (Application example 1)

[1355] 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."

[1356] It is difficult for users to easily obtain the latest summary information from the vast number of news articles on the Internet. Existing news distribution systems lack the functionality to enable users to quickly and efficiently obtain the information they need, resulting in problems such as missing the latest news and information overload. In particular, tracking the latest information for a specific article is done manually, placing a heavy burden on users. The object of this invention is to build a system that efficiently provides summary information for news articles individually selected by the user, thereby automating the acquisition of the latest information.

[1357] 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.

[1358] In this invention, the server includes means for selecting a news article and saving the URL of the selected news article, means for periodically retrieving news articles based on the URL of the saved news article, means for summarizing the content of the retrieved news article, means for saving the content of the summarized news article in a database, means for displaying the latest information on the summarized news article to a user, means for summarizing the content of the retrieved article using a natural language processing model for generating the summary, and means for notifying the user terminal of the summary information, thereby enabling automatic tracking of the latest information and provision of a summary for a specific news article selected by a user.

[1359] A "news article" is written information about current events provided on the Internet or through other electronic media.

[1360] A "user" is an individual or organization that uses this system to obtain news article information.

[1361] A "URL" is a string of characters that indicates the address on the Internet where a news article is located.

[1362] A "means" is a way or mechanism for performing a particular function or process.

[1363] "Periodic" means to repeat at regular intervals.

[1364] "Storing" means keeping data in a fixed location.

[1365] "Acquire" means to obtain information or data from an external source.

[1366] "To summarize" means to briefly summarize the contents of a news article.

[1367] A "database" is an information system for systematically storing and managing data.

[1368] "Displaying" means providing information visually to a user.

[1369] A "natural language processing model" is a machine learning algorithm for understanding and processing human language.

[1370] "Notify" means to inform a user of specific information.

[1371] To implement this invention, a user first selects a news article and sends the URL of the selected news article from their terminal to a server. The server stores this URL and the user ID in a database. Periodically, the server retrieves news articles via the Internet based on the news article URLs stored in the database. The content of the retrieved news articles is summarized on the server using a natural language processing model. For example, a machine learning algorithm such as spaCy is used as this NLP model. The summarized content of the news article is then stored again in the database.

[1372] When a user accesses the server from a device with a dedicated application installed, the server retrieves the latest summary associated with the user ID from the database and notifies the user's device of that summary information. This allows users to easily access the latest news summaries at any time. Specific hardware components include internet-enabled devices (such as smartphones or smart glasses) and a cloud server. Software components include a database management system (such as SQLite), web scraping technologies (requests, BeautifulSoup), an NLP model (spaCy), and a job scheduling library (apscheduler).

[1373] As a concrete example, consider the case where a user adds a news article about an earthquake to their favorites. When the user enters the URL of the news article and clicks the "Favorite" button, the URL (e.g., https: / / example.com / earthquake-news) is sent to the server and stored in the database. The server periodically accesses this URL and scrapes new information. For example, if new information is added to the news article stating that "the magnitude of the earthquake was 6.8, and the damage was minor," the server inputs this content into a natural language processing model and generates a summary. The summary generated is "The magnitude of the earthquake was 6.8, and the damage was minor."

[1374] When a user accesses the server from a dedicated application, the server retrieves the latest summary and notifies the user. The user can then view the latest summary information on their device screen, giving them a concise understanding of the article's content. An example of a prompt sentence generated using a generative AI model is as follows:

[1375] Prompt: "Generate a summary for the following news article text: 'The earthquake had a magnitude of 6.8 and caused minor damage. This earthquake...'"

[1376] In this way, the present invention provides a system that allows users to keep up to date with specific news articles and easily obtain summary information.

[1377] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1378] Step 1:

[1379] The user selects a news article and clicks the "Favorite" button.

[1380] Input: The URL of the news article selected by the user

[1381] Output: News article URL and user ID

[1382] Specific operation: The device sends the URL of the news article selected by the user and the user ID to the server.

[1383] Step 2:

[1384] The server stores the received URL and user ID in a database.

[1385] Input: Obtained URL and user ID

[1386] Output: URL and user ID stored in the database

[1387] Specific behavior: The server receives the URL and user ID and stores it in the database.

[1388] Step 3:

[1389] The server periodically retrieves news articles based on the URLs stored in the database.

[1390] Input: URL stored in the database

[1391] Output: HTML data of retrieved news articles

[1392] What happens: The server periodically retrieves the URL from the database and sends an HTTP request to retrieve the HTML data of the news article.

[1393] Step 4:

[1394] The server analyzes the HTML data of the retrieved news article and extracts the content of the article.

[1395] Input: Retrieved HTML data

[1396] Output: Extracted article content (text)

[1397] What happens: The server uses the BeautifulSoup library to parse the HTML data and extract the text content of the article.

[1398] Step 5:

[1399] The server inputs the extracted article content into a natural language processing model to generate a summary.

[1400] Input: Extracted article text content

[1401] Output: Summary text

[1402] What it does: The server summarizes the article text using a natural language processing model like spaCy.

[1403] Step 6:

[1404] The server stores the generated summaries in a database.

[1405] Input: Generated summary text

[1406] Output: Summary text stored in the database

[1407] Specific operation: The server stores the generated summary text in a database.

[1408] Step 7:

[1409] A user accesses the server from a terminal on which a dedicated application is installed.

[1410] Input: User request

[1411] Output: A summary list of a particular user's favorite articles

[1412] Specific operation: The terminal sends the user ID to the server and requests related summary information.

[1413] Step 8:

[1414] The server retrieves the latest summary associated with the user ID from the database and sends it to the terminal.

[1415] Input: User ID

[1416] Output: Summarized news article updates

[1417] Specific operation: The server retrieves the relevant summary text from the database and sends it to the user's terminal.

[1418] Step 9:

[1419] The user terminal displays the received summary information on the screen.

[1420] Input: Received summary text

[1421] Output: Summary text displayed on the user's screen

[1422] Specific operation: The terminal displays the received summary text on the user's screen, allowing the user to check the latest news summary.

[1423] 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.

[1424] This invention provides a system that periodically acquires the latest news article information and summarizes it by allowing a user to select a news article and save the URL of the selected news article, as well as a system that recognizes the user's emotions and optimizes the summaries and display order of news articles based on those emotions. By following the format described below, a user can always check the latest news articles that best suit their emotions.

[1425] First, when a user finds an article of interest on a news site, they select the article and click the "Favorite" button. The device (browser) then sends the URL of the news article selected by the user and their user ID to the server.

[1426] The server stores the received URL and user ID in its database, which adds the selected news article to the user's favorites and saves it.

[1427] The server periodically accesses news sites based on the URLs of news articles stored in the database to check for new information. This is done using web scraping technology. The server sends an HTTP request and receives HTML data as a response. The received HTML data is then analyzed to extract the article content (title, body, date, etc.).

[1428] The content of the retrieved news articles is summarized using natural language processing (NLP) technology. The server inputs the retrieved article text into an NLP model to generate a summary. The generated summary is stored in a database. The stored information includes the news article URL, summary content, and the date and time the summary was generated.

[1429] Furthermore, the present invention recognizes the user's emotions by combining an emotion engine. When a user accesses a dedicated web page and browses an article, the emotion engine analyzes the user's emotions based on the user's interactions and access history. This allows the user's emotional state (e.g., excitement, joy, sadness, etc.) to be recognized when reading the article.

[1430] The user's emotion data obtained by the emotion engine is used to adjust the summary content of the news article. For example, if the user's emotion is recognized as sad, the emotion engine can adjust the summary of the news article to emphasize the positive aspects.

[1431] Furthermore, the emotion engine optimizes the display order of news articles based on the user's emotional data. For example, if the user is feeling stressed, the emotion engine will prioritize articles with relaxing content.

[1432] As a concrete example, consider the case where a user adds an article about earthquakes to their favorites. When the user clicks the "Favorite" button, the URL (e.g., https: / / example.com / earthquake-news) is sent to the server and stored in a database. The server periodically accesses this URL and scrapes new information. For example, if new information is added to the article, such as "The magnitude of the earthquake was 6.8, and the damage was minor," the server inputs that information into an NLP model and generates a summary.

[1433] When a user accesses a dedicated web page, the emotion engine recognizes the user's emotions and provides summaries that emphasize positive news and news articles in an optimized viewing order according to the user's emotional state, allowing users to always view the most appropriate news summary that matches their emotions.

[1434] In this way, the present invention builds a system that not only efficiently provides the latest news information, but also realizes a personalized news browsing experience based on the user's emotions.

[1435] The processing flow will be explained below.

[1436] Step 1:

[1437] The user selects a news article and clicks the "Favorite" button. The device (browser) sends the news article URL and user ID to the server.

[1438] Step 2:

[1439] The server receives the request and stores the news article URL and the user ID in a database, which adds the selected news article to the user's favorites.

[1440] Step 3:

[1441] Set up a scheduler so that the server periodically retrieves the list of favorite article URLs from the database. For example, the scheduler can be set to perform scraping every hour.

[1442] Step 4:

[1443] At the scheduled time, the server retrieves the list of URLs from the database, sends an HTTP request to each news site, and receives the HTML response.

[1444] Step 5:

[1445] The server parses the received HTML data and extracts the article content, such as the news article title, body, date, etc. For this, it uses a scraping library (e.g. BeautifulSoup).

[1446] Step 6:

[1447] The server inputs the extracted news article content into a natural language processing (NLP) model, such as a text summarization model like BERT or GPT.

[1448] Step 7:

[1449] The server generates a summary based on the results of the NLP model and stores it in a database. The stored information includes the URL of the news article, the summary content, and the date and time the summary was generated.

[1450] Step 8:

[1451] A user accesses a dedicated web page. The device (browser) sends an HTTP request including the user ID to the server, requesting the latest summaries of favorite articles.

[1452] Step 9:

[1453] The server retrieves summaries of the latest news articles associated with the user ID from the database and returns the retrieved summaries to the terminal as an HTTP response.

[1454] Step 10:

[1455] The terminal (browser) displays the received summary on the user's viewing screen. For example, a summary such as "The magnitude of the earthquake was 6.8, and the damage was minor" may be displayed.

[1456] Step 11:

[1457] While a user is browsing a news article, the device collects the user's emotional data (facial expressions, click patterns, browsing time, etc.).

[1458] Step 12:

[1459] The device sends the collected emotion data to a server where it is analyzed by an emotion engine.

[1460] Step 13:

[1461] The server uses an emotion engine to infer the user's emotional state (e.g., happy, sad, excited, etc.).

[1462] Step 14:

[1463] The server tailors the summary of a news article based on the user's emotional state: for example, if the user is feeling sad, it generates a summary that emphasizes the positive aspects.

[1464] Step 15:

[1465] The server optimizes the display order of news articles based on the user's emotional data, for example, prioritizing articles with relaxing content.

[1466] Step 16:

[1467] When the user accesses the web page again, the server sends the adjusted summary and the optimized display order to the terminal, which then displays it on the user's viewing screen.

[1468] Through the above steps, users can efficiently track the latest updates of their favorite news articles and also check the best news summaries that match their emotions in real time.

[1469] Example 2

[1470] 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."

[1471] Conventional news article delivery systems do not adequately personalize news articles based on the user's emotions, making it difficult to provide optimal news articles based on the user's current emotional state. Furthermore, news article summarization and display order are not optimized, resulting in a poor user experience. There is a need to address these issues and provide personalized news articles based on the user's emotions.

[1472] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1473] In this invention, the server includes means for selecting a news article and saving the URL of the selected news article, means for periodically retrieving news articles based on the URL of the saved news article, means for summarizing the content of the retrieved news article, means for saving the summarized news article content in a database, means for recognizing a user's emotion and adjusting the content of the news article summary based on the emotion, means for optimizing the display order of the news articles based on the user's emotional state, and means for displaying the latest information on the summarized news articles to the user, thereby enabling personalized news article summaries according to the user's emotion and an optimal browsing experience.

[1474] "News article" refers to a description of an event or piece of information published by online or offline media.

[1475] "User" refers to an entity that uses the system to view news articles and obtain information.

[1476] "URL" is an abbreviation for Uniform Resource Locator, and is an address that indicates the location of a resource on the Internet.

[1477] A "server" refers to a computer system that provides data and services to multiple clients over a network.

[1478] A "database" refers to a structured collection of data that allows for efficient storage, retrieval, and management of large amounts of data.

[1479] "Storage methods" refers to the technologies and processes used to store data in a form that can be used for a long period of time.

[1480] "Means of acquisition" refers to the technology and processes used to search and extract stored data or information based on specific criteria.

[1481] "Summarization methods" refers to techniques or processes that extract the main points or essence from long text data and present them in a shortened form.

[1482] "Means for displaying" refers to a display device or interface for visually presenting data or information to a user.

[1483] "Emotion recognition means" refers to techniques and algorithms for detecting and classifying a user's emotional state.

[1484] "Adjustment" refers to a technique or process for optimizing some output based on input data.

[1485] "Optimization means" refers to techniques or processes that maximize the performance or efficiency of a system to achieve a certain goal.

[1486] MODE FOR CARRYING OUT THE INVENTION

[1487] This invention is a system that allows a user to select a news article and save the URL of the selected news article, thereby periodically obtaining the latest news article information, summarizing it, and providing it. Furthermore, the system has the function of recognizing the user's emotions and optimizing the summaries and display order of news articles based on those emotions. A detailed description of an embodiment of this system is provided below.

[1488] User operations

[1489] When a user finds an article of interest on a news site, they select the article and click the "Favorites" button. For example, if a user is interested in an article about earthquakes, they add the URL to their "Favorites." This causes the URL of the article to be sent to the server by the device (browser).

[1490] Terminal (browser) processing

[1491] The terminal (browser) sends the URL of the news article selected by the user and the user ID to the server using an HTTP request. The HTTP protocol is the most commonly used technology at this stage.

[1492] Example of data sent:

[1493] {

[1494] "userID": "user123",

[1495] "articleURL": "https: / / example.com / earthquake-news"

[1496] }

[1497] Server Processing

[1498] The server analyzes the received URL and user ID and stores it in a database. For example, a relational database such as MySQL is commonly used. The format of the stored data is as follows:

[1499] UserID: user123

[1500] ArticleURL: https: / / example.com / earthquake-news

[1501] Check for news article updates

[1502] At regular intervals (for example, every hour), the server accesses the URLs of each news article stored in the database to check for new information. This is done using web scraping techniques, specifically software libraries such as Python's BeautifulSoup and Scrapy.

[1503] Data extraction and summary generation

[1504] The server analyzes the HTML data obtained by web scraping and extracts the article content (title, body text, date, etc.). This text data is summarized using natural language processing (NLP) techniques. Specifically, NLP models such as the BERT model and GPT-3 are used. The summarized article data is then stored back in the database.

[1505] Example of retrieved news article summary:

[1506] An earthquake occurred, with a magnitude of 6.8, but the damage was minor and no injuries were reported.

[1507] Emotion recognition

[1508] When a user accesses a dedicated web page and browses news articles, the emotion engine analyzes the user's interactions and access history. This allows the system to recognize the user's emotional state (e.g., excitement, joy, sadness, etc.). Sentiment analysis algorithms include Python NLP libraries "TextBlob" and "VADER."

[1509] Example of recognized user emotion data:

[1510] {

[1511] "userID": "user123",

[1512] "emotion": "sadness",

[1513] "timestamp": "2023-10-01T12:34:56Z"

[1514] }

[1515] Optimizing news article display

[1516] The emotion engine adjusts the summary content and display order of news articles based on the user's emotion data. For example, if the user feels "sadness," the emotion engine adjusts the summary to emphasize positive summaries and prioritizes relaxing articles.

[1517] Display order example:

[1518] 1. Positive News Summary

[1519] 2. Articles related to user interests

[1520] 3. Other news

[1521] Prompt Sentence Examples

[1522] Here is an example prompt for generating a news article summary using an NLP model:

[1523] Prompt statement:

[1524] "Generate a summary of a news article. The content of the news article is as follows:

[1525] Article content: An earthquake occurred, with shaking of magnitude 6.8 observed. Damage was minor, and no casualties were reported.

[1526] Summary generated: An earthquake occurred, with shaking measured at magnitude 6.8, but the damage was minor and no casualties were reported.

[1527] In this way, the present invention can build a system that not only efficiently provides the latest news articles but also realizes a personalized news browsing experience based on the user's emotions, allowing the user to always see the most appropriate news summary that suits their emotions.

[1528] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1529] Program processing steps

[1530] Step 1:

[1531] A user adds a news article to their "favorites."

[1532] When a user browses a news site, they find an article they are interested in and click its "Favorite" button. The input is the user's click action, and the output is the click event of the "Favorite" button. This operation obtains the article's URL and the user ID.

[1533] Step 2:

[1534] The terminal (browser) sends the URL and user ID to the server.

[1535] The terminal (browser) receives the click event and sends the news article URL and user ID as JSON format data to the server via an HTTP request. The input is the URL and user ID, and the output is the HTTP request.

[1536] Step 3:

[1537] The server stores the URL and user ID in a database.

[1538] The server parses the received data and stores the news article URL and user ID in a database. Specifically, the data is inserted into a relational database (e.g. MySQL). The input is the received data, and the output is a new record in the database.

[1539] Step 4:

[1540] The server periodically checks for updates to news articles.

[1541] The server accesses the URLs of news articles stored in the database according to a scheduled task to check the latest information. This is done using web scraping techniques (e.g. BeautifulSoup, Scrapy). The input is the list of URLs in the database, and the output is the received HTML data.

[1542] Step 5:

[1543] The server parses the HTML data and extracts the article content.

[1544] The server analyzes the HTML data it receives and extracts information such as the article title, body, and date. Specifically, HTML parsing is performed. The input is HTML data, and the output is text data (title, body, date, etc.).

[1545] Step 6:

[1546] The server inputs the article content into a natural language processing (NLP) model to generate a summary.

[1547] The server inputs the extracted article content into an NLP model (e.g., BERT, GPT-3) to generate a summary. The model used is a generative AI model. The input is the extracted text data, and the output is the summary text.

[1548] Step 7:

[1549] The server stores the summaries in a database.

[1550] The server saves the generated summary text, the original URL, and the generation date and time in a database. The input is the summary text and metadata, and the output is a new record in the database.

[1551] Step 8:

[1552] Recognize emotions when users browse news articles.

[1553] While a user accesses a dedicated web page and browses news articles, the emotion engine analyzes the user's interactions and access history. Specifically, it performs emotion analysis using machine learning algorithms (e.g., TextBlob, VADER). The input is user behavior data, and the output is emotion data.

[1554] Step 9:

[1555] The server adjusts the summary content of the news article based on the user's emotional data.

[1556] The emotion engine can use the user's emotion data to adjust the summary of a news article and emphasize its positive aspects. The input is emotion data, and the output is the adjusted summary text.

[1557] Step 10:

[1558] The server optimizes the display order of news articles based on sentiment.

[1559] Optimize the display order of news articles based on emotion data. For example, if sadness is recognized, prioritize positive news articles. The input is emotion data, and the output is an optimized list of news articles.

[1560] Step 11:

[1561] The server displays summarized news article updates to the user.

[1562] When a user accesses a dedicated web page to view a news article, the server retrieves the latest information from the database and displays it on the user's terminal. The input is the user's request, and the output is the information displayed on the terminal.

[1563] (Application example 2)

[1564] 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."

[1565] Conventional news article summarization systems are limited to acquiring and summarizing news articles and do not optimize based on the user's emotions. As a result, they are unable to provide personalized news that adapts to the user's emotional state, resulting in an insufficient improvement in the user experience. Furthermore, even in physical stores, there is a lack of functionality to optimize in-store guidance and recommended products based on customer emotions, limiting the means for increasing customer satisfaction. The present invention aims to solve these problems and provide a system that optimizes news and store guidance based on the user's emotions.

[1566] 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.

[1567] In this invention, the server includes means for selecting a news article and saving the URL of the selected news article, means for periodically retrieving news articles based on the URL of the saved news article, means for summarizing the content of the retrieved news article, means for saving the content of the summarized news article in a database, means for displaying the latest information on the summarized news article to the user, means for recognizing the user's emotion, means for optimizing the summaries and display order of the news articles based on the recognized emotion, and means for recognizing the emotion of a customer in a physical store and optimizing in-store guidance and recommended products based on the emotion. This not only ensures that news articles are always provided in a form that is adapted to the user's emotion, but also enables optimal guidance and recommendations to be made in the physical store in accordance with the customer's emotion, thereby improving the overall user experience and increasing customer satisfaction.

[1568] A "news article" is a report of an event or information published online or in any other media.

[1569] "URL" stands for Uniform Resource Locator and is an address that identifies a specific web page or resource.

[1570] A "summary" is a document that condenses the content of the original news article, extracting only the main information and summarizing it concisely.

[1571] A "database" is a system or software for efficiently storing, retrieving, and managing large amounts of data.

[1572] "User" refers to a person who uses the system to read news articles and receive summarized information.

[1573] "Emotion" refers to the psychological reaction or state that a user expresses when reading an article.

[1574] A "brick and mortar store" refers to a retail store that has a physical presence and where customers can visit and purchase products.

[1575] "Guidance" refers to the information provided to customers in the store and the method of presenting it.

[1576] "Recommended products" are products that a store has specifically selected to encourage customers to purchase.

[1577] "Means of recognizing emotions" refers to technology that determines the psychological state of customers or users from data such as facial expressions and voice.

[1578] "Means of acquiring news articles" refers to the process and technology of acquiring articles from news sites etc. via the Internet.

[1579] "Means for optimizing the display order" refers to techniques and methods for adjusting the order of information provided based on the emotional state of the user or customer.

[1580] This invention is a system that selects news articles, saves the URLs of the selected news articles, periodically obtains the latest information, summarizes them, and provides them to users. It recognizes the user's emotions and optimizes the summaries and display order of news articles based on those emotions, and also optimizes in-store guidance and recommended products based on customer emotions in physical stores. Implementing this invention requires the following steps.

[1581] First, when a user finds an article of interest on a news site, they select the article and click the "Favorites" button. The device sends the URL of the news article selected by the user and their user ID to the server. The server saves the received URL and user ID in a database. This adds the selected news article to the user's favorites and saves it.

[1582] The server then periodically accesses news sites based on the URLs of the news articles stored in the database to check for new information. This process uses web scraping technology. The server sends an HTTP request and receives HTML data as a response. The received HTML data is then parsed to extract the article content (title, body, date, etc.).

[1583] The content of the retrieved news articles is summarized using natural language processing (NLP) technology. The server inputs the retrieved article text into an NLP model to generate a summary. The generated summary is stored in a database. The stored information includes the news article URL, summary content, and the date and time the summary was generated. Python-based libraries (e.g., BeautifulSoup, spaCy, and HuggingFace Transformers) are used to retrieve news articles and generate summaries.

[1584] Furthermore, the present invention recognizes user emotions by incorporating an emotion engine. When a user accesses a dedicated web page and browses articles, the emotion engine analyzes the user's emotions based on the user's interactions and access history. This uses image processing and machine learning models (e.g., OpenCV, Dlib, TensorFlow). This allows the user's emotional state (e.g., excitement, joy, sadness, etc.) to be recognized as they read the articles.

[1585] The user's emotion data acquired by the emotion engine is used to adjust the summary content and display order of news articles. For example, if the user's emotion of sadness is recognized, the emotion engine can adjust the summary of the news article to emphasize the positive aspects. Also, if the user is feeling stressed, articles with content that will help them relax will be displayed preferentially.

[1586] This system can also be applied in physical stores. Using cameras installed in the store and the camera on the customer's smartphone, it recognizes emotions in real time from the customer's facial expressions and voice. The emotion recognition engine determines the customer's emotions and optimizes in-store guidance and recommended products according to those emotions. For example, if a customer is feeling stressed, relaxation products will be recommended first. The selection of recommended products and the order in which guidance information is displayed are also adjusted appropriately.

[1587] An example of an input prompt for a generative AI model is:

[1588] "If the customer's current emotional state is identified as stressed, generate information to recommend relaxation products. Also include information about the latest sales."

[1589] Based on these prompts, the generative AI model can generate stress-relieving products, their features, applicable discount information, and other information, which can be delivered to a display or smart glasses in real time.

[1590] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1591] Step 1:

[1592] When a user finds an article of interest on a news site, the device clicks the "Favorites" button.

[1593] Input: URL of the news article selected by the user, user ID

[1594] Specific operation: When the "Favorites" button is clicked, the device sends the URL of the selected news article and the user ID to the server.

[1595] Output: The URL and user ID are sent to the server.

[1596] Step 2:

[1597] The server stores the received URL and user ID in a database.

[1598] Input: URL of news article, user ID

[1599] Specific operation: The server stores the received URL and user ID in a database, and the selected news article is added to the user's favorites.

[1600] Output: URL and user ID stored in the database

[1601] Step 3:

[1602] The server periodically accesses news sites based on the URLs of news articles stored in the database to check for new information.

[1603] Input: URL stored in the database

[1604] Specific operation: The server sends an HTTP request and receives HTML data as a response. The received HTML data is analyzed to extract the article content (title, body, date, etc.).

[1605] Output: Extracted article content

[1606] Step 4:

[1607] The server inputs the retrieved news article text into a natural language processing (NLP) model to generate a summary.

[1608] Input: Article content (text data)

[1609] What it does: The server uses NLP techniques (e.g., spaCy or HuggingFace Transformers) to summarize the article text.

[1610] Output: Generated summary

[1611] Step 5:

[1612] The server stores the generated summaries in a database.

[1613] Input: Generated summary, news article URL, summary generation date and time

[1614] Specific operation: The server stores the generated summary, the URL of the news article, and the date and time the summary was generated in a database.

[1615] Output: Summary information stored in a database

[1616] Step 6:

[1617] When a user accesses a dedicated web page and reads a news article, the server uses an emotion recognition engine to recognize the user's emotion.

[1618] Input: Real-time facial and voice data of the user

[1619] Specific operation: The server uses image processing and machine learning models (e.g., OpenCV, Dlib, TensorFlow) to determine the user's emotions.

[1620] Output: Recognized emotion data

[1621] Step 7:

[1622] The server adjusts the summary content and display order of news articles based on the recognized emotion data.

[1623] Input: News article summary information, sentiment data

[1624] What it does: The emotion engine uses emotion data to adjust the summary content, highlighting positive aspects and changing the display order.

[1625] Output: Adjusted news article display information

[1626] Step 8:

[1627] The server transmits the adjusted news article display information to the user's terminal, and displays the latest news articles.

[1628] Input: Adjusted news article display information

[1629] Specific operation: Based on the received information, the device provides the user with a summary of the news articles and a display order that reflects the adjusted information.

[1630] Output: News article displayed on user's device

[1631] Step 9:

[1632] In a physical store, the device captures the customer's facial expressions and voice and sends them to an emotion recognition engine.

[1633] Input: Real-time facial and voice data

[1634] Specific operation: Cameras and smartphones in physical stores capture customers' facial expressions and voices and send them to a server.

[1635] Output: Facial expression and voice data sent to the server

[1636] Step 10:

[1637] The server optimizes in-store guidance and recommended products based on the customer's emotional state.

[1638] Input: Customer emotion data, in-store product information

[1639] Specific operation: The emotion recognition engine selects appropriate guidance and recommended products based on emotional data and displays them on the display or smart device.

[1640] Output: Optimized navigation information and recommended products

[1641] Step 11:

[1642] Smart displays in physical stores present emotion-based information.

[1643] Input: Optimized guidance information and recommended product information

[1644] Specific operation: Displays in physical stores display optimized guidance and recommended products to customers based on information received from the server.

[1645] Output: Optimization information displayed on the screen

[1646] Step 12:

[1647] It uses input prompt sentences to the generative AI model to generate that information.

[1648] Input: Prompt (e.g., "If the customer's current emotional state is identified as stressed, generate information to recommend relaxation products. Also include information about the latest sales.")

[1649] What it does: Based on this prompt, the generative AI model generates and delivers appropriate information in real time.

[1650] Output: Generated relaxation product information and special sale information

[1651] 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.

[1652] 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.

[1653] 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.

[1654] 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.

[1655] 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.

[1656] 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.

[1657] 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).

[1658] 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.

[1659] 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."

[1660] 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.

[1661] 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).

[1662] 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.

[1663] 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.

[1664] 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.

[1665] 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.

[1666] 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.

[1667] 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.

[1668] 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.

[1669] 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.

[1670] 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.

[1671] 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.

[1672] The following is further disclosed regarding the above embodiment.

[1673] (Claim 1)

[1674] means for selecting a news article and saving the URL of the selected news article;

[1675] a means for periodically retrieving news articles based on the URLs of the stored news articles;

[1676] means for summarizing the content of the retrieved news articles;

[1677] a means for storing the summarized news article content in a database;

[1678] A system including means for displaying updated summarized news articles to a user.

[1679] (Claim 2)

[1680] 2. The system according to claim 1, wherein the news article acquisition means acquires the news articles via the Internet.

[1681] (Claim 3)

[1682] 2. The system according to claim 1, wherein the means for displaying the summarized content of news articles retrieves the latest information from a database in response to a user request and displays it on the user's terminal.

[1683] "Example 1"

[1684] (Claim 1)

[1685] a means for saving the URL of the news article selected by the user;

[1686] a means for periodically retrieving news articles based on the URLs of the stored news articles;

[1687] A means for summarizing the content of the acquired news articles using a natural language processing model;

[1688] a means for storing the summarized news article content in a database;

[1689] A system including means for displaying updated summarized news articles to a user.

[1690] (Claim 2)

[1691] 2. The system according to claim 1, wherein the news article acquisition means acquires the news articles via the Internet.

[1692] (Claim 3)

[1693] 2. The system according to claim 1, wherein the means for displaying the summarized content of news articles retrieves the latest information from a database in response to a user request and displays it on the user's terminal.

[1694] "Application Example 1"

[1695] (Claim 1)

[1696] means for selecting a news article and saving the URL of the selected news article;

[1697] a means for periodically retrieving news articles based on the URLs of the stored news articles;

[1698] means for summarizing the content of the retrieved news articles;

[1699] a means for storing the summarized news article content in a database;

[1700] means for displaying updated summarized news articles to a user;

[1701] a means for summarizing the content of the retrieved articles using a natural language processing model for summary generation;

[1702] A system including means for notifying a user terminal of summary information.

[1703] (Claim 2)

[1704] 2. The system according to claim 1, wherein the news article acquisition means acquires the news articles via the Internet.

[1705] (Claim 3)

[1706] 2. The system according to claim 1, wherein the means for displaying the summarized content of news articles retrieves the latest information from a database in response to a user request and displays it on the user's terminal.

[1707] "Example 2: Combining Emotion Engines"

[1708] (Claim 1)

[1709] means for selecting a news article and saving the URL of the selected news article;

[1710] a means for periodically retrieving news articles based on the URLs of the stored news articles;

[1711] means for summarizing the content of the retrieved news articles;

[1712] a means for storing the summarized news article content in a database;

[1713] means for recognizing a user's emotion and adjusting the content of a news article summary based on the emotion;

[1714] means for optimizing the display order of news articles based on the emotional state of a user;

[1715] A system including means for displaying summarized news article updates to a user.

[1716] (Claim 2)

[1717] 2. The system according to claim 1, wherein the news article acquisition means acquires the news articles via the Internet.

[1718] (Claim 3)

[1719] 2. The system according to claim 1, wherein the means for displaying the summarized content of news articles retrieves the latest information from a database in response to a user request and displays it on the user's terminal.

[1720] "Application example 2 when combining emotion engines"

[1721] (Claim 1)

[1722] means for selecting a news article and saving the URL of the selected news article;

[1723] a means for periodically retrieving news articles based on the URLs of the stored news articles;

[1724] means for summarizing the content of the retrieved news articles;

[1725] a means for storing the summarized news article content in a database;

[1726] means for displaying updated summarized news articles to a user;

[1727] means for recognizing a user's emotion;

[1728] A means for optimizing summarization and display order of news articles based on perceived sentiment;

[1729] A system that recognizes customer emotions in a physical store and includes a means for optimizing in-store guidance and recommended products based on those emotions.

[1730] (Claim 2)

[1731] 2. The system according to claim 1, wherein the news article acquisition means acquires the news articles via the Internet.

[1732] (Claim 3)

[1733] 2. The system according to claim 1, wherein the means for displaying the summarized content of news articles retrieves the latest information from a database in response to a user request and displays it on the user's terminal. [Explanation of symbols]

[1734] 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 for selecting a news article and saving the URL of the selected news article; a means for periodically retrieving news articles based on the URLs of the stored news articles; means for summarizing the content of the retrieved news articles; a means for storing the summarized news article content in a database; and means for displaying updated information on summarized news articles to a user.

2. 2. The system according to claim 1, wherein the news article acquisition means acquires news articles via the Internet.

3. 2. The system according to claim 1, wherein the means for displaying the summarized content of news articles retrieves the latest information from the database in response to a user request and displays it on the user's terminal.

Images