System
The system addresses the inefficiencies of current search systems by filtering and categorizing information based on user keywords, ensuring quick and reliable access to relevant data through a portal site format.
Patent Information
- Application Number
- JP2024137231
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2024-08-16
- Publication Date
- 2026-02-27
AI Technical Summary
Current internet search systems face challenges in providing users with reliable, localized, and up-to-date information efficiently, often leading to misinformation and difficulty in finding relevant information, especially for users with low IT literacy.
A system that collects information based on user keywords, filters out misinformation, organizes it chronologically and categorically, and presents it in a portal site format using generative AI, ensuring quick and efficient access.
Enables users to obtain accurate and organized information quickly, eliminating false content and enhancing user experience by providing reliable, easily accessible information.
Smart Images

Figure 2026034110000001_ABST
Abstract
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] Current internet search systems make it difficult for users to find the information they need quickly and efficiently. In particular, the following problems exist: low reliability of information, excessive misinformation, dispersion of information, and overlooking of important information. Furthermore, users with low IT literacy have a lot of trouble finding the right information. This means that users spend a lot of time trying to find the information they need. Furthermore, search results are biased toward general ranking algorithms, making it difficult to obtain localized and up-to-date information. [Means for solving the problem]
[0005] In order to solve the above-mentioned problems, the present invention provides a system that collects information based on search keywords entered by a user, eliminates unreliable information, organizes the information in chronological order and by category, and provides it to users in the form of a portal site. The system includes the following means:
[0006] A means for receiving search keywords from users
[0007] A crawling engine that collects information related to search keywords from online sources
[0008] How to filter collected information and eliminate misinformation
[0009] A means of sorting filtered information chronologically and categorically
[0010] Generative AI means for displaying generated information in a portal site format
[0011] A means of providing information to a user's device
[0012] This allows users to quickly and efficiently obtain the information they need. Furthermore, by eliminating false information, it is possible to provide more reliable information. Furthermore, by organizing information in chronological order and by category, it is possible to prevent information from being scattered and important information from being overlooked.
[0013] The "means for receiving search keywords from a user" is a function that allows the system to receive search keywords entered by a user.
[0014] "Online information sources" is a general term for various databases, websites, social media sites, blogs, etc. that can be accessed on the Internet.
[0015] The "crawling engine means for collecting information related to search keywords" is an engine function that automatically collects information related to specified keywords from the Internet.
[0016] "Means for filtering collected information and eliminating false information" refers to a function for analyzing collected information and eliminating false or unreliable information.
[0017] "Means for classifying filtered information chronologically and by category" refers to a function for organizing filtered information based on the date and time it was collected, and further separating it into different categories according to its content.
[0018] "Generative AI means for displaying generated information in a portal site format" refers to an artificial intelligence function that formats classified information into a portal site format that is easy for users to view.
[0019] The "means for providing information to a user's terminal" is a function for transmitting formatted information to a terminal used by the user and displaying it. [Brief explanation of the drawings]
[0020] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION
[0021] 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.
[0022] First, the terms used in the following description will be explained.
[0023] 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).
[0024] 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.
[0025] 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.
[0026] 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.
[0027] 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."
[0028] [First embodiment]
[0029] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.
[0030] 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.
[0031] 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).
[0032] 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.
[0033] 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.
[0034] 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.
[0035] 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.
[0036] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.
[0037] 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.
[0038] 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.
[0039] 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.
[0040] 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."
[0041] This invention is a system that collects necessary information based on search keywords entered by a user, eliminates unreliable information, and provides the information in a format that is easy for the user to view. Specific embodiments of this system are described below.
[0042] Server Processing
[0043] 1. Receiving search keywords from users
[0044] When a user enters search keywords on their device and presses the search button, the request is sent to the server. The server receives this request and starts the search process. For example, if a user enters "COVID-19 vaccine information," the server analyzes the keywords and prepares to collect related information.
[0045] 2. Starting the crawling engine
[0046] The server starts a crawling engine to collect information related to the search keyword from multiple sources on the Internet (news sites, official websites, social media, blogs, etc.). The engine searches sources as needed to obtain the latest data.
[0047] 3. Information filtering
[0048] The collected information is processed through a filtering system within the server. The filtering system checks the reliability of the information, the reputation of the source, and the consistency of the content, and filters out rumors and misinformation. For example, anonymous posts on social media and information from unreliable sources are filtered out.
[0049] 4. Classification and chronological organization of information
[0050] The filtered information is organized chronologically and by category: specifically, it is sorted based on the date and time of collection and categorized based on the content of the information (e.g., latest news, side effects, research data).
[0051] 5. Information Shaping with Generative AI
[0052] The organized information is then formatted into a portal site by a generative AI. This generative AI understands the meaning of search keywords and determines the layout so that users can easily browse related information. For example, the latest news is placed at the top of the page, and information on side effects and research data is displayed as the main content.
[0053] 6. Transmission of Information
[0054] The formatted information is sent to the user's device using web technologies such as HTML and JSON in a format suitable for the user's device.
[0055] Terminal handling
[0056] 1. Enter search keywords
[0057] Users enter keywords related to the information they want to find into the device's search box. For example, they enter "COVID-19 vaccine information."
[0058] 2. Receiving and Displaying Information
[0059] The terminal receives the formatted information sent from the server. This information is displayed in a portal site format, allowing the user to easily browse the categorized information.
[0060] User Activity
[0061] 1. Enter keywords
[0062] The user inputs the information they want to know into the device. For example, they input specific keywords such as "COVID-19 vaccine information."
[0063] 2. Verification and Use of Information
[0064] They can browse the displayed portal-style information and access the information they need, for example, by reading the latest news about vaccines, checking side-effect data and scientific research results.
[0065] Specific examples
[0066] (Search keyword: "COVID-19 vaccine information")
[0067] 1. Receiving a request
[0068] The server receives a search request for "COVID-19 vaccine information."
[0069] 2. Starting the crawling engine
[0070] The crawling engine starts up and collects information about vaccines from news sites, official websites, social media, etc.
[0071] 3. Information filtering
[0072] The filtering system evaluates the reliability of collected information and filters out false information, for example, giving priority to information from reliable news sites and official organizations.
[0073] 4. Classifying and Formatting Information
[0074] The filtered information is categorized into the latest news, side effect data, research results, etc. and organized in chronological order. Generative AI then formats this information into a portal site format and prepares it for delivery to users.
[0075] 5. Provision of Information
[0076] The formatted information is sent to the user's terminal, where the user can view the information displayed in a portal site format and access the information they need.
[0077] As described above, this invention is designed to enable users to efficiently obtain the information they need. By eliminating false information and categorizing information by time series and category, reliability and ease of use are improved. This system allows users to quickly access appropriate information.
[0078] The processing flow will be explained below.
[0079] Server Processing
[0080] Step 1:
[0081] The server receives the search keywords sent by the user, along with the user's settings information (language settings, region settings, etc.).
[0082] Step 2:
[0083] The server starts a crawling engine to collect information, and the crawling engine prepares to collect information related to the specified keyword from a wide range of sources.
[0084] Step 3:
[0085] The crawling engine accesses predefined sources such as news sites, official sites, social media, blogs, etc. to gather information related to the search keyword, exploring multiple sources in parallel during this process.
[0086] Step 4:
[0087] The collected information passes through a filtering system within the server, which evaluates the reliability of the information and filters out hoaxes and unreliable information. For example, unreliable anonymous social media posts are filtered out.
[0088] Step 5:
[0089] The filtered information is sorted chronologically and by category. Specifically, it is organized in chronological order based on the date and time the information was collected, and then divided into categories such as "latest news," "side effects," and "research data" according to its content.
[0090] Step 6:
[0091] The AI then organizes the information and lays it out in a portal site format. The AI understands the meaning of the search keywords and places relevant information in key positions. For example, the latest news is placed at the top of the page, and detailed research results are displayed in the center.
[0092] Step 7:
[0093] The formatted information is then converted into a format appropriate for the user's device (such as HTML or JSON), and the converted data is ready to be sent to the user's device.
[0094] Terminal handling
[0095] Step 1:
[0096] The user enters keywords into the search box on the device and presses the search button. This input information is sent to the server.
[0097] Step 2:
[0098] The terminal receives the formatted information sent from the server, parses the received data, and prepares it for display.
[0099] Step 3:
[0100] The device displays the received information in a portal site format, with the latest news displayed at the top of the screen, side effect statistics in the middle, and links to research papers at the bottom.
[0101] User Action
[0102] Step 1:
[0103] Users enter the keywords they want to know in the search box on their device and press the search button. For example, they could enter "COVID-19 vaccine information."
[0104] Step 2:
[0105] Users can check the information displayed on their devices in the form of a portal site and access the information they need, for example, by clicking on the latest news headline to read a detailed article.
[0106] The above is a specific processing flow of the present invention. This system allows users to obtain the information they need efficiently and quickly.
[0107] Example 1
[0108] 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."
[0109] There is a lot of information scattered on the Internet, making it difficult for users to quickly and accurately obtain the information they need. Furthermore, the Internet sometimes contains inaccurate or false information, making it difficult for users to access reliable information. There is a particular need for a system that can efficiently provide information for urgent situations or when specialized information is required.
[0110] 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.
[0111] In this invention, the server includes means for receiving search keywords from users, a crawling engine for collecting information related to the search keywords from online information sources, means for filtering the collected information and eliminating false information, means for classifying the filtered information in chronological order and by category, means for formatting the classified information into a portal site format using a generative AI model, and means for transmitting the information to the user's terminal, thereby enabling users to quickly and efficiently obtain reliable information.
[0112] The "means for receiving search keywords from a user" refers to a mechanism for transmitting search keywords entered by a user via an Internet terminal to a server and receiving the keywords.
[0113] The "crawling engine means" is a program that automatically collects information related to a specified keyword from multiple information sources on the Internet.
[0114] A "means for filtering out misinformation" is an algorithm or system that detects and filters out unreliable or false information from collected information.
[0115] "Chronological and categorical sorting" refers to the process of sorting the filtered information in chronological order based on the date and time of collection and categorizing it according to its content.
[0116] "Means for formatting into a portal site format using a generative AI model" refers to a technology that uses an artificial intelligence model to reconstruct classified information into a portal site format that is easy for users to view.
[0117] "Means for sending information to the user's device" refers to a mechanism for sending formatted information to the user's device using an appropriate communication means (e.g., HTML, JSON).
[0118] This invention is a system that collects necessary information based on search keywords entered by a user, eliminates unreliable information, and provides the information in a format that is easy for the user to view. Specific embodiments of this system are described below.
[0119] Server Processing
[0120] Receiving search keywords from users
[0121] When a user enters specific keywords, such as "COVID-19 vaccine information," into their device and presses the search button, the request is sent to the server. The server receives the HTTP request and extracts the search keywords from it. This prepares the system to begin collecting information.
[0122] Starting the crawling engine
[0123] The server runs a crawling engine using Python tools such as BeautifulSoup and Scrapy to collect information from multiple sources on the Internet. Based on the input keywords, the crawling engine searches news sites, official websites, social media, blogs, etc. to obtain the latest relevant data.
[0124] Information Filtering
[0125] The collected information is stored in a database, and then processed by a filtering system on the server. The filtering system evaluates the reliability of the source and its content, and runs an algorithm to filter out misinformation and unreliable information. Based on the source's reliability score, information with a score below a certain level is filtered out.
[0126] Classification and chronological organization of information
[0127] The filtered information is sorted chronologically and by category (e.g., latest news, side effects, research data). The server retrieves this information from the database and sorts it chronologically using an SQL query.
[0128] Shaping information with generative AI
[0129] The classified information is then formatted into a portal site using a generative AI model (e.g., GPT-4 (registered trademark)). The generative AI understands the context of the search keywords and generates a layout that is easy for users to view. For example, the latest news is placed at the top of the page, and side effects and research data are placed in different sections.
[0130] Sending information
[0131] The formatted information is sent to the user's device in HTML or JSON format. The server generates an HTTP response and sends the information to the user's browser. The browser receives the response and displays the information.
[0132] Terminal handling
[0133] Enter search keywords
[0134] The user enters the information of interest (e.g., "COVID-19 vaccine information") into the search box on the device and presses the search button. The device then sends the user's input to the server.
[0135] Receiving and displaying information
[0136] The user's device receives the formatted information sent from the server and displays it on the screen. Information sent in HTML format is rendered by the browser and displayed so that the user can easily check the content. In the case of JSON format, JavaScript (registered trademark) manipulates the DOM to dynamically display the information.
[0137] User Activity
[0138] Enter keywords
[0139] Users enter keywords related to the information they want to know into the device. For example, they can enter "COVID-19 vaccine information."
[0140] Verifying and Using Information
[0141] Users can review the displayed information and access the information they need, such as reading the latest vaccine news or viewing data and research results on side effects.
[0142] Specific examples
[0143] Search keyword: "COVID-19 vaccine information"
[0144] 1. Request reception: The server receives a search request for "COVID-19 vaccine information" and analyzes the keywords.
[0145] 2. Launching the crawling engine: Collect information about vaccines from news sites, official websites, social media, etc.
[0146] 3. Information filtering: Select reliable information and eliminate misinformation.
[0147] 4. Classification and formatting of information: The information is classified into the latest news, side effect data, research results, etc. and organized in chronological order. Generative AI formats this information into a portal site format.
[0148] 5. Providing information: The formatted information is sent to the user's device and displayed in portal site format.
[0149] As described above, the present invention is a system that efficiently collects information required by users and provides it in a highly reliable manner.
[0150] The flow of the identification process in the first embodiment will be described with reference to FIG.
[0151] Step 1:
[0152] Receiving search keywords from users
[0153] A user enters "COVID-19 vaccine information" into the search box on their device and presses the search button. The device generates an HTTP request and sends it to the server. The server receives this request and extracts the search keywords. It then prepares to collect information based on the analyzed keywords (Input: User's search keyword "COVID-19 vaccine information", Output: Analyzed search keywords).
[0154] Step 2:
[0155] Starting the crawling engine
[0156] The server starts a Python crawling engine (e.g., BeautifulSoup or Scrapy) using the parsed search keywords as input. This engine searches multiple sources (news sites, official website, social media, blogs, etc.) for information related to the specified keywords and obtains HTML data. The crawling engine then crawls these sites and returns the collected HTML content to the server (input: parsed search keywords, output: collected HTML data).
[0157] Step 3:
[0158] Information Filtering
[0159] The filtering system in the server receives the HTML data collected in the previous step as input. The filtering system runs an algorithm to evaluate the reliability of the information and eliminate hoaxes and unreliable information. For example, it analyzes the source domain and calculates a reliability score. Information from unreliable sites and anonymous social media posts will be eliminated due to their low scores. (Input: Collected HTML data, Output: Filtered information)
[0160] Step 4:
[0161] Classification and chronological organization of information
[0162] The server then sorts the filtered information into chronological order and categories. This process uses SQL queries to retrieve information stored in the database, sorts it based on the collection date and time, and categorizes it by content. Specifically, the information is classified into latest news, side effects, and research data, and then organized in chronological order within each category (input: filtered information, output: classified and organized information).
[0163] Step 5:
[0164] Shaping information with generative AI
[0165] The server takes the classified and organized information as input and uses a generative AI model (e.g., GPT-4) to format it into a portal site. The generative AI model understands the context of the search keywords and lays it out in a format that is easy for users to read. For example, the latest news is placed at the top of the page, and side effects and research data are placed in dedicated sections (input: classified and organized information, output: formatted HTML content).
[0166] Step 6:
[0167] Sending information
[0168] The formatted HTML content is sent from the server to the user's device. The server generates an HTTP response and delivers the information to the user's device in HTML or JSON format. The user's browser receives this and displays it optimally (input: formatted HTML content, output: information sent to the user's device).
[0169] Step 7:
[0170] Receiving and displaying information
[0171] The user's device receives the HTML or JSON data sent from the server and renders it in the browser. In the case of HTML format, the browser's rendering engine displays the page, allowing the user to easily browse the categorized information. In the case of JSON format, JavaScript manipulates the DOM to dynamically display the information (input: information sent from the server, output: information displayed on the device).
[0172] Step 8:
[0173] User review and use of information
[0174] The user checks the displayed information and accesses the information they need, such as reading the latest news about vaccines, checking side effect data, and scientific research results (input: information displayed on the device, output: information obtained by the user).
[0175] Through these steps, the system is able to efficiently collect and organize reliable information and provide it to users.
[0176] (Application example 1)
[0177] 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."
[0178] Conventional information retrieval systems have issues such as low reliability of information collected based on search keywords and lack of an appropriate display format for users to easily view and use information. Furthermore, especially when providing information for smart glasses and other wearable devices, the lack of a voice-based interface limits the user experience. To address these issues, a system is needed that can provide real-time, reliable information and display that information in an optimized audio and visual format.
[0179] 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.
[0180] In this invention, the server includes means for receiving search keywords from users, crawling engine means for collecting information related to the search keywords from online sources, means for filtering the collected information and eliminating false information, means for classifying the filtered information in chronological order and by type, generation AI means for displaying the generated information in a format suitable for a display device, means for analyzing voice input and generating search keywords, a speech synthesizer for outputting the classified information by voice, and means for providing the information to the user's display device. This allows users to obtain reliable information in real time and easily view and use it in a format optimized for audio and visuals.
[0181] A "user" is an entity that searches for and uses information.
[0182] "Search keywords" are words or phrases related to the information a user wants to find.
[0183] "Online sources" are information sources such as websites and databases that can be accessed via the Internet.
[0184] A "crawling engine" is software that automatically collects specific information from online sources.
[0185] "Filtering" is the process of removing unreliable information or false information from collected information.
[0186] "Chronological order" refers to arranging information in order based on the date and time it was collected.
[0187] "By type" refers to classifying information into categories according to its content.
[0188] "Generative AI" is artificial intelligence that formats collected and filtered information into a format suitable for the user's device or display device.
[0189] "Voice input" is a method in which a user provides information to a system by speaking into a microphone.
[0190] A "speech synthesizer" is a device that converts text information into speech and makes it audible to the user.
[0191] A "display device" is a device for visually presenting collected and generated information to a user.
[0192] This invention is a system that collects necessary information based on search keywords entered by the user through input or voice, eliminates unreliable information, and uses a generation AI to provide the information in a format suitable for the user's device. Specific embodiments of this system are described below.
[0193] Server Processing
[0194] 1. Receiving search keywords from users
[0195] The user enters a search keyword by voice or text, and the request is sent to the server. An example search keyword is "new spring clothes." The server receives this request and starts the search process.
[0196] 2. Starting the crawling engine
[0197] The server uses a crawling engine to collect information related to the search keywords from online sources, such as fashion websites, official brand websites, and social media.
[0198] 3. Information filtering
[0199] The information collected by the crawling engine is processed through a filtering system within the server. The filtering system evaluates the reliability of the information and removes unnecessary data such as false information. For example, anonymous social media posts and data from less reliable sources are excluded.
[0200] 4. Classification and chronological organization of information
[0201] The server then organizes the filtered information chronologically and by type, specifically into categories like tops, bottoms, and outerwear.
[0202] 5. Information Shaping with Generative AI
[0203] The organized information is then formatted into a suitable format for the display device by the generation AI, which understands the meaning of the search keywords and determines the layout so that the user can easily view related information.
[0204] 6. Transmission of Information
[0205] The formatted information is sent from the server to the user's display device using web technologies such as HTML and JSON.
[0206] Terminal handling
[0207] 1. Enter search keywords
[0208] The user inputs keywords related to the information they want to search for into a search box on the device or provides them by voice. For example, the user inputs or provides by voice a specific keyword such as "new spring clothes."
[0209] 2. Receiving and Displaying Information
[0210] The terminal receives the formatted information sent from the server, displays the information in a format optimal for the display device, and uses a speech synthesizer to provide the information to the user aloud.
[0211] User Activity
[0212] 1. Enter search keywords
[0213] Users input the information they want to know into the terminal. Voice input is also supported, so for example, keywords such as "new spring clothes" can be provided by voice.
[0214] 2. Verification and Use of Information
[0215] Users can access the information they need by viewing the information displayed in a portal format on the display device. Audio output is also provided, allowing users to view and hear the information.
[0216] Specific examples
[0217] For example, if a user uses smart glasses to say, "I want new spring clothes," the system will analyze the speech, collect relevant fashion information, filter and organize it, and use the generative AI to visually display the information in a format that is optimal for the smart glasses, while the speech synthesizer will provide the information aloud.
[0218] Prompt Sentence Examples
[0219] What's the latest information on spring fashion?
[0220] The hardware used includes smart glasses (e.g., microphone, speaker, display) and software including Python, Requests, BeautifulSoup, OpenAI® GPT-3®, and google_speech, enabling users to efficiently receive up-to-date, reliable information through visual and audio channels.
[0221] The flow of the specific processing in the application example 1 will be described with reference to FIG.
[0222] Step 1:
[0223] Enter and submit search keywords
[0224] The user enters search keywords using the voice input function of the smart glasses.
[0225] For example: "I want some new spring clothes."
[0226] Speech recognition software converts this voice input into text data.
[0227] The converted text data is sent to the server.
[0228] Step 2:
[0229] Receiving search keywords
[0230] The server receives the text data sent by the user.
[0231] Input: A keyword spoken by the user (e.g., "new spring clothes")
[0232] Output: Search keyword data for analysis
[0233] Step 3:
[0234] Starting the crawling engine
[0235] The server launches a crawling engine targeting multiple online information sources (fashion sites, official brand websites, social media, etc.).
[0236] Input: Search keyword ("new spring clothes")
[0237] Data processing: Obtain relevant information from sources and collect it as text data.
[0238] Output: Collected text data
[0239] Step 4:
[0240] Information Filtering
[0241] The server evaluates the reliability of the collected information using a filtering system.
[0242] Input: Collected text data
[0243] Data calculation: Remove unreliable information and false information.
[0244] What it does: Identify and filter out low-trust sources (e.g., anonymous social media posts).
[0245] Output: Filtered reliable information data
[0246] Step 5:
[0247] Classification and chronological organization of information
[0248] The server categorizes the filtered information chronologically and by type (tops, bottoms, outerwear, etc.).
[0249] Input: Filtered reliable information data
[0250] Data processing: Sort the information based on the date and time it was collected and divide it into categories.
[0251] Output: Categorized and organized information data
[0252] Step 6:
[0253] Shaping information with generative AI
[0254] The server uses generative AI to format the organized information into a format suitable for display on smart glasses.
[0255] Input: Classified and organized information data
[0256] Data calculation: Generative AI determines the optimal layout for the user and converts it into a format that is easy to display visually.
[0257] What it does: Sorts information by category, with the most recent information at the top.
[0258] Output: Formatted information data
[0259] Step 7:
[0260] Sending information
[0261] The server sends the formatted information to the user's smart glasses.
[0262] Input: Formatted information data
[0263] Data processing: Converts data into a transmission format (e.g., HTML, JSON) and executes transmission logic.
[0264] Output: Information data sent to the user terminal
[0265] Step 8:
[0266] Display and speech output of information
[0267] The terminal provides the received information to the user visually and audibly.
[0268] Input: Information data sent from the server
[0269] Specific operation: Information is displayed on the HUD of the smart glasses, and at the same time, the voice synthesizer plays the information aloud.
[0270] Output: Visual and audio information provided to the user
[0271] 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.
[0272] The present invention is a system that collects information on the Internet based on search keywords entered by users, eliminates unreliable information, and recognizes the user's emotions using an emotion engine to provide the user with appropriate information in the form of a portal site. A specific embodiment of this system is described below.
[0273] Server Processing
[0274] 1. Receiving search keywords from users
[0275] When a user enters search keywords on their device and presses the search button, the request is sent to the server. The server receives this request and starts the search process. For example, if a user enters "COVID-19 vaccine information," the server analyzes the keywords and prepares to collect related information.
[0276] 2. Starting the crawling engine
[0277] The server starts a crawling engine to collect information related to the search keyword from multiple sources on the Internet (news sites, official websites, social media, blogs, etc.). The engine searches sources as needed to obtain the latest data.
[0278] 3. Information filtering
[0279] The collected information is processed through a filtering system within the server. The filtering system checks the reliability of the information, the reputation of the source, and the consistency of the content, and filters out rumors and misinformation. For example, anonymous posts on social media and information from unreliable sources are filtered out.
[0280] 4. Classification and chronological organization of information
[0281] The filtered information is organized chronologically and by category. Specifically, it is organized in chronological order based on the date and time of collection, and then divided into categories such as "latest news," "side effects," and "research data" according to its content.
[0282] 5. Information Shaping with Generative AI
[0283] The organized information is then laid out in a portal site format by a generative AI. This generative AI understands the meaning of search keywords and arranges relevant information so that users can easily browse it. For example, the latest news is placed at the top of the page, and information on side effects and research data are displayed as the main content.
[0284] 6. User Emotion Recognition by Emotion Engine
[0285] The server uses an emotion engine to analyze the user's search keywords and behavioral history to recognize the user's emotions. Based on this emotion recognition, the priority of the information displayed is changed. For example, if the user is worried, information that gives a sense of security is displayed at the top.
[0286] 7. Transmission of Information
[0287] The formatted information is then converted into a format appropriate for the user's device (such as HTML or JSON), and the converted data is ready to be sent to the user's device.
[0288] Terminal handling
[0289] 1. Enter search keywords
[0290] Users enter keywords related to the information they want to find into the device's search box. For example, they enter "COVID-19 vaccine information."
[0291] 2. Receiving and Displaying Information
[0292] The device receives the formatted information sent from the server. This information is displayed in a portal site format, allowing users to easily browse categorized information. For example, the top of the screen displays the latest news, the middle section displays side effect statistics, and the bottom section displays links to research papers. In addition, information appropriate to the user's current emotional state is displayed based on the analysis results of the emotion engine.
[0293] User Activity
[0294] 1. Enter keywords
[0295] The user inputs the information they want to know into the device. For example, they input specific keywords such as "COVID-19 vaccine information."
[0296] 2. Verification and Use of Information
[0297] Users can review the displayed portal-style information and access the information they need. For example, they can read the latest news about vaccines, check side effect data, and check scientific research results. Based on the analysis results of the emotion engine, information appropriate to the user's emotional state is recommended, providing information to reduce stress.
[0298] Specific examples
[0299] (Search keyword: "COVID-19 vaccine information")
[0300] 1. Receiving a request
[0301] The server receives a search request for "COVID-19 vaccine information."
[0302] 2. Starting the crawling engine
[0303] The crawling engine starts up and collects information about vaccines from news sites, official websites, social media, etc.
[0304] 3. Information filtering
[0305] The filtering system evaluates the reliability of collected information and filters out false information, for example, giving priority to information from reliable news sites and official organizations.
[0306] 4. Classifying and Formatting Information
[0307] The filtered information is categorized into the latest news, side effect data, research results, etc. and organized in chronological order. Generative AI then formats this information into a portal site format and prepares it for delivery to users.
[0308] 5. Emotion Recognition by Emotion Engine
[0309] If a user is worried, information that provides reassurance (e.g., vaccine effectiveness or reassuring statistics) will be prioritized based on that emotion.
[0310] 6. Transmission of Information
[0311] The formatted information is sent to the user's terminal, where the user can view the information displayed in a portal site format and access the information they need.
[0312] As described above, this invention is designed to enable users to efficiently obtain the information they need. In addition to eliminating false information and classifying information by time series and category, reliability, ease of use, and psychological compatibility are improved by recognizing user emotions using an emotion engine. This system allows users to quickly access appropriate information.
[0313] The processing flow will be explained below.
[0314] Server Processing
[0315] Step 1:
[0316] The server receives the search keywords sent by the user, along with the user's settings information (language settings, region settings, etc.).
[0317] Step 2:
[0318] The server starts a crawling engine to collect information, and the crawling engine prepares to collect information related to the specified keyword from a wide range of sources.
[0319] Step 3:
[0320] The crawling engine accesses predefined sources such as news sites, official sites, social media, blogs, etc. to gather information related to the search keyword, exploring multiple sources in parallel during this process.
[0321] Step 4:
[0322] The collected information passes through a filtering system within the server, which evaluates the reliability of the information and filters out hoaxes and unreliable information. For example, unreliable anonymous social media posts are filtered out.
[0323] Step 5:
[0324] The filtered information is sorted chronologically and by category. Specifically, it is organized in chronological order based on the date and time the information was collected, and then divided into categories such as "latest news," "side effects," and "research data" according to its content.
[0325] Step 6:
[0326] The AI then organizes the information and lays it out in a portal site format. The AI understands the meaning of the search keywords and places relevant information in key positions. For example, the latest news is placed at the top of the page, and detailed research results are displayed in the center.
[0327] Step 7:
[0328] The server starts the emotion engine, which analyzes the user's search keywords and history to recognize the user's emotions. Based on the recognition results, the display priority is changed.
[0329] Step 8:
[0330] The formatted information is then converted into a format appropriate for the user's device (such as HTML or JSON), and the converted data is ready to be sent to the user's device.
[0331] Terminal handling
[0332] Step 1:
[0333] The user enters keywords into the search box on the device and presses the search button. This input information is sent to the server.
[0334] Step 2:
[0335] The terminal receives the formatted information sent from the server, parses the received data, and prepares it for display.
[0336] Step 3:
[0337] The device displays the received information in a portal site format. For example, the top of the screen displays the latest news, side effect statistics in the middle, and links to research papers in the bottom. Based on the analysis results of the emotion engine, the device also displays information appropriate to the user's current psychological state.
[0338] User Action
[0339] Step 1:
[0340] Users enter the keywords they want to know in the search box on their device and press the search button. For example, they could enter "COVID-19 vaccine information."
[0341] Step 2:
[0342] Users can check the portal site-style information displayed on their device and access the information they need. For example, they can click on the latest news headline to read a detailed article. They can also view recommended reassuring information based on the analysis results of the emotion engine.
[0343] Specific examples
[0344] (Search keyword: "COVID-19 vaccine information")
[0345] Server details:
[0346] Step 1:
[0347] The server receives a search request for "COVID-19 vaccine information."
[0348] Step 2:
[0349] The crawling engine starts up and collects information about vaccines from news sites, official websites, social media, etc.
[0350] Step 3:
[0351] The collected information is filtered out using a filtering system to remove false information, such as anonymous posts on social media and unreliable information.
[0352] Step 4:
[0353] The filtered information is categorized into the latest news, side effect data, research results, etc. and organized in chronological order.
[0354] Step 5:
[0355] The generating AI lays out the classified information in a portal site format, placing important information at the top.
[0356] Step 6:
[0357] The server activates an emotion engine, analyzes the user's search history and behavior, and prioritizes displaying information that will reassure worried users.
[0358] Step 7:
[0359] The formatted information is converted into HTML or JSON format and sent to the user's device.
[0360] Detailed terminal processing:
[0361] Step 1:
[0362] The user enters "COVID-19 vaccine information" and presses the search button.
[0363] Step 2:
[0364] The terminal receives the information sent from the server and analyzes it.
[0365] Step 3:
[0366] The terminal displays the received information in a portal site format, and displays information according to the user's emotions with priority.
[0367] User detailed processing:
[0368] Step 1:
[0369] A user enters "COVID-19 vaccine information" into the search box and presses the search button.
[0370] Step 2:
[0371] Users can check the displayed information in a portal site format and click on the item they are interested in to see more details. For example, a worried user will be given priority in seeing reassuring information demonstrating the effectiveness of vaccines.
[0372] As described above, the present invention not only enables users to efficiently obtain the information they need, but also improves user satisfaction and reliability, since the information provided by the emotion engine is adapted to the user's emotional state.
[0373] Example 2
[0374] 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."
[0375] In today's Internet environment, a huge amount of information is generated every day, making it difficult for users to quickly obtain reliable information from that information. Furthermore, if the information is unreliable or contains false information, users may act based on the incorrect information. Furthermore, conventional information gathering systems lack a mechanism for changing the priority of information taking into account the user's emotions, which makes it difficult for users to quickly access the information they want to see.
[0376] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving a search keyword from a user, information collection means for collecting information related to the search keyword from online information sources, means for filtering the collected information and eliminating unreliable information, means for classifying the filtered information in chronological order and by category, generation AI means for displaying the generated information in a portal site format, emotion recognition means for recognizing the user's emotion and changing the priority of information based on the emotion, and means for providing information to the user's terminal. This allows the user to quickly access reliable information and provides appropriate information according to the user's emotional state, thereby improving the convenience and reliability of information.
[0377] The "means for receiving search keywords from a user" is a function that allows the server to receive search keywords that a user has input on a terminal.
[0378] "Information gathering means for gathering information related to search keywords from online sources" is a function for gathering information related to specific search keywords from multiple sources on the Internet.
[0379] "Means for filtering collected information and eliminating unreliable information" refers to a function for removing unreliable information or false information from the collected information.
[0380] "Means for classifying filtered information chronologically and by category" refers to a function for organizing filtered information in chronological order based on the date and time of collection and classifying it into different categories depending on its content.
[0381] "Generative AI means for displaying generated information in a portal site format" is a function that utilizes artificial intelligence technology to format organized and classified information into a portal site format and display it in a way that is easy for users to view.
[0382] "Emotion recognition means that recognizes the user's emotions and changes the priority of information based on those emotions" is a function that analyzes the user's search keywords and behavioral history to recognize the user's emotions, and dynamically changes the priority of the information to be displayed based on the results.
[0383] The "means for providing information to a user's terminal" is a function for transmitting formatted information to a user's terminal in an appropriate format.
[0384] This invention is a system that collects information on the Internet based on search keywords entered by users, eliminates unreliable information, recognizes the user's emotions using an emotion engine, and provides appropriate information in the form of a portal site. This system is composed of three entities: a server, a terminal, and a user.
[0385] Server Action:
[0386] First, the server receives the search keywords entered by the user on the device. For example, if a specific keyword such as "COVID-19 vaccine information" is entered, the server analyzes this information using a dedicated analysis module.
[0387] Next, the server launches a crawling engine as a means of gathering information. This uses software such as Apache Nutch or Scrapy to collect relevant information from multiple sources on the Internet (news sites, official website, social media, blogs, etc.). The engine generates a list of URLs as needed and picks out relevant information.
[0388] The collected information is then passed through a filtering system to check its reliability. This system analyzes the reputation of the information source and the consistency of its content, and eliminates unreliable information and hoaxes. For example, anonymous posts on social media and unreliable blogs are excluded.
[0389] After filtering, the information is further sorted chronologically and by category. Specifically, it is organized based on the date and time of collection and divided into categories such as "latest news," "side effects," and "research data" according to its content.
[0390] The information is then formatted using generative AI and laid out in the form of a portal site. Models such as GPT-3 and BERT are used for this generative AI. The generative AI understands the context of the search keywords and arranges the information so that it is easy for users to view. The layout is designed with usability in mind, such as placing the latest news at the top of the page and displaying information about side effects and research data as the main content.
[0391] Furthermore, the server uses an emotion engine to analyze the user's search keywords and behavioral history to recognize the user's emotions. The emotion engine uses natural language processing technology to determine whether the user is anxious. For example, if the user is anxious, it can prioritize the display of information that will give a sense of reassurance (e.g., the effectiveness of vaccines or reassuring statistical data).
[0392] Finally, the formatted information is converted into a format suitable for the user's device (such as HTML or JSON) and sent to the device using the HTTP protocol.
[0393] Terminal handling:
[0394] Users enter keywords for the information they want to find into the search box on their device. For example, they might enter "COVID-19 vaccine information." This information is sent to the server, where it is analyzed and processed.
[0395] The device receives and displays the formatted data sent from the server. This information is displayed in a portal site format, allowing users to easily browse categorized information. For example, the top of the screen displays the latest news, the middle section displays side effect statistics, and the bottom section displays links to research papers. Additionally, information appropriate to the user's current psychological state is displayed based on the analysis results of an emotion engine.
[0396] User Activity:
[0397] Users enter the information they want to know into their device and perform a search. They then check the portal-site-style information displayed on the device and access the information they need. For example, they can browse the latest news about vaccines and check side effect data and scientific research results. Based on the analysis results of the emotion engine, information that is appropriate to the user's emotional state is highlighted, allowing the user to use the information with peace of mind.
[0398] Examples of prompts:
[0399] "New Coronavirus Vaccine Information"
[0400] The flow of the identification process in the second embodiment will be described with reference to FIG.
[0401] Step 1: User enters keywords
[0402] Users enter keywords for the information they want to find into the device's search box, such as "COVID-19 vaccine information."
[0403] Input: The search term entered by the user.
[0404] Output: The data sent to the server as a search request.
[0405] Step 2: Receiving and analyzing search keywords
[0406] The server receives search keywords sent by the user, which are then input to an analysis module and analyzed as search keywords.
[0407] Input: The search keyword submitted by the user.
[0408] Output: Parsed search keywords data format.
[0409] Specifically, the server extracts keywords as string patterns using an analysis module and passes them to the next processing step.
[0410] Step 3: Start the crawling engine and gather information
[0411] The server runs a crawling engine to collect relevant information from online sources (news sites, official websites, social media, blogs, etc.) using crawling engines such as Apache Nutch and Scrapy.
[0412] Input: The parsed search keywords.
[0413] Output: Collected raw information data.
[0414] Specifically, the crawling engine generates a list of URLs to search and sends HTTP requests to collect relevant information.
[0415] Step 4: Filtering the information
[0416] The collected information is then input into a filtering system to check its reliability, which evaluates the source of the information and eliminates unreliable or false information.
[0417] Input: Collected information data.
[0418] Output: filtered reliable information data.
[0419] Specifically, the filtering system evaluates information sources using reputation scores and eliminates information from unreliable social media posts and unreliable blogs.
[0420] Step 5: Categorize and organize information
[0421] The filtered information is sorted chronologically and by category. Specifically, it is organized based on the date and time of collection and sorted into categories such as "latest news," "side effects," and "research data" according to its content.
[0422] Input: Filtered information data.
[0423] Output: Information data organized in categorized and chronological order.
[0424] Specifically, the server uses the information's metadata (collection date and time, information category, etc.) to store the information in a database and then categorizes and organizes it appropriately.
[0425] Step 6: Formatting information with generative AI
[0426] The organized information is then laid out in a portal site format by a generative AI, using models such as GPT-3 and BERT. The generative AI understands the context of search keywords and arranges the information in a way that makes it easy for users to browse.
[0427] Input: Information data organized in categorized and chronological order.
[0428] Output: Information data formatted for a portal site.
[0429] Specifically, the generative AI model analyzes the input data, places the latest news at the top of the page, and arranges information about side effects and research data as the main content.
[0430] Step 7: Recognizing user emotions with the emotion engine
[0431] The server uses an emotion engine to analyze the user's search keywords and behavioral history to recognize the user's emotions. Based on the emotion recognition, the priority of the information to be displayed is changed.
[0432] Input: User search keywords and behavioral history data.
[0433] Output: Emotion recognition results and reprioritized information data.
[0434] Specifically, the emotion engine uses natural language processing technology to analyze the user's emotional state (e.g., worry, relief), and optimizes the display order of information based on the results.
[0435] Step 8: Send and view information
[0436] The formatted information is converted into a format suitable for the user's device (HTML, JSON, etc.) and sent to the device. The device displays the received information in portal site format.
[0437] Input: Information data with changed priority.
[0438] Output: Data sent to the user's device.
[0439] Specifically, the server converts the information into an appropriate format and sends it to the user's device using the HTTP protocol. The device then renders the received data using a display engine, allowing the user to view the information in a portal site format.
[0440] (Application example 2)
[0441] 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."
[0442] In today's information society, a huge amount of information is circulating on the Internet, including unreliable and false information. It is difficult for users to quickly and accurately obtain the information they need. Furthermore, there are few systems that prioritize and provide appropriate information based on the user's search behavior and psychological state. This can cause users to feel stressed or anxious during the information acquisition process. The present invention aims to solve these problems and efficiently provide reliable information that corresponds to the user's emotional state.
[0443] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: means for receiving a search keyword from a user; crawling engine means for collecting information related to the search keyword from online information sources; means for filtering the collected information and eliminating false information; means for classifying the filtered information in chronological order and by category; generation AI means for displaying the generated information in a portal site format; means for providing the information to the user's terminal; emotion engine means for recognizing the user's emotions and changing the priority of the information to be displayed based on the emotions; means for classifying the filtered information by reliability, content, and category and rearranging it in chronological order; and means for prioritizing the generated portal site-formatted information based on the user's emotions. This allows users to efficiently obtain reliable information without stress and make decisions based on appropriate information.
[0444] "User" refers to an end user who uses this system to search for information.
[0445] A "search keyword" is a word or phrase that a user enters to identify information in which they are interested.
[0446] A "crawling engine" is software that automatically collects information related to specified search keywords from various sources on the Internet.
[0447] "Filtering" is the process of evaluating the reliability and relevance of collected information and eliminating or filtering out false or unnecessary information.
[0448] "Chronological order" is a method of ordering information based on the date and time it was collected, with the most recent information being displayed first.
[0449] "Categorization" is the process of separating collected information into different categories based on its content.
[0450] "Generative AI" is artificial intelligence that takes collected and organized information and formats it into a form that users can easily understand and use (e.g., portal site format).
[0451] An "emotion engine" is a system that analyzes a user's input and behavioral history to recognize the user's current emotions.
[0452] "Changing the priority" means appropriately adjusting the display order of information based on the emotional state of the user.
[0453] The "portal site format" is a format that integrates various information and allows users to access all information on a single interface.
[0454] This invention is a system that collects information on the Internet based on search keywords entered by a user, evaluates the reliability of that information, and provides appropriate information according to the user's emotional state. Specific embodiments of this invention are described below.
[0455] System Overview
[0456] The system includes a server that receives search keywords from users and collects related information from the internet, and a terminal that allows users to access the information. The information is classified through reliability evaluation and filtering, and then formatted into a portal site format by a generative AI. In addition, the emotional engine recognizes the user's emotional state and adjusts the priority of the information to be displayed.
[0457] Hardware used
[0458] 1. Server: Oversees information collection, filtering, classification, formatting by generative AI, and processing by the emotion engine.
[0459] 2. Terminal: A device such as a smartphone or tablet that allows users to input, receive, and view information.
[0460] Software used
[0461] 1. Crawling engine: Collects data related to the search keywords from sources.
[0462] 2. Filtering system: Evaluate the reliability of collected data and eliminate false information.
[0463] 3. Classification system: Organize the filtered data chronologically and by category.
[0464] 4. Generative AI: Layout information into a portal site format for easy viewing by users.
[0465] 5. Emotion engine: Analyzes the user's search behavior and history, recognizes the user's emotional state, and adjusts the information that is displayed as a priority.
[0466] Data processing and calculation
[0467] 1. Receiving search keywords:
[0468] The user enters search keywords into the terminal and sends them to the server.
[0469] 2. Collection of Information:
[0470] The crawling engine on the server starts up and collects relevant information from news sites, official websites, social media, blogs, and other sources on the Internet.
[0471] 3. Information filtering:
[0472] The filtering system evaluates the reliability of collected information and filters out false or inaccurate information.
[0473] 4. Classifying and organizing information:
[0474] The filtered information is sorted chronologically and into categories such as "latest news" and "word of mouth."
[0475] 5. Emotion recognition:
[0476] The emotion engine analyzes users' search history and behavior to recognize their emotional state and prioritize information based on the recognized emotion.
[0477] 6. Formatting Information:
[0478] The organized information is laid out in a portal site format by the generation AI and converted into a format suitable for the user's device.
[0479] 7. Transmission and Display of Information:
[0480] The converted data is sent to the user's device and displayed in a portal site format, allowing the user to easily browse the classified information and access the information they need.
[0481] Specific examples
[0482] When a user enters the keyword "COVID-19 vaccine information," the server receives this keyword and the crawling engine collects related information. The filtering system eliminates unreliable information and classifies it into categories such as "latest news," "side effect information," and "research data," and the generative AI layouts the information appropriately. The emotion engine analyzes the user's emotions and prioritizes displaying information that provides reassurance if the user is feeling anxious.
[0483] Example prompt for generative AI model:
[0484] Provide reliable, up-to-date information on the following keyword: "COVID-19 vaccine information." Categorize information into news, reviews, and statistics, and prioritize information that reassures users based on emotion recognition.
[0485] Based on this example, it becomes easier for users to efficiently obtain reliable information and access data categories that have already been specified.
[0486] The flow of the specific processing in the application example 2 will be described with reference to FIG.
[0487] Step 1:
[0488] Input: The user enters a search keyword into the device's search box. Example: "COVID-19 vaccine information."
[0489] How it works: The user presses the search button and a search request is sent to the server.
[0490] Output: The server receives the search keywords and starts the search process.
[0491] Step 2:
[0492] Input: The server is ready to process the search keywords received from the user.
[0493] How it works: The server launches a crawling engine to collect information related to the search keywords from numerous online sources (news sites, official websites, social media, blogs, etc.).
[0494] Output: The server generates an initial list of the relevant information it has collected.
[0495] Step 3:
[0496] Input: The server proceeds based on the initial list of relevant information it has collected.
[0497] How it works: The filtering system is activated and evaluates the reliability of the collected information. Unreliable information and hoaxes are filtered out. For example, anonymous posts on social media and data from low-rated sources are filtered out.
[0498] Output: A filtered list of trusted information is generated.
[0499] Step 4:
[0500] Input: The server receives the filtered list of information.
[0501] How it works: Organizes information chronologically and by category. It categorizes information into breaking news, side effects, research data, etc., and sorts them chronologically within each category.
[0502] Output: A complete list of information organized chronologically and by category.
[0503] Step 5:
[0504] Input: The server performs the following process based on the organized information list.
[0505] How it works: The generative AI starts up and lays out the organized information in a portal site format. To make it easier for users to understand visually, the latest news is placed at the top of the page, side effect information in the middle, and research data at the bottom.
[0506] Output: Information in the form of a portal site is generated, laid out by the generation AI.
[0507] Step 6:
[0508] Input: The server receives the information laid out by the generation AI.
[0509] How it works: The emotion engine analyzes the user's search keywords and behavioral history to recognize their emotions. It then changes the priority of the information displayed based on their emotional state. For example, if the user is worried, it displays reassuring information at the top.
[0510] Output: A prioritized list of information tailored by the emotion engine is created.
[0511] Step 7:
[0512] Input: The server receives the priority list adjusted by the emotion engine.
[0513] What it does: Converts information into the appropriate data format (e.g. HTML, JSON, etc.) to send to the user's device.
[0514] Output: Data is formatted for sending to the terminal.
[0515] Step 8:
[0516] Input: The user's terminal receives the data sent from the server.
[0517] Operation: The device displays the received data in a portal site format. The user browses through the categorized information and accesses the information they need. In addition, the emotion engine prioritizes display, providing information appropriate to the user's psychological state.
[0518] Output: The portal site that the user can view is displayed.
[0519] 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.
[0520] 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.
[0521] 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.
[0522] [Second embodiment]
[0523] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.
[0524] 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.
[0525] 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).
[0526] 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.
[0527] 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.
[0528] 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).
[0529] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.
[0530] 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.
[0531] 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.
[0532] 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.
[0533] 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.
[0534] 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."
[0535] This invention is a system that collects necessary information based on search keywords entered by a user, eliminates unreliable information, and provides the information in a format that is easy for the user to view. Specific embodiments of this system are described below.
[0536] Server Processing
[0537] 1. Receiving search keywords from users
[0538] When a user enters search keywords on their device and presses the search button, the request is sent to the server. The server receives this request and starts the search process. For example, if a user enters "COVID-19 vaccine information," the server analyzes the keywords and prepares to collect related information.
[0539] 2. Starting the crawling engine
[0540] The server starts a crawling engine to collect information related to the search keyword from multiple sources on the Internet (news sites, official websites, social media, blogs, etc.). The engine searches sources as needed to obtain the latest data.
[0541] 3. Information filtering
[0542] The collected information is processed through a filtering system within the server. The filtering system checks the reliability of the information, the reputation of the source, and the consistency of the content, and filters out rumors and misinformation. For example, anonymous posts on social media and information from unreliable sources are filtered out.
[0543] 4. Classification and chronological organization of information
[0544] The filtered information is organized chronologically and by category: specifically, it is sorted based on the date and time of collection and categorized based on the content of the information (e.g., latest news, side effects, research data).
[0545] 5. Information Shaping with Generative AI
[0546] The organized information is then formatted into a portal site by a generative AI. This generative AI understands the meaning of search keywords and determines the layout so that users can easily browse related information. For example, the latest news is placed at the top of the page, and information on side effects and research data is displayed as the main content.
[0547] 6. Transmission of Information
[0548] The formatted information is sent to the user's device using web technologies such as HTML and JSON in a format suitable for the user's device.
[0549] Terminal handling
[0550] 1. Enter search keywords
[0551] Users enter keywords related to the information they want to find into the device's search box. For example, they enter "COVID-19 vaccine information."
[0552] 2. Receiving and Displaying Information
[0553] The terminal receives the formatted information sent from the server. This information is displayed in a portal site format, allowing the user to easily browse the categorized information.
[0554] User Activity
[0555] 1. Enter keywords
[0556] The user inputs the information they want to know into the device. For example, they input specific keywords such as "COVID-19 vaccine information."
[0557] 2. Verification and Use of Information
[0558] They can browse the displayed portal-style information and access the information they need, for example, by reading the latest news about vaccines, checking side-effect data and scientific research results.
[0559] Specific examples
[0560] (Search keyword: "COVID-19 vaccine information")
[0561] 1. Receiving a request
[0562] The server receives a search request for "COVID-19 vaccine information."
[0563] 2. Starting the crawling engine
[0564] The crawling engine starts up and collects information about vaccines from news sites, official websites, social media, etc.
[0565] 3. Information filtering
[0566] The filtering system evaluates the reliability of collected information and filters out false information, for example, giving priority to information from reliable news sites and official organizations.
[0567] 4. Classifying and Formatting Information
[0568] The filtered information is categorized into the latest news, side effect data, research results, etc. and organized in chronological order. Generative AI then formats this information into a portal site format and prepares it for delivery to users.
[0569] 5. Provision of Information
[0570] The formatted information is sent to the user's terminal, where the user can view the information displayed in a portal site format and access the information they need.
[0571] As described above, this invention is designed to enable users to efficiently obtain the information they need. By eliminating false information and categorizing information by time series and category, reliability and ease of use are improved. This system allows users to quickly access appropriate information.
[0572] The processing flow will be explained below.
[0573] Server Processing
[0574] Step 1:
[0575] The server receives the search keywords sent by the user, along with the user's settings information (language settings, region settings, etc.).
[0576] Step 2:
[0577] The server starts a crawling engine to collect information, and the crawling engine prepares to collect information related to the specified keyword from a wide range of sources.
[0578] Step 3:
[0579] The crawling engine accesses predefined sources such as news sites, official sites, social media, blogs, etc. to gather information related to the search keyword, exploring multiple sources in parallel during this process.
[0580] Step 4:
[0581] The collected information passes through a filtering system within the server, which evaluates the reliability of the information and filters out hoaxes and unreliable information. For example, unreliable anonymous social media posts are filtered out.
[0582] Step 5:
[0583] The filtered information is sorted chronologically and by category. Specifically, it is organized in chronological order based on the date and time the information was collected, and then divided into categories such as "latest news," "side effects," and "research data" according to its content.
[0584] Step 6:
[0585] The AI then organizes the information and lays it out in a portal site format. The AI understands the meaning of the search keywords and places relevant information in key positions. For example, the latest news is placed at the top of the page, and detailed research results are displayed in the center.
[0586] Step 7:
[0587] The formatted information is then converted into a format appropriate for the user's device (such as HTML or JSON), and the converted data is ready to be sent to the user's device.
[0588] Terminal handling
[0589] Step 1:
[0590] The user enters keywords into the search box on the device and presses the search button. This input information is sent to the server.
[0591] Step 2:
[0592] The terminal receives the formatted information sent from the server, parses the received data, and prepares it for display.
[0593] Step 3:
[0594] The device displays the received information in a portal site format, with the latest news displayed at the top of the screen, side effect statistics in the middle, and links to research papers at the bottom.
[0595] User Action
[0596] Step 1:
[0597] Users enter the keywords they want to know in the search box on their device and press the search button. For example, they could enter "COVID-19 vaccine information."
[0598] Step 2:
[0599] Users can check the information displayed on their devices in the form of a portal site and access the information they need, for example, by clicking on the latest news headline to read a detailed article.
[0600] The above is a specific processing flow of the present invention. This system allows users to obtain the information they need efficiently and quickly.
[0601] Example 1
[0602] 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."
[0603] There is a lot of information scattered on the Internet, making it difficult for users to quickly and accurately obtain the information they need. Furthermore, the Internet sometimes contains inaccurate or false information, making it difficult for users to access reliable information. There is a particular need for a system that can efficiently provide information for urgent situations or when specialized information is required.
[0604] 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.
[0605] In this invention, the server includes means for receiving search keywords from users, a crawling engine for collecting information related to the search keywords from online information sources, means for filtering the collected information and eliminating false information, means for classifying the filtered information in chronological order and by category, means for formatting the classified information into a portal site format using a generative AI model, and means for transmitting the information to the user's terminal, thereby enabling users to quickly and efficiently obtain reliable information.
[0606] The "means for receiving search keywords from a user" refers to a mechanism for transmitting search keywords entered by a user via an Internet terminal to a server and receiving the keywords.
[0607] The "crawling engine means" is a program that automatically collects information related to a specified keyword from multiple information sources on the Internet.
[0608] A "means for filtering out misinformation" is an algorithm or system that detects and filters out unreliable or false information from collected information.
[0609] "Chronological and categorical sorting" refers to the process of sorting the filtered information in chronological order based on the date and time of collection and categorizing it according to its content.
[0610] "Means for formatting into a portal site format using a generative AI model" refers to a technology that uses an artificial intelligence model to reconstruct classified information into a portal site format that is easy for users to view.
[0611] "Means for sending information to the user's device" refers to a mechanism for sending formatted information to the user's device using an appropriate communication means (e.g., HTML, JSON).
[0612] This invention is a system that collects necessary information based on search keywords entered by a user, eliminates unreliable information, and provides the information in a format that is easy for the user to view. Specific embodiments of this system are described below.
[0613] Server Processing
[0614] Receiving search keywords from users
[0615] When a user enters specific keywords, such as "COVID-19 vaccine information," into their device and presses the search button, the request is sent to the server. The server receives the HTTP request and extracts the search keywords from it. This prepares the system to begin collecting information.
[0616] Starting the crawling engine
[0617] The server runs a crawling engine using Python tools such as BeautifulSoup and Scrapy to collect information from multiple sources on the Internet. Based on the input keywords, the crawling engine searches news sites, official websites, social media, blogs, etc. to obtain the latest relevant data.
[0618] Information Filtering
[0619] The collected information is stored in a database, and then processed by a filtering system on the server. The filtering system evaluates the reliability of the source and its content, and runs an algorithm to filter out misinformation and unreliable information. Based on the source's reliability score, information with a score below a certain level is filtered out.
[0620] Classification and chronological organization of information
[0621] The filtered information is sorted chronologically and by category (e.g., latest news, side effects, research data). The server retrieves this information from the database and sorts it chronologically using an SQL query.
[0622] Shaping information with generative AI
[0623] The classified information is then formatted into a portal site using a generative AI model (e.g., GPT-4). The generative AI understands the context of the search keywords and generates a user-friendly layout. For example, the latest news is placed at the top of the page, and side effects and research data are placed in different sections.
[0624] Sending information
[0625] The formatted information is sent to the user's device in HTML or JSON format. The server generates an HTTP response and sends the information to the user's browser. The browser receives the response and displays the information.
[0626] Terminal handling
[0627] Enter search keywords
[0628] The user enters the information of interest (e.g., "COVID-19 vaccine information") into the search box on the device and presses the search button. The device then sends the user's input to the server.
[0629] Receiving and displaying information
[0630] The user's device receives the formatted information sent from the server and displays it on the screen. Information sent in HTML format is rendered by the browser and displayed so that the user can easily check the content. In the case of JSON format, JavaScript manipulates the DOM to dynamically display the information.
[0631] User Activity
[0632] Enter keywords
[0633] Users enter keywords related to the information they want to know into the device. For example, they can enter "COVID-19 vaccine information."
[0634] Verifying and Using Information
[0635] Users can review the displayed information and access the information they need, such as reading the latest vaccine news or viewing data and research results on side effects.
[0636] Specific examples
[0637] Search keyword: "COVID-19 vaccine information"
[0638] 1. Request reception: The server receives a search request for "COVID-19 vaccine information" and analyzes the keywords.
[0639] 2. Launching the crawling engine: Collect information about vaccines from news sites, official websites, social media, etc.
[0640] 3. Information filtering: Select reliable information and eliminate misinformation.
[0641] 4. Classification and formatting of information: The information is classified into the latest news, side effect data, research results, etc. and organized in chronological order. Generative AI formats this information into a portal site format.
[0642] 5. Providing information: The formatted information is sent to the user's device and displayed in portal site format.
[0643] As described above, the present invention is a system that efficiently collects information required by users and provides it in a highly reliable manner.
[0644] The flow of the identification process in the first embodiment will be described with reference to FIG.
[0645] Step 1:
[0646] Receiving search keywords from users
[0647] A user enters "COVID-19 vaccine information" into the search box on their device and presses the search button. The device generates an HTTP request and sends it to the server. The server receives this request and extracts the search keywords. It then prepares to collect information based on the analyzed keywords (Input: User's search keyword "COVID-19 vaccine information", Output: Analyzed search keywords).
[0648] Step 2:
[0649] Starting the crawling engine
[0650] The server starts a Python crawling engine (e.g., BeautifulSoup or Scrapy) using the parsed search keywords as input. This engine searches multiple sources (news sites, official website, social media, blogs, etc.) for information related to the specified keywords and obtains HTML data. The crawling engine then crawls these sites and returns the collected HTML content to the server (input: parsed search keywords, output: collected HTML data).
[0651] Step 3:
[0652] Information Filtering
[0653] The filtering system in the server receives the HTML data collected in the previous step as input. The filtering system runs an algorithm to evaluate the reliability of the information and eliminate hoaxes and unreliable information. For example, it analyzes the source domain and calculates a reliability score. Information from unreliable sites and anonymous social media posts will be eliminated due to their low scores. (Input: Collected HTML data, Output: Filtered information)
[0654] Step 4:
[0655] Classification and chronological organization of information
[0656] The server then sorts the filtered information into chronological order and categories. This process uses SQL queries to retrieve information stored in the database, sorts it based on the collection date and time, and categorizes it by content. Specifically, the information is classified into latest news, side effects, and research data, and then organized in chronological order within each category (input: filtered information, output: classified and organized information).
[0657] Step 5:
[0658] Shaping information with generative AI
[0659] The server takes the classified and organized information as input and uses a generative AI model (e.g., GPT-4) to format it into a portal site. The generative AI model understands the context of the search keywords and lays it out in a format that is easy for users to read. For example, the latest news is placed at the top of the page, and side effects and research data are placed in dedicated sections (input: classified and organized information, output: formatted HTML content).
[0660] Step 6:
[0661] Sending information
[0662] The formatted HTML content is sent from the server to the user's device. The server generates an HTTP response and delivers the information to the user's device in HTML or JSON format. The user's browser receives this and displays it optimally (input: formatted HTML content, output: information sent to the user's device).
[0663] Step 7:
[0664] Receiving and displaying information
[0665] The user's device receives the HTML or JSON data sent from the server and renders it in the browser. In the case of HTML format, the browser's rendering engine displays the page, allowing the user to easily browse the categorized information. In the case of JSON format, JavaScript manipulates the DOM to dynamically display the information (input: information sent from the server, output: information displayed on the device).
[0666] Step 8:
[0667] User review and use of information
[0668] The user checks the displayed information and accesses the information they need, such as reading the latest news about vaccines, checking side effect data, and scientific research results (input: information displayed on the device, output: information obtained by the user).
[0669] Through these steps, the system is able to efficiently collect and organize reliable information and provide it to users.
[0670] (Application example 1)
[0671] 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."
[0672] Conventional information retrieval systems have issues such as low reliability of information collected based on search keywords and lack of an appropriate display format for users to easily view and use information. Furthermore, especially when providing information for smart glasses and other wearable devices, the lack of a voice-based interface limits the user experience. To address these issues, a system is needed that can provide real-time, reliable information and display that information in an optimized audio and visual format.
[0673] 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.
[0674] In this invention, the server includes means for receiving search keywords from users, crawling engine means for collecting information related to the search keywords from online sources, means for filtering the collected information and eliminating false information, means for classifying the filtered information in chronological order and by type, generation AI means for displaying the generated information in a format suitable for a display device, means for analyzing voice input and generating search keywords, a speech synthesizer for outputting the classified information by voice, and means for providing the information to the user's display device. This allows users to obtain reliable information in real time and easily view and use it in a format optimized for audio and visuals.
[0675] A "user" is an entity that searches for and uses information.
[0676] "Search keywords" are words or phrases related to the information a user wants to find.
[0677] "Online sources" are information sources such as websites and databases that can be accessed via the Internet.
[0678] A "crawling engine" is software that automatically collects specific information from online sources.
[0679] "Filtering" is the process of removing unreliable information or false information from collected information.
[0680] "Chronological order" refers to arranging information in order based on the date and time it was collected.
[0681] "By type" refers to classifying information into categories according to its content.
[0682] "Generative AI" is artificial intelligence that formats collected and filtered information into a format suitable for the user's device or display device.
[0683] "Voice input" is a method in which a user provides information to a system by speaking into a microphone.
[0684] A "speech synthesizer" is a device that converts text information into speech and makes it audible to the user.
[0685] A "display device" is a device for visually presenting collected and generated information to a user.
[0686] This invention is a system that collects necessary information based on search keywords entered by the user through input or voice, eliminates unreliable information, and uses a generation AI to provide the information in a format suitable for the user's device. Specific embodiments of this system are described below.
[0687] Server Processing
[0688] 1. Receiving search keywords from users
[0689] The user enters a search keyword by voice or text, and the request is sent to the server. An example search keyword is "new spring clothes." The server receives this request and starts the search process.
[0690] 2. Starting the crawling engine
[0691] The server uses a crawling engine to collect information related to the search keywords from online sources, such as fashion websites, official brand websites, and social media.
[0692] 3. Information filtering
[0693] The information collected by the crawling engine is processed through a filtering system within the server. The filtering system evaluates the reliability of the information and removes unnecessary data such as false information. For example, anonymous social media posts and data from less reliable sources are excluded.
[0694] 4. Classification and chronological organization of information
[0695] The server then organizes the filtered information chronologically and by type, specifically into categories like tops, bottoms, and outerwear.
[0696] 5. Information Shaping with Generative AI
[0697] The organized information is then formatted into a suitable format for the display device by the generation AI, which understands the meaning of the search keywords and determines the layout so that the user can easily view related information.
[0698] 6. Transmission of Information
[0699] The formatted information is sent from the server to the user's display device using web technologies such as HTML and JSON.
[0700] Terminal handling
[0701] 1. Enter search keywords
[0702] The user inputs keywords related to the information they want to search for into a search box on the device or provides them by voice. For example, the user inputs or provides by voice a specific keyword such as "new spring clothes."
[0703] 2. Receiving and Displaying Information
[0704] The terminal receives the formatted information sent from the server, displays the information in a format optimal for the display device, and uses a speech synthesizer to provide the information to the user aloud.
[0705] User Activity
[0706] 1. Enter search keywords
[0707] Users input the information they want to know into the terminal. Voice input is also supported, so for example, keywords such as "new spring clothes" can be provided by voice.
[0708] 2. Verification and Use of Information
[0709] Users can access the information they need by viewing the information displayed in a portal format on the display device. Audio output is also provided, allowing users to view and hear the information.
[0710] Specific examples
[0711] For example, if a user uses smart glasses to say, "I want new spring clothes," the system will analyze the speech, collect relevant fashion information, filter and organize it, and use the generative AI to visually display the information in a format that is optimal for the smart glasses, while the speech synthesizer will provide the information aloud.
[0712] Prompt Sentence Examples
[0713] What's the latest information on spring fashion?
[0714] The hardware used includes smart glasses (e.g., microphone, speaker, display), and the software includes Python, Requests, BeautifulSoup, OpenAI GPT-3, and google_speech, allowing users to efficiently receive up-to-date, reliable information through visual and audio.
[0715] The flow of the specific processing in the application example 1 will be described with reference to FIG.
[0716] Step 1:
[0717] Enter and submit search keywords
[0718] The user enters search keywords using the voice input function of the smart glasses.
[0719] For example: "I want some new spring clothes."
[0720] Speech recognition software converts this voice input into text data.
[0721] The converted text data is sent to the server.
[0722] Step 2:
[0723] Receiving search keywords
[0724] The server receives the text data sent by the user.
[0725] Input: A keyword spoken by the user (e.g., "new spring clothes")
[0726] Output: Search keyword data for analysis
[0727] Step 3:
[0728] Starting the crawling engine
[0729] The server launches a crawling engine targeting multiple online information sources (fashion sites, official brand websites, social media, etc.).
[0730] Input: Search keyword ("new spring clothes")
[0731] Data processing: Obtain relevant information from sources and collect it as text data.
[0732] Output: Collected text data
[0733] Step 4:
[0734] Information Filtering
[0735] The server evaluates the reliability of the collected information using a filtering system.
[0736] Input: Collected text data
[0737] Data calculation: Remove unreliable information and false information.
[0738] What it does: Identify and filter out low-trust sources (e.g., anonymous social media posts).
[0739] Output: Filtered reliable information data
[0740] Step 5:
[0741] Classification and chronological organization of information
[0742] The server categorizes the filtered information chronologically and by type (tops, bottoms, outerwear, etc.).
[0743] Input: Filtered reliable information data
[0744] Data processing: Sort the information based on the date and time it was collected and divide it into categories.
[0745] Output: Categorized and organized information data
[0746] Step 6:
[0747] Shaping information with generative AI
[0748] The server uses generative AI to format the organized information into a format suitable for display on smart glasses.
[0749] Input: Classified and organized information data
[0750] Data calculation: Generative AI determines the optimal layout for the user and converts it into a format that is easy to display visually.
[0751] What it does: Sorts information by category, with the most recent information at the top.
[0752] Output: Formatted information data
[0753] Step 7:
[0754] Sending information
[0755] The server sends the formatted information to the user's smart glasses.
[0756] Input: Formatted information data
[0757] Data processing: Converts data into a transmission format (e.g., HTML, JSON) and executes transmission logic.
[0758] Output: Information data sent to the user terminal
[0759] Step 8:
[0760] Display and speech output of information
[0761] The terminal provides the received information to the user visually and audibly.
[0762] Input: Information data sent from the server
[0763] Specific operation: Information is displayed on the HUD of the smart glasses, and at the same time, the voice synthesizer plays the information aloud.
[0764] Output: Visual and audio information provided to the user
[0765] 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.
[0766] The present invention is a system that collects information on the Internet based on search keywords entered by users, eliminates unreliable information, and recognizes the user's emotions using an emotion engine to provide the user with appropriate information in the form of a portal site. A specific embodiment of this system is described below.
[0767] Server Processing
[0768] 1. Receiving search keywords from users
[0769] When a user enters search keywords on their device and presses the search button, the request is sent to the server. The server receives this request and starts the search process. For example, if a user enters "COVID-19 vaccine information," the server analyzes the keywords and prepares to collect related information.
[0770] 2. Starting the crawling engine
[0771] The server starts a crawling engine to collect information related to the search keyword from multiple sources on the Internet (news sites, official websites, social media, blogs, etc.). The engine searches sources as needed to obtain the latest data.
[0772] 3. Information filtering
[0773] The collected information is processed through a filtering system within the server. The filtering system checks the reliability of the information, the reputation of the source, and the consistency of the content, and filters out rumors and misinformation. For example, anonymous posts on social media and information from unreliable sources are filtered out.
[0774] 4. Classification and chronological organization of information
[0775] The filtered information is organized chronologically and by category. Specifically, it is organized in chronological order based on the date and time of collection, and then divided into categories such as "latest news," "side effects," and "research data" according to its content.
[0776] 5. Information Shaping with Generative AI
[0777] The organized information is then laid out in a portal site format by a generative AI. This generative AI understands the meaning of search keywords and arranges relevant information so that users can easily browse it. For example, the latest news is placed at the top of the page, and information on side effects and research data are displayed as the main content.
[0778] 6. User Emotion Recognition by Emotion Engine
[0779] The server uses an emotion engine to analyze the user's search keywords and behavioral history to recognize the user's emotions. Based on this emotion recognition, the priority of the information displayed is changed. For example, if the user is worried, information that gives a sense of security is displayed at the top.
[0780] 7. Transmission of Information
[0781] The formatted information is then converted into a format appropriate for the user's device (such as HTML or JSON), and the converted data is ready to be sent to the user's device.
[0782] Terminal handling
[0783] 1. Enter search keywords
[0784] Users enter keywords related to the information they want to find into the device's search box. For example, they enter "COVID-19 vaccine information."
[0785] 2. Receiving and Displaying Information
[0786] The device receives the formatted information sent from the server. This information is displayed in a portal site format, allowing users to easily browse categorized information. For example, the top of the screen displays the latest news, the middle section displays side effect statistics, and the bottom section displays links to research papers. In addition, information appropriate to the user's current emotional state is displayed based on the analysis results of the emotion engine.
[0787] User Activity
[0788] 1. Enter keywords
[0789] The user inputs the information they want to know into the device. For example, they input specific keywords such as "COVID-19 vaccine information."
[0790] 2. Verification and Use of Information
[0791] Users can review the displayed portal-style information and access the information they need. For example, they can read the latest news about vaccines, check side effect data, and check scientific research results. Based on the analysis results of the emotion engine, information appropriate to the user's emotional state is recommended, providing information to reduce stress.
[0792] Specific examples
[0793] (Search keyword: "COVID-19 vaccine information")
[0794] 1. Receiving a request
[0795] The server receives a search request for "COVID-19 vaccine information."
[0796] 2. Starting the crawling engine
[0797] The crawling engine starts up and collects information about vaccines from news sites, official websites, social media, etc.
[0798] 3. Information filtering
[0799] The filtering system evaluates the reliability of collected information and filters out false information, for example, giving priority to information from reliable news sites and official organizations.
[0800] 4. Classifying and Formatting Information
[0801] The filtered information is categorized into the latest news, side effect data, research results, etc. and organized in chronological order. Generative AI then formats this information into a portal site format and prepares it for delivery to users.
[0802] 5. Emotion Recognition by Emotion Engine
[0803] If a user is worried, information that provides reassurance (e.g., vaccine effectiveness or reassuring statistics) will be prioritized based on that emotion.
[0804] 6. Transmission of Information
[0805] The formatted information is sent to the user's terminal, where the user can view the information displayed in a portal site format and access the information they need.
[0806] As described above, this invention is designed to enable users to efficiently obtain the information they need. In addition to eliminating false information and classifying information by time series and category, reliability, ease of use, and psychological compatibility are improved by recognizing user emotions using an emotion engine. This system allows users to quickly access appropriate information.
[0807] The processing flow will be explained below.
[0808] Server Processing
[0809] Step 1:
[0810] The server receives the search keywords sent by the user, along with the user's settings information (language settings, region settings, etc.).
[0811] Step 2:
[0812] The server starts a crawling engine to collect information, and the crawling engine prepares to collect information related to the specified keyword from a wide range of sources.
[0813] Step 3:
[0814] The crawling engine accesses predefined sources such as news sites, official sites, social media, blogs, etc. to gather information related to the search keyword, exploring multiple sources in parallel during this process.
[0815] Step 4:
[0816] The collected information passes through a filtering system within the server, which evaluates the reliability of the information and filters out hoaxes and unreliable information. For example, unreliable anonymous social media posts are filtered out.
[0817] Step 5:
[0818] The filtered information is sorted chronologically and by category. Specifically, it is organized in chronological order based on the date and time the information was collected, and then divided into categories such as "latest news," "side effects," and "research data" according to its content.
[0819] Step 6:
[0820] The AI then organizes the information and lays it out in a portal site format. The AI understands the meaning of the search keywords and places relevant information in key positions. For example, the latest news is placed at the top of the page, and detailed research results are displayed in the center.
[0821] Step 7:
[0822] The server starts the emotion engine, which analyzes the user's search keywords and history to recognize the user's emotions. Based on the recognition results, the display priority is changed.
[0823] Step 8:
[0824] The formatted information is then converted into a format appropriate for the user's device (such as HTML or JSON), and the converted data is ready to be sent to the user's device.
[0825] Terminal handling
[0826] Step 1:
[0827] The user enters keywords into the search box on the device and presses the search button. This input information is sent to the server.
[0828] Step 2:
[0829] The terminal receives the formatted information sent from the server, parses the received data, and prepares it for display.
[0830] Step 3:
[0831] The device displays the received information in a portal site format. For example, the top of the screen displays the latest news, side effect statistics in the middle, and links to research papers in the bottom. Based on the analysis results of the emotion engine, the device also displays information appropriate to the user's current psychological state.
[0832] User Action
[0833] Step 1:
[0834] Users enter the keywords they want to know in the search box on their device and press the search button. For example, they could enter "COVID-19 vaccine information."
[0835] Step 2:
[0836] Users can check the portal site-style information displayed on their device and access the information they need. For example, they can click on the latest news headline to read a detailed article. They can also view recommended reassuring information based on the analysis results of the emotion engine.
[0837] Specific examples
[0838] (Search keyword: "COVID-19 vaccine information")
[0839] Server details:
[0840] Step 1:
[0841] The server receives a search request for "COVID-19 vaccine information."
[0842] Step 2:
[0843] The crawling engine starts up and collects information about vaccines from news sites, official websites, social media, etc.
[0844] Step 3:
[0845] The collected information is filtered out using a filtering system to remove false information, such as anonymous posts on social media and unreliable information.
[0846] Step 4:
[0847] The filtered information is categorized into the latest news, side effect data, research results, etc. and organized in chronological order.
[0848] Step 5:
[0849] The generating AI lays out the classified information in a portal site format, placing important information at the top.
[0850] Step 6:
[0851] The server activates an emotion engine, analyzes the user's search history and behavior, and prioritizes displaying information that will reassure worried users.
[0852] Step 7:
[0853] The formatted information is converted into HTML or JSON format and sent to the user's device.
[0854] Detailed terminal processing:
[0855] Step 1:
[0856] The user enters "COVID-19 vaccine information" and presses the search button.
[0857] Step 2:
[0858] The terminal receives the information sent from the server and analyzes it.
[0859] Step 3:
[0860] The terminal displays the received information in a portal site format, and displays information according to the user's emotions with priority.
[0861] User detailed processing:
[0862] Step 1:
[0863] A user enters "COVID-19 vaccine information" into the search box and presses the search button.
[0864] Step 2:
[0865] Users can check the displayed information in a portal site format and click on the item they are interested in to see more details. For example, a worried user will be given priority in seeing reassuring information demonstrating the effectiveness of vaccines.
[0866] As described above, the present invention not only enables users to efficiently obtain the information they need, but also improves user satisfaction and reliability, since the information provided by the emotion engine is adapted to the user's emotional state.
[0867] Example 2
[0868] 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."
[0869] In today's Internet environment, a huge amount of information is generated every day, making it difficult for users to quickly obtain reliable information from that information. Furthermore, if the information is unreliable or contains false information, users may act based on the incorrect information. Furthermore, conventional information gathering systems lack a mechanism for changing the priority of information taking into account the user's emotions, which makes it difficult for users to quickly access the information they want to see.
[0870] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving a search keyword from a user, information collection means for collecting information related to the search keyword from online information sources, means for filtering the collected information and eliminating unreliable information, means for classifying the filtered information in chronological order and by category, generation AI means for displaying the generated information in a portal site format, emotion recognition means for recognizing the user's emotion and changing the priority of information based on the emotion, and means for providing information to the user's terminal. This allows the user to quickly access reliable information and provides appropriate information according to the user's emotional state, thereby improving the convenience and reliability of information.
[0871] The "means for receiving search keywords from a user" is a function that allows the server to receive search keywords that a user has input on a terminal.
[0872] "Information gathering means for gathering information related to search keywords from online sources" is a function for gathering information related to specific search keywords from multiple sources on the Internet.
[0873] "Means for filtering collected information and eliminating unreliable information" refers to a function for removing unreliable information or false information from the collected information.
[0874] "Means for classifying filtered information chronologically and by category" refers to a function for organizing filtered information in chronological order based on the date and time of collection and classifying it into different categories depending on its content.
[0875] "Generative AI means for displaying generated information in a portal site format" is a function that utilizes artificial intelligence technology to format organized and classified information into a portal site format and display it in a way that is easy for users to view.
[0876] "Emotion recognition means that recognizes the user's emotions and changes the priority of information based on those emotions" is a function that analyzes the user's search keywords and behavioral history to recognize the user's emotions, and dynamically changes the priority of the information to be displayed based on the results.
[0877] The "means for providing information to a user's terminal" is a function for transmitting formatted information to a user's terminal in an appropriate format.
[0878] This invention is a system that collects information on the Internet based on search keywords entered by users, eliminates unreliable information, recognizes the user's emotions using an emotion engine, and provides appropriate information in the form of a portal site. This system is composed of three entities: a server, a terminal, and a user.
[0879] Server Action:
[0880] First, the server receives the search keywords entered by the user on the device. For example, if a specific keyword such as "COVID-19 vaccine information" is entered, the server analyzes this information using a dedicated analysis module.
[0881] The server then launches a crawling engine to gather relevant information from multiple sources on the Internet (news sites, official websites, social media, blogs, etc.) using software such as Apache Nutch or Scrapy. The engine generates a list of URLs as needed and selects relevant information.
[0882] The collected information is then passed through a filtering system to check its reliability. This system analyzes the reputation of the information source and the consistency of its content, and eliminates unreliable information and hoaxes. For example, anonymous posts on social media and unreliable blogs are excluded.
[0883] After filtering, the information is further sorted chronologically and by category. Specifically, it is organized based on the date and time of collection and divided into categories such as "latest news," "side effects," and "research data" according to its content.
[0884] The information is then formatted using generative AI and laid out in the form of a portal site. Models such as GPT-3 and BERT are used for this generative AI. The generative AI understands the context of the search keywords and arranges the information so that it is easy for users to view. The layout is designed with usability in mind, such as placing the latest news at the top of the page and displaying information about side effects and research data as the main content.
[0885] Furthermore, the server uses an emotion engine to analyze the user's search keywords and behavioral history to recognize the user's emotions. The emotion engine uses natural language processing technology to determine whether the user is anxious. For example, if the user is anxious, it can prioritize the display of information that will give a sense of reassurance (e.g., the effectiveness of vaccines or reassuring statistical data).
[0886] Finally, the formatted information is converted into a format suitable for the user's device (such as HTML or JSON) and sent to the device using the HTTP protocol.
[0887] Terminal handling:
[0888] Users enter keywords for the information they want to find into the search box on their device. For example, they might enter "COVID-19 vaccine information." This information is sent to the server, where it is analyzed and processed.
[0889] The device receives and displays the formatted data sent from the server. This information is displayed in a portal site format, allowing users to easily browse categorized information. For example, the top of the screen displays the latest news, the middle section displays side effect statistics, and the bottom section displays links to research papers. Additionally, information appropriate to the user's current psychological state is displayed based on the analysis results of an emotion engine.
[0890] User Activity:
[0891] Users enter the information they want to know into their device and perform a search. They then check the portal-site-style information displayed on the device and access the information they need. For example, they can browse the latest news about vaccines and check side effect data and scientific research results. Based on the analysis results of the emotion engine, information that is appropriate to the user's emotional state is highlighted, allowing the user to use the information with peace of mind.
[0892] Examples of prompts:
[0893] "New Coronavirus Vaccine Information"
[0894] The flow of the identification process in the second embodiment will be described with reference to FIG.
[0895] Step 1: User enters keywords
[0896] Users enter keywords for the information they want to find into the device's search box, such as "COVID-19 vaccine information."
[0897] Input: The search term entered by the user.
[0898] Output: The data sent to the server as a search request.
[0899] Step 2: Receiving and analyzing search keywords
[0900] The server receives search keywords sent by the user, which are then input to an analysis module and analyzed as search keywords.
[0901] Input: The search keyword submitted by the user.
[0902] Output: Parsed search keywords data format.
[0903] Specifically, the server extracts keywords as string patterns using an analysis module and passes them to the next processing step.
[0904] Step 3: Start the crawling engine and gather information
[0905] The server runs a crawling engine to collect relevant information from online sources (news sites, official websites, social media, blogs, etc.) using crawling engines such as Apache Nutch and Scrapy.
[0906] Input: The parsed search keywords.
[0907] Output: Collected raw information data.
[0908] Specifically, the crawling engine generates a list of URLs to search and sends HTTP requests to collect relevant information.
[0909] Step 4: Filtering the information
[0910] The collected information is then input into a filtering system to check its reliability, which evaluates the source of the information and eliminates unreliable or false information.
[0911] Input: Collected information data.
[0912] Output: filtered reliable information data.
[0913] Specifically, the filtering system evaluates information sources using reputation scores and eliminates information from unreliable social media posts and unreliable blogs.
[0914] Step 5: Categorize and organize information
[0915] The filtered information is sorted chronologically and by category. Specifically, it is organized based on the date and time of collection and sorted into categories such as "latest news," "side effects," and "research data" according to its content.
[0916] Input: Filtered information data.
[0917] Output: Information data organized in categorized and chronological order.
[0918] Specifically, the server uses the information's metadata (collection date and time, information category, etc.) to store the information in a database and then categorizes and organizes it appropriately.
[0919] Step 6: Formatting information with generative AI
[0920] The organized information is then laid out in a portal site format by a generative AI, using models such as GPT-3 and BERT. The generative AI understands the context of search keywords and arranges the information in a way that makes it easy for users to browse.
[0921] Input: Information data organized in categorized and chronological order.
[0922] Output: Information data formatted for a portal site.
[0923] Specifically, the generative AI model analyzes the input data, places the latest news at the top of the page, and arranges information about side effects and research data as the main content.
[0924] Step 7: Recognizing user emotions with the emotion engine
[0925] The server uses an emotion engine to analyze the user's search keywords and behavioral history to recognize the user's emotions. Based on the emotion recognition, the priority of the information to be displayed is changed.
[0926] Input: User search keywords and behavioral history data.
[0927] Output: Emotion recognition results and reprioritized information data.
[0928] Specifically, the emotion engine uses natural language processing technology to analyze the user's emotional state (e.g., worry, relief), and optimizes the display order of information based on the results.
[0929] Step 8: Send and view information
[0930] The formatted information is converted into a format suitable for the user's device (HTML, JSON, etc.) and sent to the device. The device displays the received information in portal site format.
[0931] Input: Information data with changed priority.
[0932] Output: Data sent to the user's device.
[0933] Specifically, the server converts the information into an appropriate format and sends it to the user's device using the HTTP protocol. The device then renders the received data using a display engine, allowing the user to view the information in a portal site format.
[0934] (Application example 2)
[0935] 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."
[0936] In today's information society, a huge amount of information is circulating on the Internet, including unreliable and false information. It is difficult for users to quickly and accurately obtain the information they need. Furthermore, there are few systems that prioritize and provide appropriate information based on the user's search behavior and psychological state. This can cause users to feel stressed or anxious during the information acquisition process. The present invention aims to solve these problems and efficiently provide reliable information that corresponds to the user's emotional state.
[0937] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: means for receiving a search keyword from a user; crawling engine means for collecting information related to the search keyword from online information sources; means for filtering the collected information and eliminating false information; means for classifying the filtered information in chronological order and by category; generation AI means for displaying the generated information in a portal site format; means for providing the information to the user's terminal; emotion engine means for recognizing the user's emotions and changing the priority of the information to be displayed based on the emotions; means for classifying the filtered information by reliability, content, and category and rearranging it in chronological order; and means for prioritizing the generated portal site-formatted information based on the user's emotions. This allows users to efficiently obtain reliable information without stress and make decisions based on appropriate information.
[0938] "User" refers to an end user who uses this system to search for information.
[0939] A "search keyword" is a word or phrase that a user enters to identify information in which they are interested.
[0940] A "crawling engine" is software that automatically collects information related to specified search keywords from various sources on the Internet.
[0941] "Filtering" is the process of evaluating the reliability and relevance of collected information and eliminating or filtering out false or unnecessary information.
[0942] "Chronological order" is a method of ordering information based on the date and time it was collected, with the most recent information being displayed first.
[0943] "Categorization" is the process of separating collected information into different categories based on its content.
[0944] "Generative AI" is artificial intelligence that takes collected and organized information and formats it into a form that users can easily understand and use (e.g., portal site format).
[0945] An "emotion engine" is a system that analyzes a user's input and behavioral history to recognize the user's current emotions.
[0946] "Changing the priority" means appropriately adjusting the display order of information based on the emotional state of the user.
[0947] The "portal site format" is a format that integrates various information and allows users to access all information on a single interface.
[0948] This invention is a system that collects information on the Internet based on search keywords entered by a user, evaluates the reliability of that information, and provides appropriate information according to the user's emotional state. Specific embodiments of this invention are described below.
[0949] System Overview
[0950] The system includes a server that receives search keywords from users and collects related information from the internet, and a terminal that allows users to access the information. The information is classified through reliability evaluation and filtering, and then formatted into a portal site format by a generative AI. In addition, the emotional engine recognizes the user's emotional state and adjusts the priority of the information to be displayed.
[0951] Hardware used
[0952] 1. Server: Oversees information collection, filtering, classification, formatting by generative AI, and processing by the emotion engine.
[0953] 2. Terminal: A device such as a smartphone or tablet that allows users to input, receive, and view information.
[0954] Software used
[0955] 1. Crawling engine: Collects data related to the search keywords from sources.
[0956] 2. Filtering system: Evaluate the reliability of collected data and eliminate false information.
[0957] 3. Classification system: Organize the filtered data chronologically and by category.
[0958] 4. Generative AI: Layout information into a portal site format for easy viewing by users.
[0959] 5. Emotion engine: Analyzes the user's search behavior and history, recognizes the user's emotional state, and adjusts the information that is displayed as a priority.
[0960] Data processing and calculation
[0961] 1. Receiving search keywords:
[0962] The user enters search keywords into the terminal and sends them to the server.
[0963] 2. Collection of Information:
[0964] The crawling engine on the server starts up and collects relevant information from news sites, official websites, social media, blogs, and other sources on the Internet.
[0965] 3. Information filtering:
[0966] The filtering system evaluates the reliability of collected information and filters out false or inaccurate information.
[0967] 4. Classifying and organizing information:
[0968] The filtered information is sorted chronologically and into categories such as "latest news" and "word of mouth."
[0969] 5. Emotion recognition:
[0970] The emotion engine analyzes users' search history and behavior to recognize their emotional state and prioritize information based on the recognized emotion.
[0971] 6. Formatting Information:
[0972] The organized information is laid out in a portal site format by the generation AI and converted into a format suitable for the user's device.
[0973] 7. Transmission and Display of Information:
[0974] The converted data is sent to the user's device and displayed in a portal site format, allowing the user to easily browse the classified information and access the information they need.
[0975] Specific examples
[0976] When a user enters the keyword "COVID-19 vaccine information," the server receives this keyword and the crawling engine collects related information. The filtering system eliminates unreliable information and classifies it into categories such as "latest news," "side effect information," and "research data," and the generative AI layouts the information appropriately. The emotion engine analyzes the user's emotions and prioritizes displaying information that provides reassurance if the user is feeling anxious.
[0977] Example prompt for generative AI model:
[0978] Provide reliable, up-to-date information on the following keyword: "COVID-19 vaccine information." Categorize information into news, reviews, and statistics, and prioritize information that reassures users based on emotion recognition.
[0979] Based on this example, it becomes easier for users to efficiently obtain reliable information and access data categories that have already been specified.
[0980] The flow of the specific processing in the application example 2 will be described with reference to FIG.
[0981] Step 1:
[0982] Input: The user enters a search keyword into the device's search box. Example: "COVID-19 vaccine information."
[0983] How it works: The user presses the search button and a search request is sent to the server.
[0984] Output: The server receives the search keywords and starts the search process.
[0985] Step 2:
[0986] Input: The server is ready to process the search keywords received from the user.
[0987] How it works: The server launches a crawling engine to collect information related to the search keywords from numerous online sources (news sites, official websites, social media, blogs, etc.).
[0988] Output: The server generates an initial list of the relevant information it has collected.
[0989] Step 3:
[0990] Input: The server proceeds based on the initial list of relevant information it has collected.
[0991] How it works: The filtering system is activated and evaluates the reliability of the collected information. Unreliable information and hoaxes are filtered out. For example, anonymous posts on social media and data from low-rated sources are filtered out.
[0992] Output: A filtered list of trusted information is generated.
[0993] Step 4:
[0994] Input: The server receives the filtered list of information.
[0995] How it works: Organizes information chronologically and by category. It categorizes information into breaking news, side effects, research data, etc., and sorts them chronologically within each category.
[0996] Output: A complete list of information organized chronologically and by category.
[0997] Step 5:
[0998] Input: The server performs the following process based on the organized information list.
[0999] How it works: The generative AI starts up and lays out the organized information in a portal site format. To make it easier for users to understand visually, the latest news is placed at the top of the page, side effect information in the middle, and research data at the bottom.
[1000] Output: Information in the form of a portal site is generated, laid out by the generation AI.
[1001] Step 6:
[1002] Input: The server receives the information laid out by the generation AI.
[1003] How it works: The emotion engine analyzes the user's search keywords and behavioral history to recognize their emotions. It then changes the priority of the information displayed based on their emotional state. For example, if the user is worried, it displays reassuring information at the top.
[1004] Output: A prioritized list of information tailored by the emotion engine is created.
[1005] Step 7:
[1006] Input: The server receives the priority list adjusted by the emotion engine.
[1007] What it does: Converts information into the appropriate data format (e.g. HTML, JSON, etc.) to send to the user's device.
[1008] Output: Data is formatted for sending to the terminal.
[1009] Step 8:
[1010] Input: The user's terminal receives the data sent from the server.
[1011] Operation: The device displays the received data in a portal site format. The user browses through the categorized information and accesses the information they need. In addition, the emotion engine prioritizes display, providing information appropriate to the user's psychological state.
[1012] Output: The portal site that the user can view is displayed.
[1013] 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.
[1014] 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.
[1015] 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.
[1016] [Third embodiment]
[1017] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.
[1018] 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.
[1019] 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).
[1020] 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.
[1021] 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.
[1022] 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).
[1023] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.
[1024] 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.
[1025] 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.
[1026] 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.
[1027] 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.
[1028] 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."
[1029] This invention is a system that collects necessary information based on search keywords entered by a user, eliminates unreliable information, and provides the information in a format that is easy for the user to view. Specific embodiments of this system are described below.
[1030] Server Processing
[1031] 1. Receiving search keywords from users
[1032] When a user enters search keywords on their device and presses the search button, the request is sent to the server. The server receives this request and starts the search process. For example, if a user enters "COVID-19 vaccine information," the server analyzes the keywords and prepares to collect related information.
[1033] 2. Starting the crawling engine
[1034] The server starts a crawling engine to collect information related to the search keyword from multiple sources on the Internet (news sites, official websites, social media, blogs, etc.). The engine searches sources as needed to obtain the latest data.
[1035] 3. Information filtering
[1036] The collected information is processed through a filtering system within the server. The filtering system checks the reliability of the information, the reputation of the source, and the consistency of the content, and filters out rumors and misinformation. For example, anonymous posts on social media and information from unreliable sources are filtered out.
[1037] 4. Classification and chronological organization of information
[1038] The filtered information is organized chronologically and by category: specifically, it is sorted based on the date and time of collection and categorized based on the content of the information (e.g., latest news, side effects, research data).
[1039] 5. Information Shaping with Generative AI
[1040] The organized information is then formatted into a portal site by a generative AI. This generative AI understands the meaning of search keywords and determines the layout so that users can easily browse related information. For example, the latest news is placed at the top of the page, and information on side effects and research data is displayed as the main content.
[1041] 6. Transmission of Information
[1042] The formatted information is sent to the user's device using web technologies such as HTML and JSON in a format suitable for the user's device.
[1043] Terminal handling
[1044] 1. Enter search keywords
[1045] Users enter keywords related to the information they want to find into the device's search box. For example, they enter "COVID-19 vaccine information."
[1046] 2. Receiving and Displaying Information
[1047] The terminal receives the formatted information sent from the server. This information is displayed in a portal site format, allowing the user to easily browse the categorized information.
[1048] User Activity
[1049] 1. Enter keywords
[1050] The user inputs the information they want to know into the device. For example, they input specific keywords such as "COVID-19 vaccine information."
[1051] 2. Verification and Use of Information
[1052] They can browse the displayed portal-style information and access the information they need, for example, by reading the latest news about vaccines, checking side-effect data and scientific research results.
[1053] Specific examples
[1054] (Search keyword: "COVID-19 vaccine information")
[1055] 1. Receiving a request
[1056] The server receives a search request for "COVID-19 vaccine information."
[1057] 2. Starting the crawling engine
[1058] The crawling engine starts up and collects information about vaccines from news sites, official websites, social media, etc.
[1059] 3. Information filtering
[1060] The filtering system evaluates the reliability of collected information and filters out false information, for example, giving priority to information from reliable news sites and official organizations.
[1061] 4. Classifying and Formatting Information
[1062] The filtered information is categorized into the latest news, side effect data, research results, etc. and organized in chronological order. Generative AI then formats this information into a portal site format and prepares it for delivery to users.
[1063] 5. Provision of Information
[1064] The formatted information is sent to the user's terminal, where the user can view the information displayed in a portal site format and access the information they need.
[1065] As described above, this invention is designed to enable users to efficiently obtain the information they need. By eliminating false information and categorizing information by time series and category, reliability and ease of use are improved. This system allows users to quickly access appropriate information.
[1066] The processing flow will be explained below.
[1067] Server Processing
[1068] Step 1:
[1069] The server receives the search keywords sent by the user, along with the user's settings information (language settings, region settings, etc.).
[1070] Step 2:
[1071] The server starts a crawling engine to collect information, and the crawling engine prepares to collect information related to the specified keyword from a wide range of sources.
[1072] Step 3:
[1073] The crawling engine accesses predefined sources such as news sites, official sites, social media, blogs, etc. to gather information related to the search keyword, exploring multiple sources in parallel during this process.
[1074] Step 4:
[1075] The collected information passes through a filtering system within the server, which evaluates the reliability of the information and filters out hoaxes and unreliable information. For example, unreliable anonymous social media posts are filtered out.
[1076] Step 5:
[1077] The filtered information is sorted chronologically and by category. Specifically, it is organized in chronological order based on the date and time the information was collected, and then divided into categories such as "latest news," "side effects," and "research data" according to its content.
[1078] Step 6:
[1079] The AI then organizes the information and lays it out in a portal site format. The AI understands the meaning of the search keywords and places relevant information in key positions. For example, the latest news is placed at the top of the page, and detailed research results are displayed in the center.
[1080] Step 7:
[1081] The formatted information is then converted into a format appropriate for the user's device (such as HTML or JSON), and the converted data is ready to be sent to the user's device.
[1082] Terminal handling
[1083] Step 1:
[1084] The user enters keywords into the search box on the device and presses the search button. This input information is sent to the server.
[1085] Step 2:
[1086] The terminal receives the formatted information sent from the server, parses the received data, and prepares it for display.
[1087] Step 3:
[1088] The device displays the received information in a portal site format, with the latest news displayed at the top of the screen, side effect statistics in the middle, and links to research papers at the bottom.
[1089] User Action
[1090] Step 1:
[1091] Users enter the keywords they want to know in the search box on their device and press the search button. For example, they could enter "COVID-19 vaccine information."
[1092] Step 2:
[1093] Users can check the information displayed on their devices in the form of a portal site and access the information they need, for example, by clicking on the latest news headline to read a detailed article.
[1094] The above is a specific processing flow of the present invention. This system allows users to obtain the information they need efficiently and quickly.
[1095] Example 1
[1096] 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."
[1097] There is a lot of information scattered on the Internet, making it difficult for users to quickly and accurately obtain the information they need. Furthermore, the Internet sometimes contains inaccurate or false information, making it difficult for users to access reliable information. There is a particular need for a system that can efficiently provide information for urgent situations or when specialized information is required.
[1098] 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.
[1099] In this invention, the server includes means for receiving search keywords from users, a crawling engine for collecting information related to the search keywords from online information sources, means for filtering the collected information and eliminating false information, means for classifying the filtered information in chronological order and by category, means for formatting the classified information into a portal site format using a generative AI model, and means for transmitting the information to the user's terminal, thereby enabling users to quickly and efficiently obtain reliable information.
[1100] The "means for receiving search keywords from a user" refers to a mechanism for transmitting search keywords entered by a user via an Internet terminal to a server and receiving the keywords.
[1101] The "crawling engine means" is a program that automatically collects information related to a specified keyword from multiple information sources on the Internet.
[1102] A "means for filtering out misinformation" is an algorithm or system that detects and filters out unreliable or false information from collected information.
[1103] "Chronological and categorical sorting" refers to the process of sorting the filtered information in chronological order based on the date and time of collection and categorizing it according to its content.
[1104] "Means for formatting into a portal site format using a generative AI model" refers to a technology that uses an artificial intelligence model to reconstruct classified information into a portal site format that is easy for users to view.
[1105] "Means for sending information to the user's device" refers to a mechanism for sending formatted information to the user's device using an appropriate communication means (e.g., HTML, JSON).
[1106] This invention is a system that collects necessary information based on search keywords entered by a user, eliminates unreliable information, and provides the information in a format that is easy for the user to view. Specific embodiments of this system are described below.
[1107] Server Processing
[1108] Receiving search keywords from users
[1109] When a user enters specific keywords, such as "COVID-19 vaccine information," into their device and presses the search button, the request is sent to the server. The server receives the HTTP request and extracts the search keywords from it. This prepares the system to begin collecting information.
[1110] Starting the crawling engine
[1111] The server runs a crawling engine using Python tools such as BeautifulSoup and Scrapy to collect information from multiple sources on the Internet. Based on the input keywords, the crawling engine searches news sites, official websites, social media, blogs, etc. to obtain the latest relevant data.
[1112] Information Filtering
[1113] The collected information is stored in a database, and then processed by a filtering system on the server. The filtering system evaluates the reliability of the source and its content, and runs an algorithm to filter out misinformation and unreliable information. Based on the source's reliability score, information with a score below a certain level is filtered out.
[1114] Classification and chronological organization of information
[1115] The filtered information is sorted chronologically and by category (e.g., latest news, side effects, research data). The server retrieves this information from the database and sorts it chronologically using an SQL query.
[1116] Shaping information with generative AI
[1117] The classified information is then formatted into a portal site using a generative AI model (e.g., GPT-4). The generative AI understands the context of the search keywords and generates a user-friendly layout. For example, the latest news is placed at the top of the page, and side effects and research data are placed in different sections.
[1118] Sending information
[1119] The formatted information is sent to the user's device in HTML or JSON format. The server generates an HTTP response and sends the information to the user's browser. The browser receives the response and displays the information.
[1120] Terminal handling
[1121] Enter search keywords
[1122] The user enters the information of interest (e.g., "COVID-19 vaccine information") into the search box on the device and presses the search button. The device then sends the user's input to the server.
[1123] Receiving and displaying information
[1124] The user's device receives the formatted information sent from the server and displays it on the screen. Information sent in HTML format is rendered by the browser and displayed so that the user can easily check the content. In the case of JSON format, JavaScript manipulates the DOM to dynamically display the information.
[1125] User Activity
[1126] Enter keywords
[1127] Users enter keywords related to the information they want to know into the device. For example, they can enter "COVID-19 vaccine information."
[1128] Verifying and Using Information
[1129] Users can review the displayed information and access the information they need, such as reading the latest vaccine news or viewing data and research results on side effects.
[1130] Specific examples
[1131] Search keyword: "COVID-19 vaccine information"
[1132] 1. Request reception: The server receives a search request for "COVID-19 vaccine information" and analyzes the keywords.
[1133] 2. Launching the crawling engine: Collect information about vaccines from news sites, official websites, social media, etc.
[1134] 3. Information filtering: Select reliable information and eliminate misinformation.
[1135] 4. Classification and formatting of information: The information is classified into the latest news, side effect data, research results, etc. and organized in chronological order. Generative AI formats this information into a portal site format.
[1136] 5. Providing information: The formatted information is sent to the user's device and displayed in portal site format.
[1137] As described above, the present invention is a system that efficiently collects information required by users and provides it in a highly reliable manner.
[1138] The flow of the identification process in the first embodiment will be described with reference to FIG.
[1139] Step 1:
[1140] Receiving search keywords from users
[1141] A user enters "COVID-19 vaccine information" into the search box on their device and presses the search button. The device generates an HTTP request and sends it to the server. The server receives this request and extracts the search keywords. It then prepares to collect information based on the analyzed keywords (Input: User's search keyword "COVID-19 vaccine information", Output: Analyzed search keywords).
[1142] Step 2:
[1143] Starting the crawling engine
[1144] The server starts a Python crawling engine (e.g., BeautifulSoup or Scrapy) using the parsed search keywords as input. This engine searches multiple sources (news sites, official website, social media, blogs, etc.) for information related to the specified keywords and obtains HTML data. The crawling engine then crawls these sites and returns the collected HTML content to the server (input: parsed search keywords, output: collected HTML data).
[1145] Step 3:
[1146] Information Filtering
[1147] The filtering system in the server receives the HTML data collected in the previous step as input. The filtering system runs an algorithm to evaluate the reliability of the information and eliminate hoaxes and unreliable information. For example, it analyzes the source domain and calculates a reliability score. Information from unreliable sites and anonymous social media posts will be eliminated due to their low scores. (Input: Collected HTML data, Output: Filtered information)
[1148] Step 4:
[1149] Classification and chronological organization of information
[1150] The server then sorts the filtered information into chronological order and categories. This process uses SQL queries to retrieve information stored in the database, sorts it based on the collection date and time, and categorizes it by content. Specifically, the information is classified into latest news, side effects, and research data, and then organized in chronological order within each category (input: filtered information, output: classified and organized information).
[1151] Step 5:
[1152] Shaping information with generative AI
[1153] The server takes the classified and organized information as input and uses a generative AI model (e.g., GPT-4) to format it into a portal site. The generative AI model understands the context of the search keywords and lays it out in a format that is easy for users to read. For example, the latest news is placed at the top of the page, and side effects and research data are placed in dedicated sections (input: classified and organized information, output: formatted HTML content).
[1154] Step 6:
[1155] Sending information
[1156] The formatted HTML content is sent from the server to the user's device. The server generates an HTTP response and delivers the information to the user's device in HTML or JSON format. The user's browser receives this and displays it optimally (input: formatted HTML content, output: information sent to the user's device).
[1157] Step 7:
[1158] Receiving and displaying information
[1159] The user's device receives the HTML or JSON data sent from the server and renders it in the browser. In the case of HTML format, the browser's rendering engine displays the page, allowing the user to easily browse the categorized information. In the case of JSON format, JavaScript manipulates the DOM to dynamically display the information (input: information sent from the server, output: information displayed on the device).
[1160] Step 8:
[1161] User review and use of information
[1162] The user checks the displayed information and accesses the information they need, such as reading the latest news about vaccines, checking side effect data, and scientific research results (input: information displayed on the device, output: information obtained by the user).
[1163] Through these steps, the system is able to efficiently collect and organize reliable information and provide it to users.
[1164] (Application example 1)
[1165] 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."
[1166] Conventional information retrieval systems have issues such as low reliability of information collected based on search keywords and lack of an appropriate display format for users to easily view and use information. Furthermore, especially when providing information for smart glasses and other wearable devices, the lack of a voice-based interface limits the user experience. To address these issues, a system is needed that can provide real-time, reliable information and display that information in an optimized audio and visual format.
[1167] 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.
[1168] In this invention, the server includes means for receiving search keywords from users, crawling engine means for collecting information related to the search keywords from online sources, means for filtering the collected information and eliminating false information, means for classifying the filtered information in chronological order and by type, generation AI means for displaying the generated information in a format suitable for a display device, means for analyzing voice input and generating search keywords, a speech synthesizer for outputting the classified information by voice, and means for providing the information to the user's display device. This allows users to obtain reliable information in real time and easily view and use it in a format optimized for audio and visuals.
[1169] A "user" is an entity that searches for and uses information.
[1170] "Search keywords" are words or phrases related to the information a user wants to find.
[1171] "Online sources" are information sources such as websites and databases that can be accessed via the Internet.
[1172] A "crawling engine" is software that automatically collects specific information from online sources.
[1173] "Filtering" is the process of removing unreliable information or false information from collected information.
[1174] "Chronological order" refers to arranging information in order based on the date and time it was collected.
[1175] "By type" refers to classifying information into categories according to its content.
[1176] "Generative AI" is artificial intelligence that formats collected and filtered information into a format suitable for the user's device or display device.
[1177] "Voice input" is a method in which a user provides information to a system by speaking into a microphone.
[1178] A "speech synthesizer" is a device that converts text information into speech and makes it audible to the user.
[1179] A "display device" is a device for visually presenting collected and generated information to a user.
[1180] This invention is a system that collects necessary information based on search keywords entered by the user through input or voice, eliminates unreliable information, and uses a generation AI to provide the information in a format suitable for the user's device. Specific embodiments of this system are described below.
[1181] Server Processing
[1182] 1. Receiving search keywords from users
[1183] The user enters a search keyword by voice or text, and the request is sent to the server. An example search keyword is "new spring clothes." The server receives this request and starts the search process.
[1184] 2. Starting the crawling engine
[1185] The server uses a crawling engine to collect information related to the search keywords from online sources, such as fashion websites, official brand websites, and social media.
[1186] 3. Information filtering
[1187] The information collected by the crawling engine is processed through a filtering system within the server. The filtering system evaluates the reliability of the information and removes unnecessary data such as false information. For example, anonymous social media posts and data from less reliable sources are excluded.
[1188] 4. Classification and chronological organization of information
[1189] The server then organizes the filtered information chronologically and by type, specifically into categories like tops, bottoms, and outerwear.
[1190] 5. Information Shaping with Generative AI
[1191] The organized information is then formatted into a suitable format for the display device by the generation AI, which understands the meaning of the search keywords and determines the layout so that the user can easily view related information.
[1192] 6. Transmission of Information
[1193] The formatted information is sent from the server to the user's display device using web technologies such as HTML and JSON.
[1194] Terminal handling
[1195] 1. Enter search keywords
[1196] The user inputs keywords related to the information they want to search for into a search box on the device or provides them by voice. For example, the user inputs or provides by voice a specific keyword such as "new spring clothes."
[1197] 2. Receiving and Displaying Information
[1198] The terminal receives the formatted information sent from the server, displays the information in a format optimal for the display device, and uses a speech synthesizer to provide the information to the user aloud.
[1199] User Activity
[1200] 1. Enter search keywords
[1201] Users input the information they want to know into the terminal. Voice input is also supported, so for example, keywords such as "new spring clothes" can be provided by voice.
[1202] 2. Verification and Use of Information
[1203] Users can access the information they need by viewing the information displayed in a portal format on the display device. Audio output is also provided, allowing users to view and hear the information.
[1204] Specific examples
[1205] For example, if a user uses smart glasses to say, "I want new spring clothes," the system will analyze the speech, collect relevant fashion information, filter and organize it, and use the generative AI to visually display the information in a format that is optimal for the smart glasses, while the speech synthesizer will provide the information aloud.
[1206] Prompt Sentence Examples
[1207] What's the latest information on spring fashion?
[1208] The hardware used includes smart glasses (e.g., microphone, speaker, display), and the software includes Python, Requests, BeautifulSoup, OpenAI GPT-3, and google_speech, allowing users to efficiently receive up-to-date, reliable information through visual and audio.
[1209] The flow of the specific processing in the application example 1 will be described with reference to FIG.
[1210] Step 1:
[1211] Enter and submit search keywords
[1212] The user enters search keywords using the voice input function of the smart glasses.
[1213] For example: "I want some new spring clothes."
[1214] Speech recognition software converts this voice input into text data.
[1215] The converted text data is sent to the server.
[1216] Step 2:
[1217] Receiving search keywords
[1218] The server receives the text data sent by the user.
[1219] Input: A keyword spoken by the user (e.g., "new spring clothes")
[1220] Output: Search keyword data for analysis
[1221] Step 3:
[1222] Starting the crawling engine
[1223] The server launches a crawling engine targeting multiple online information sources (fashion sites, official brand websites, social media, etc.).
[1224] Input: Search keyword ("new spring clothes")
[1225] Data processing: Obtain relevant information from sources and collect it as text data.
[1226] Output: Collected text data
[1227] Step 4:
[1228] Information Filtering
[1229] The server evaluates the reliability of the collected information using a filtering system.
[1230] Input: Collected text data
[1231] Data calculation: Remove unreliable information and false information.
[1232] What it does: Identify and filter out low-trust sources (e.g., anonymous social media posts).
[1233] Output: Filtered reliable information data
[1234] Step 5:
[1235] Classification and chronological organization of information
[1236] The server categorizes the filtered information chronologically and by type (tops, bottoms, outerwear, etc.).
[1237] Input: Filtered reliable information data
[1238] Data processing: Sort the information based on the date and time it was collected and divide it into categories.
[1239] Output: Categorized and organized information data
[1240] Step 6:
[1241] Shaping information with generative AI
[1242] The server uses generative AI to format the organized information into a format suitable for display on smart glasses.
[1243] Input: Classified and organized information data
[1244] Data calculation: Generative AI determines the optimal layout for the user and converts it into a format that is easy to display visually.
[1245] What it does: Sorts information by category, with the most recent information at the top.
[1246] Output: Formatted information data
[1247] Step 7:
[1248] Sending information
[1249] The server sends the formatted information to the user's smart glasses.
[1250] Input: Formatted information data
[1251] Data processing: Converts data into a transmission format (e.g., HTML, JSON) and executes transmission logic.
[1252] Output: Information data sent to the user terminal
[1253] Step 8:
[1254] Display and speech output of information
[1255] The terminal provides the received information to the user visually and audibly.
[1256] Input: Information data sent from the server
[1257] Specific operation: Information is displayed on the HUD of the smart glasses, and at the same time, the voice synthesizer plays the information aloud.
[1258] Output: Visual and audio information provided to the user
[1259] 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.
[1260] The present invention is a system that collects information on the Internet based on search keywords entered by users, eliminates unreliable information, and recognizes the user's emotions using an emotion engine to provide the user with appropriate information in the form of a portal site. A specific embodiment of this system is described below.
[1261] Server Processing
[1262] 1. Receiving search keywords from users
[1263] When a user enters search keywords on their device and presses the search button, the request is sent to the server. The server receives this request and starts the search process. For example, if a user enters "COVID-19 vaccine information," the server analyzes the keywords and prepares to collect related information.
[1264] 2. Starting the crawling engine
[1265] The server starts a crawling engine to collect information related to the search keyword from multiple sources on the Internet (news sites, official websites, social media, blogs, etc.). The engine searches sources as needed to obtain the latest data.
[1266] 3. Information filtering
[1267] The collected information is processed through a filtering system within the server. The filtering system checks the reliability of the information, the reputation of the source, and the consistency of the content, and filters out rumors and misinformation. For example, anonymous posts on social media and information from unreliable sources are filtered out.
[1268] 4. Classification and chronological organization of information
[1269] The filtered information is organized chronologically and by category. Specifically, it is organized in chronological order based on the date and time of collection, and then divided into categories such as "latest news," "side effects," and "research data" according to its content.
[1270] 5. Information Shaping with Generative AI
[1271] The organized information is then laid out in a portal site format by a generative AI. This generative AI understands the meaning of search keywords and arranges relevant information so that users can easily browse it. For example, the latest news is placed at the top of the page, and information on side effects and research data are displayed as the main content.
[1272] 6. User Emotion Recognition by Emotion Engine
[1273] The server uses an emotion engine to analyze the user's search keywords and behavioral history to recognize the user's emotions. Based on this emotion recognition, the priority of the information displayed is changed. For example, if the user is worried, information that gives a sense of security is displayed at the top.
[1274] 7. Transmission of Information
[1275] The formatted information is then converted into a format appropriate for the user's device (such as HTML or JSON), and the converted data is ready to be sent to the user's device.
[1276] Terminal handling
[1277] 1. Enter search keywords
[1278] Users enter keywords related to the information they want to find into the device's search box. For example, they enter "COVID-19 vaccine information."
[1279] 2. Receiving and Displaying Information
[1280] The device receives the formatted information sent from the server. This information is displayed in a portal site format, allowing users to easily browse categorized information. For example, the top of the screen displays the latest news, the middle section displays side effect statistics, and the bottom section displays links to research papers. In addition, information appropriate to the user's current emotional state is displayed based on the analysis results of the emotion engine.
[1281] User Activity
[1282] 1. Enter keywords
[1283] The user inputs the information they want to know into the device. For example, they input specific keywords such as "COVID-19 vaccine information."
[1284] 2. Verification and Use of Information
[1285] Users can review the displayed portal-style information and access the information they need. For example, they can read the latest news about vaccines, check side effect data, and check scientific research results. Based on the analysis results of the emotion engine, information appropriate to the user's emotional state is recommended, providing information to reduce stress.
[1286] Specific examples
[1287] (Search keyword: "COVID-19 vaccine information")
[1288] 1. Receiving a request
[1289] The server receives a search request for "COVID-19 vaccine information."
[1290] 2. Starting the crawling engine
[1291] The crawling engine starts up and collects information about vaccines from news sites, official websites, social media, etc.
[1292] 3. Information filtering
[1293] The filtering system evaluates the reliability of collected information and filters out false information, for example, giving priority to information from reliable news sites and official organizations.
[1294] 4. Classifying and Formatting Information
[1295] The filtered information is categorized into the latest news, side effect data, research results, etc. and organized in chronological order. Generative AI then formats this information into a portal site format and prepares it for delivery to users.
[1296] 5. Emotion Recognition by Emotion Engine
[1297] If a user is worried, information that provides reassurance (e.g., vaccine effectiveness or reassuring statistics) will be prioritized based on that emotion.
[1298] 6. Transmission of Information
[1299] The formatted information is sent to the user's terminal, where the user can view the information displayed in a portal site format and access the information they need.
[1300] As described above, this invention is designed to enable users to efficiently obtain the information they need. In addition to eliminating false information and classifying information by time series and category, reliability, ease of use, and psychological compatibility are improved by recognizing user emotions using an emotion engine. This system allows users to quickly access appropriate information.
[1301] The processing flow will be explained below.
[1302] Server Processing
[1303] Step 1:
[1304] The server receives the search keywords sent by the user, along with the user's settings information (language settings, region settings, etc.).
[1305] Step 2:
[1306] The server starts a crawling engine to collect information, and the crawling engine prepares to collect information related to the specified keyword from a wide range of sources.
[1307] Step 3:
[1308] The crawling engine accesses predefined sources such as news sites, official sites, social media, blogs, etc. to gather information related to the search keyword, exploring multiple sources in parallel during this process.
[1309] Step 4:
[1310] The collected information passes through a filtering system within the server, which evaluates the reliability of the information and filters out hoaxes and unreliable information. For example, unreliable anonymous social media posts are filtered out.
[1311] Step 5:
[1312] The filtered information is sorted chronologically and by category. Specifically, it is organized in chronological order based on the date and time the information was collected, and then divided into categories such as "latest news," "side effects," and "research data" according to its content.
[1313] Step 6:
[1314] The AI then organizes the information and lays it out in a portal site format. The AI understands the meaning of the search keywords and places relevant information in key positions. For example, the latest news is placed at the top of the page, and detailed research results are displayed in the center.
[1315] Step 7:
[1316] The server starts the emotion engine, which analyzes the user's search keywords and history to recognize the user's emotions. Based on the recognition results, the display priority is changed.
[1317] Step 8:
[1318] The formatted information is then converted into a format appropriate for the user's device (such as HTML or JSON), and the converted data is ready to be sent to the user's device.
[1319] Terminal handling
[1320] Step 1:
[1321] The user enters keywords into the search box on the device and presses the search button. This input information is sent to the server.
[1322] Step 2:
[1323] The terminal receives the formatted information sent from the server, parses the received data, and prepares it for display.
[1324] Step 3:
[1325] The device displays the received information in a portal site format. For example, the top of the screen displays the latest news, side effect statistics in the middle, and links to research papers in the bottom. Based on the analysis results of the emotion engine, the device also displays information appropriate to the user's current psychological state.
[1326] User Action
[1327] Step 1:
[1328] Users enter the keywords they want to know in the search box on their device and press the search button. For example, they could enter "COVID-19 vaccine information."
[1329] Step 2:
[1330] Users can check the portal site-style information displayed on their device and access the information they need. For example, they can click on the latest news headline to read a detailed article. They can also view recommended reassuring information based on the analysis results of the emotion engine.
[1331] Specific examples
[1332] (Search keyword: "COVID-19 vaccine information")
[1333] Server details:
[1334] Step 1:
[1335] The server receives a search request for "COVID-19 vaccine information."
[1336] Step 2:
[1337] The crawling engine starts up and collects information about vaccines from news sites, official websites, social media, etc.
[1338] Step 3:
[1339] The collected information is filtered out using a filtering system to remove false information, such as anonymous posts on social media and unreliable information.
[1340] Step 4:
[1341] The filtered information is categorized into the latest news, side effect data, research results, etc. and organized in chronological order.
[1342] Step 5:
[1343] The generating AI lays out the classified information in a portal site format, placing important information at the top.
[1344] Step 6:
[1345] The server activates an emotion engine, analyzes the user's search history and behavior, and prioritizes displaying information that will reassure worried users.
[1346] Step 7:
[1347] The formatted information is converted into HTML or JSON format and sent to the user's device.
[1348] Detailed terminal processing:
[1349] Step 1:
[1350] The user enters "COVID-19 vaccine information" and presses the search button.
[1351] Step 2:
[1352] The terminal receives the information sent from the server and analyzes it.
[1353] Step 3:
[1354] The terminal displays the received information in a portal site format, and displays information according to the user's emotions with priority.
[1355] User detailed processing:
[1356] Step 1:
[1357] A user enters "COVID-19 vaccine information" into the search box and presses the search button.
[1358] Step 2:
[1359] Users can check the displayed information in a portal site format and click on the item they are interested in to see more details. For example, a worried user will be given priority in seeing reassuring information demonstrating the effectiveness of vaccines.
[1360] As described above, the present invention not only enables users to efficiently obtain the information they need, but also improves user satisfaction and reliability, since the information provided by the emotion engine is adapted to the user's emotional state.
[1361] Example 2
[1362] 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."
[1363] In today's Internet environment, a huge amount of information is generated every day, making it difficult for users to quickly obtain reliable information from that information. Furthermore, if the information is unreliable or contains false information, users may act based on the incorrect information. Furthermore, conventional information gathering systems lack a mechanism for changing the priority of information taking into account the user's emotions, which makes it difficult for users to quickly access the information they want to see.
[1364] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving a search keyword from a user, information collection means for collecting information related to the search keyword from online information sources, means for filtering the collected information and eliminating unreliable information, means for classifying the filtered information in chronological order and by category, generation AI means for displaying the generated information in a portal site format, emotion recognition means for recognizing the user's emotion and changing the priority of information based on the emotion, and means for providing information to the user's terminal. This allows the user to quickly access reliable information and provides appropriate information according to the user's emotional state, thereby improving the convenience and reliability of information.
[1365] The "means for receiving search keywords from a user" is a function that allows the server to receive search keywords that a user has input on a terminal.
[1366] "Information gathering means for gathering information related to search keywords from online sources" is a function for gathering information related to specific search keywords from multiple sources on the Internet.
[1367] "Means for filtering collected information and eliminating unreliable information" refers to a function for removing unreliable information or false information from the collected information.
[1368] "Means for classifying filtered information chronologically and by category" refers to a function for organizing filtered information in chronological order based on the date and time of collection and classifying it into different categories depending on its content.
[1369] "Generative AI means for displaying generated information in a portal site format" is a function that utilizes artificial intelligence technology to format organized and classified information into a portal site format and display it in a way that is easy for users to view.
[1370] "Emotion recognition means that recognizes the user's emotions and changes the priority of information based on those emotions" is a function that analyzes the user's search keywords and behavioral history to recognize the user's emotions, and dynamically changes the priority of the information to be displayed based on the results.
[1371] The "means for providing information to a user's terminal" is a function for transmitting formatted information to a user's terminal in an appropriate format.
[1372] This invention is a system that collects information on the Internet based on search keywords entered by users, eliminates unreliable information, recognizes the user's emotions using an emotion engine, and provides appropriate information in the form of a portal site. This system is composed of three entities: a server, a terminal, and a user.
[1373] Server Action:
[1374] First, the server receives the search keywords entered by the user on the device. For example, if a specific keyword such as "COVID-19 vaccine information" is entered, the server analyzes this information using a dedicated analysis module.
[1375] The server then launches a crawling engine to gather relevant information from multiple sources on the Internet (news sites, official websites, social media, blogs, etc.) using software such as Apache Nutch or Scrapy. The engine generates a list of URLs as needed and selects relevant information.
[1376] The collected information is then passed through a filtering system to check its reliability. This system analyzes the reputation of the information source and the consistency of its content, and eliminates unreliable information and hoaxes. For example, anonymous posts on social media and unreliable blogs are excluded.
[1377] After filtering, the information is further sorted chronologically and by category. Specifically, it is organized based on the date and time of collection and divided into categories such as "latest news," "side effects," and "research data" according to its content.
[1378] The information is then formatted using generative AI and laid out in the form of a portal site. Models such as GPT-3 and BERT are used for this generative AI. The generative AI understands the context of the search keywords and arranges the information so that it is easy for users to view. The layout is designed with usability in mind, such as placing the latest news at the top of the page and displaying information about side effects and research data as the main content.
[1379] Furthermore, the server uses an emotion engine to analyze the user's search keywords and behavioral history to recognize the user's emotions. The emotion engine uses natural language processing technology to determine whether the user is anxious. For example, if the user is anxious, it can prioritize the display of information that will give a sense of reassurance (e.g., the effectiveness of vaccines or reassuring statistical data).
[1380] Finally, the formatted information is converted into a format suitable for the user's device (such as HTML or JSON) and sent to the device using the HTTP protocol.
[1381] Terminal handling:
[1382] Users enter keywords for the information they want to find into the search box on their device. For example, they might enter "COVID-19 vaccine information." This information is sent to the server, where it is analyzed and processed.
[1383] The device receives and displays the formatted data sent from the server. This information is displayed in a portal site format, allowing users to easily browse categorized information. For example, the top of the screen displays the latest news, the middle section displays side effect statistics, and the bottom section displays links to research papers. Additionally, information appropriate to the user's current psychological state is displayed based on the analysis results of an emotion engine.
[1384] User Activity:
[1385] Users enter the information they want to know into their device and perform a search. They then check the portal-site-style information displayed on the device and access the information they need. For example, they can browse the latest news about vaccines and check side effect data and scientific research results. Based on the analysis results of the emotion engine, information that is appropriate to the user's emotional state is highlighted, allowing the user to use the information with peace of mind.
[1386] Examples of prompts:
[1387] "New Coronavirus Vaccine Information"
[1388] The flow of the identification process in the second embodiment will be described with reference to FIG.
[1389] Step 1: User enters keywords
[1390] Users enter keywords for the information they want to find into the device's search box, such as "COVID-19 vaccine information."
[1391] Input: The search term entered by the user.
[1392] Output: The data sent to the server as a search request.
[1393] Step 2: Receiving and analyzing search keywords
[1394] The server receives search keywords sent by the user, which are then input to an analysis module and analyzed as search keywords.
[1395] Input: The search keyword submitted by the user.
[1396] Output: Parsed search keywords data format.
[1397] Specifically, the server extracts keywords as string patterns using an analysis module and passes them to the next processing step.
[1398] Step 3: Start the crawling engine and gather information
[1399] The server runs a crawling engine to collect relevant information from online sources (news sites, official websites, social media, blogs, etc.) using crawling engines such as Apache Nutch and Scrapy.
[1400] Input: The parsed search keywords.
[1401] Output: Collected raw information data.
[1402] Specifically, the crawling engine generates a list of URLs to search and sends HTTP requests to collect relevant information.
[1403] Step 4: Filtering the information
[1404] The collected information is then input into a filtering system to check its reliability, which evaluates the source of the information and eliminates unreliable or false information.
[1405] Input: Collected information data.
[1406] Output: filtered reliable information data.
[1407] Specifically, the filtering system evaluates information sources using reputation scores and eliminates information from unreliable social media posts and unreliable blogs.
[1408] Step 5: Categorize and organize information
[1409] The filtered information is sorted chronologically and by category. Specifically, it is organized based on the date and time of collection and sorted into categories such as "latest news," "side effects," and "research data" according to its content.
[1410] Input: Filtered information data.
[1411] Output: Information data organized in categorized and chronological order.
[1412] Specifically, the server uses the information's metadata (collection date and time, information category, etc.) to store the information in a database and then categorizes and organizes it appropriately.
[1413] Step 6: Formatting information with generative AI
[1414] The organized information is then laid out in a portal site format by a generative AI, using models such as GPT-3 and BERT. The generative AI understands the context of search keywords and arranges the information in a way that makes it easy for users to browse.
[1415] Input: Information data organized in categorized and chronological order.
[1416] Output: Information data formatted for a portal site.
[1417] Specifically, the generative AI model analyzes the input data, places the latest news at the top of the page, and arranges information about side effects and research data as the main content.
[1418] Step 7: Recognizing user emotions with the emotion engine
[1419] The server uses an emotion engine to analyze the user's search keywords and behavioral history to recognize the user's emotions. Based on the emotion recognition, the priority of the information to be displayed is changed.
[1420] Input: User search keywords and behavioral history data.
[1421] Output: Emotion recognition results and reprioritized information data.
[1422] Specifically, the emotion engine uses natural language processing technology to analyze the user's emotional state (e.g., worry, relief), and optimizes the display order of information based on the results.
[1423] Step 8: Send and view information
[1424] The formatted information is converted into a format suitable for the user's device (HTML, JSON, etc.) and sent to the device. The device displays the received information in portal site format.
[1425] Input: Information data with changed priority.
[1426] Output: Data sent to the user's device.
[1427] Specifically, the server converts the information into an appropriate format and sends it to the user's device using the HTTP protocol. The device then renders the received data using a display engine, allowing the user to view the information in a portal site format.
[1428] (Application example 2)
[1429] 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."
[1430] In today's information society, a huge amount of information is circulating on the Internet, including unreliable and false information. It is difficult for users to quickly and accurately obtain the information they need. Furthermore, there are few systems that prioritize and provide appropriate information based on the user's search behavior and psychological state. This can cause users to feel stressed or anxious during the information acquisition process. The present invention aims to solve these problems and efficiently provide reliable information that corresponds to the user's emotional state.
[1431] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: means for receiving a search keyword from a user; crawling engine means for collecting information related to the search keyword from online information sources; means for filtering the collected information and eliminating false information; means for classifying the filtered information in chronological order and by category; generation AI means for displaying the generated information in a portal site format; means for providing the information to the user's terminal; emotion engine means for recognizing the user's emotions and changing the priority of the information to be displayed based on the emotions; means for classifying the filtered information by reliability, content, and category and rearranging it in chronological order; and means for prioritizing the generated portal site-formatted information based on the user's emotions. This allows users to efficiently obtain reliable information without stress and make decisions based on appropriate information.
[1432] "User" refers to an end user who uses this system to search for information.
[1433] A "search keyword" is a word or phrase that a user enters to identify information in which they are interested.
[1434] A "crawling engine" is software that automatically collects information related to specified search keywords from various sources on the Internet.
[1435] "Filtering" is the process of evaluating the reliability and relevance of collected information and eliminating or filtering out false or unnecessary information.
[1436] "Chronological order" is a method of ordering information based on the date and time it was collected, with the most recent information being displayed first.
[1437] "Categorization" is the process of separating collected information into different categories based on its content.
[1438] "Generative AI" is artificial intelligence that takes collected and organized information and formats it into a form that users can easily understand and use (e.g., portal site format).
[1439] An "emotion engine" is a system that analyzes a user's input and behavioral history to recognize the user's current emotions.
[1440] "Changing the priority" means appropriately adjusting the display order of information based on the emotional state of the user.
[1441] The "portal site format" is a format that integrates various information and allows users to access all information on a single interface.
[1442] This invention is a system that collects information on the Internet based on search keywords entered by a user, evaluates the reliability of that information, and provides appropriate information according to the user's emotional state. Specific embodiments of this invention are described below.
[1443] System Overview
[1444] The system includes a server that receives search keywords from users and collects related information from the internet, and a terminal that allows users to access the information. The information is classified through reliability evaluation and filtering, and then formatted into a portal site format by a generative AI. In addition, the emotional engine recognizes the user's emotional state and adjusts the priority of the information to be displayed.
[1445] Hardware used
[1446] 1. Server: Oversees information collection, filtering, classification, formatting by generative AI, and processing by the emotion engine.
[1447] 2. Terminal: A device such as a smartphone or tablet that allows users to input, receive, and view information.
[1448] Software used
[1449] 1. Crawling engine: Collects data related to the search keywords from sources.
[1450] 2. Filtering system: Evaluate the reliability of collected data and eliminate false information.
[1451] 3. Classification system: Organize the filtered data chronologically and by category.
[1452] 4. Generative AI: Layout information into a portal site format for easy viewing by users.
[1453] 5. Emotion engine: Analyzes the user's search behavior and history, recognizes the user's emotional state, and adjusts the information that is displayed as a priority.
[1454] Data processing and calculation
[1455] 1. Receiving search keywords:
[1456] The user enters search keywords into the terminal and sends them to the server.
[1457] 2. Collection of Information:
[1458] The crawling engine on the server starts up and collects relevant information from news sites, official websites, social media, blogs, and other sources on the Internet.
[1459] 3. Information filtering:
[1460] The filtering system evaluates the reliability of collected information and filters out false or inaccurate information.
[1461] 4. Classifying and organizing information:
[1462] The filtered information is sorted chronologically and into categories such as "latest news" and "word of mouth."
[1463] 5. Emotion recognition:
[1464] The emotion engine analyzes users' search history and behavior to recognize their emotional state and prioritize information based on the recognized emotion.
[1465] 6. Formatting Information:
[1466] The organized information is laid out in a portal site format by the generation AI and converted into a format suitable for the user's device.
[1467] 7. Transmission and Display of Information:
[1468] The converted data is sent to the user's device and displayed in a portal site format, allowing the user to easily browse the classified information and access the information they need.
[1469] Specific examples
[1470] When a user enters the keyword "COVID-19 vaccine information," the server receives this keyword and the crawling engine collects related information. The filtering system eliminates unreliable information and classifies it into categories such as "latest news," "side effect information," and "research data," and the generative AI layouts the information appropriately. The emotion engine analyzes the user's emotions and prioritizes displaying information that provides reassurance if the user is feeling anxious.
[1471] Example prompt for generative AI model:
[1472] Provide reliable, up-to-date information on the following keyword: "COVID-19 vaccine information." Categorize information into news, reviews, and statistics, and prioritize information that reassures users based on emotion recognition.
[1473] Based on this example, it becomes easier for users to efficiently obtain reliable information and access data categories that have already been specified.
[1474] The flow of the specific processing in the application example 2 will be described with reference to FIG.
[1475] Step 1:
[1476] Input: The user enters a search keyword into the device's search box. Example: "COVID-19 vaccine information."
[1477] How it works: The user presses the search button and a search request is sent to the server.
[1478] Output: The server receives the search keywords and starts the search process.
[1479] Step 2:
[1480] Input: The server is ready to process the search keywords received from the user.
[1481] How it works: The server launches a crawling engine to collect information related to the search keywords from numerous online sources (news sites, official websites, social media, blogs, etc.).
[1482] Output: The server generates an initial list of the relevant information it has collected.
[1483] Step 3:
[1484] Input: The server proceeds based on the initial list of relevant information it has collected.
[1485] How it works: The filtering system is activated and evaluates the reliability of the collected information. Unreliable information and hoaxes are filtered out. For example, anonymous posts on social media and data from low-rated sources are filtered out.
[1486] Output: A filtered list of trusted information is generated.
[1487] Step 4:
[1488] Input: The server receives the filtered list of information.
[1489] How it works: Organizes information chronologically and by category. It categorizes information into breaking news, side effects, research data, etc., and sorts them chronologically within each category.
[1490] Output: A complete list of information organized chronologically and by category.
[1491] Step 5:
[1492] Input: The server performs the following process based on the organized information list.
[1493] How it works: The generative AI starts up and lays out the organized information in a portal site format. To make it easier for users to understand visually, the latest news is placed at the top of the page, side effect information in the middle, and research data at the bottom.
[1494] Output: Information in the form of a portal site is generated, laid out by the generation AI.
[1495] Step 6:
[1496] Input: The server receives the information laid out by the generation AI.
[1497] How it works: The emotion engine analyzes the user's search keywords and behavioral history to recognize their emotions. It then changes the priority of the information displayed based on their emotional state. For example, if the user is worried, it displays reassuring information at the top.
[1498] Output: A prioritized list of information tailored by the emotion engine is created.
[1499] Step 7:
[1500] Input: The server receives the priority list adjusted by the emotion engine.
[1501] What it does: Converts information into the appropriate data format (e.g. HTML, JSON, etc.) to send to the user's device.
[1502] Output: Data is formatted for sending to the terminal.
[1503] Step 8:
[1504] Input: The user's terminal receives the data sent from the server.
[1505] Operation: The device displays the received data in a portal site format. The user browses through the categorized information and accesses the information they need. In addition, the emotion engine prioritizes display, providing information appropriate to the user's psychological state.
[1506] Output: The portal site that the user can view is displayed.
[1507] 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.
[1508] 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.
[1509] 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.
[1510] [Fourth embodiment]
[1511] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.
[1512] 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.
[1513] 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).
[1514] 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.
[1515] 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.
[1516] 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).
[1517] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.
[1518] 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.
[1519] 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.
[1520] 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.
[1521] 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.
[1522] 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.
[1523] 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."
[1524] This invention is a system that collects necessary information based on search keywords entered by a user, eliminates unreliable information, and provides the information in a format that is easy for the user to view. Specific embodiments of this system are described below.
[1525] Server Processing
[1526] 1. Receiving search keywords from users
[1527] When a user enters search keywords on their device and presses the search button, the request is sent to the server. The server receives this request and starts the search process. For example, if a user enters "COVID-19 vaccine information," the server analyzes the keywords and prepares to collect related information.
[1528] 2. Starting the crawling engine
[1529] The server starts a crawling engine to collect information related to the search keyword from multiple sources on the Internet (news sites, official websites, social media, blogs, etc.). The engine searches sources as needed to obtain the latest data.
[1530] 3. Information filtering
[1531] The collected information is processed through a filtering system within the server. The filtering system checks the reliability of the information, the reputation of the source, and the consistency of the content, and filters out rumors and misinformation. For example, anonymous posts on social media and information from unreliable sources are filtered out.
[1532] 4. Classification and chronological organization of information
[1533] The filtered information is organized chronologically and by category: specifically, it is sorted based on the date and time of collection and categorized based on the content of the information (e.g., latest news, side effects, research data).
[1534] 5. Information Shaping with Generative AI
[1535] The organized information is then formatted into a portal site by a generative AI. This generative AI understands the meaning of search keywords and determines the layout so that users can easily browse related information. For example, the latest news is placed at the top of the page, and information on side effects and research data is displayed as the main content.
[1536] 6. Transmission of Information
[1537] The formatted information is sent to the user's device using web technologies such as HTML and JSON in a format suitable for the user's device.
[1538] Terminal handling
[1539] 1. Enter search keywords
[1540] Users enter keywords related to the information they want to find into the device's search box. For example, they enter "COVID-19 vaccine information."
[1541] 2. Receiving and Displaying Information
[1542] The terminal receives the formatted information sent from the server. This information is displayed in a portal site format, allowing the user to easily browse the categorized information.
[1543] User Activity
[1544] 1. Enter keywords
[1545] The user inputs the information they want to know into the device. For example, they input specific keywords such as "COVID-19 vaccine information."
[1546] 2. Verification and Use of Information
[1547] They can browse the displayed portal-style information and access the information they need, for example, by reading the latest news about vaccines, checking side-effect data and scientific research results.
[1548] Specific examples
[1549] (Search keyword: "COVID-19 vaccine information")
[1550] 1. Receiving a request
[1551] The server receives a search request for "COVID-19 vaccine information."
[1552] 2. Starting the crawling engine
[1553] The crawling engine starts up and collects information about vaccines from news sites, official websites, social media, etc.
[1554] 3. Information filtering
[1555] The filtering system evaluates the reliability of collected information and filters out false information, for example, giving priority to information from reliable news sites and official organizations.
[1556] 4. Classifying and Formatting Information
[1557] The filtered information is categorized into the latest news, side effect data, research results, etc. and organized in chronological order. Generative AI then formats this information into a portal site format and prepares it for delivery to users.
[1558] 5. Provision of Information
[1559] The formatted information is sent to the user's terminal, where the user can view the information displayed in a portal site format and access the information they need.
[1560] As described above, this invention is designed to enable users to efficiently obtain the information they need. By eliminating false information and categorizing information by time series and category, reliability and ease of use are improved. This system allows users to quickly access appropriate information.
[1561] The processing flow will be explained below.
[1562] Server Processing
[1563] Step 1:
[1564] The server receives the search keywords sent by the user, along with the user's settings information (language settings, region settings, etc.).
[1565] Step 2:
[1566] The server starts a crawling engine to collect information, and the crawling engine prepares to collect information related to the specified keyword from a wide range of sources.
[1567] Step 3:
[1568] The crawling engine accesses predefined sources such as news sites, official sites, social media, blogs, etc. to gather information related to the search keyword, exploring multiple sources in parallel during this process.
[1569] Step 4:
[1570] The collected information passes through a filtering system within the server, which evaluates the reliability of the information and filters out hoaxes and unreliable information. For example, unreliable anonymous social media posts are filtered out.
[1571] Step 5:
[1572] The filtered information is sorted chronologically and by category. Specifically, it is organized in chronological order based on the date and time the information was collected, and then divided into categories such as "latest news," "side effects," and "research data" according to its content.
[1573] Step 6:
[1574] The AI then organizes the information and lays it out in a portal site format. The AI understands the meaning of the search keywords and places relevant information in key positions. For example, the latest news is placed at the top of the page, and detailed research results are displayed in the center.
[1575] Step 7:
[1576] The formatted information is then converted into a format appropriate for the user's device (such as HTML or JSON), and the converted data is ready to be sent to the user's device.
[1577] Terminal handling
[1578] Step 1:
[1579] The user enters keywords into the search box on the device and presses the search button. This input information is sent to the server.
[1580] Step 2:
[1581] The terminal receives the formatted information sent from the server, parses the received data, and prepares it for display.
[1582] Step 3:
[1583] The device displays the received information in a portal site format, with the latest news displayed at the top of the screen, side effect statistics in the middle, and links to research papers at the bottom.
[1584] User Action
[1585] Step 1:
[1586] Users enter the keywords they want to know in the search box on their device and press the search button. For example, they could enter "COVID-19 vaccine information."
[1587] Step 2:
[1588] Users can check the information displayed on their devices in the form of a portal site and access the information they need, for example, by clicking on the latest news headline to read a detailed article.
[1589] The above is a specific processing flow of the present invention. This system allows users to obtain the information they need efficiently and quickly.
[1590] Example 1
[1591] 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."
[1592] There is a lot of information scattered on the Internet, making it difficult for users to quickly and accurately obtain the information they need. Furthermore, the Internet sometimes contains inaccurate or false information, making it difficult for users to access reliable information. There is a particular need for a system that can efficiently provide information for urgent situations or when specialized information is required.
[1593] 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.
[1594] In this invention, the server includes means for receiving search keywords from users, a crawling engine for collecting information related to the search keywords from online information sources, means for filtering the collected information and eliminating false information, means for classifying the filtered information in chronological order and by category, means for formatting the classified information into a portal site format using a generative AI model, and means for transmitting the information to the user's terminal, thereby enabling users to quickly and efficiently obtain reliable information.
[1595] The "means for receiving search keywords from a user" refers to a mechanism for transmitting search keywords entered by a user via an Internet terminal to a server and receiving the keywords.
[1596] The "crawling engine means" is a program that automatically collects information related to a specified keyword from multiple information sources on the Internet.
[1597] A "means for filtering out misinformation" is an algorithm or system that detects and filters out unreliable or false information from collected information.
[1598] "Chronological and categorical sorting" refers to the process of sorting the filtered information in chronological order based on the date and time of collection and categorizing it according to its content.
[1599] "Means for formatting into a portal site format using a generative AI model" refers to a technology that uses an artificial intelligence model to reconstruct classified information into a portal site format that is easy for users to view.
[1600] "Means for sending information to the user's device" refers to a mechanism for sending formatted information to the user's device using an appropriate communication means (e.g., HTML, JSON).
[1601] This invention is a system that collects necessary information based on search keywords entered by a user, eliminates unreliable information, and provides the information in a format that is easy for the user to view. Specific embodiments of this system are described below.
[1602] Server Processing
[1603] Receiving search keywords from users
[1604] When a user enters specific keywords, such as "COVID-19 vaccine information," into their device and presses the search button, the request is sent to the server. The server receives the HTTP request and extracts the search keywords from it. This prepares the system to begin collecting information.
[1605] Starting the crawling engine
[1606] The server runs a crawling engine using Python tools such as BeautifulSoup and Scrapy to collect information from multiple sources on the Internet. Based on the input keywords, the crawling engine searches news sites, official websites, social media, blogs, etc. to obtain the latest relevant data.
[1607] Information Filtering
[1608] The collected information is stored in a database, and then processed by a filtering system on the server. The filtering system evaluates the reliability of the source and its content, and runs an algorithm to filter out misinformation and unreliable information. Based on the source's reliability score, information with a score below a certain level is filtered out.
[1609] Classification and chronological organization of information
[1610] The filtered information is sorted chronologically and by category (e.g., breaking news, side effects, research data). The server retrieves this information from the database and sorts it chronologically using an SQL query.
[1611] Shaping information with generative AI
[1612] The classified information is then formatted into a portal site using a generative AI model (e.g., GPT-4). The generative AI understands the context of the search keywords and generates a user-friendly layout. For example, the latest news is placed at the top of the page, and side effects and research data are placed in different sections.
[1613] Sending information
[1614] The formatted information is sent to the user's device in HTML or JSON format. The server generates an HTTP response and sends the information to the user's browser. The browser receives the response and displays the information.
[1615] Terminal handling
[1616] Enter search keywords
[1617] The user enters the information of interest (e.g., "COVID-19 vaccine information") into the search box on the device and presses the search button. The device then sends the user's input to the server.
[1618] Receiving and displaying information
[1619] The user's device receives the formatted information sent from the server and displays it on the screen. Information sent in HTML format is rendered by the browser and displayed so that the user can easily check the content. In the case of JSON format, JavaScript manipulates the DOM to dynamically display the information.
[1620] User Activity
[1621] Enter keywords
[1622] Users enter keywords related to the information they want to know into the device. For example, they can enter "COVID-19 vaccine information."
[1623] Verifying and Using Information
[1624] Users can review the displayed information and access the information they need, such as reading the latest vaccine news or viewing data and research results on side effects.
[1625] Specific examples
[1626] Search keyword: "COVID-19 vaccine information"
[1627] 1. Request reception: The server receives a search request for "COVID-19 vaccine information" and analyzes the keywords.
[1628] 2. Launching the crawling engine: Collect information about vaccines from news sites, official websites, social media, etc.
[1629] 3. Information filtering: Select reliable information and eliminate misinformation.
[1630] 4. Classification and formatting of information: The information is classified into the latest news, side effect data, research results, etc. and organized in chronological order. Generative AI formats this information into a portal site format.
[1631] 5. Providing information: The formatted information is sent to the user's device and displayed in portal site format.
[1632] As described above, the present invention is a system that efficiently collects information required by users and provides it in a highly reliable manner.
[1633] The flow of the identification process in the first embodiment will be described with reference to FIG.
[1634] Step 1:
[1635] Receiving search keywords from users
[1636] A user enters "COVID-19 vaccine information" into the search box on their device and presses the search button. The device generates an HTTP request and sends it to the server. The server receives this request and extracts the search keywords. It then prepares to collect information based on the analyzed keywords (Input: User's search keyword "COVID-19 vaccine information", Output: Analyzed search keywords).
[1637] Step 2:
[1638] Starting the crawling engine
[1639] The server starts a Python crawling engine (e.g., BeautifulSoup or Scrapy) using the parsed search keywords as input. This engine searches multiple sources (news sites, official website, social media, blogs, etc.) for information related to the specified keywords and obtains HTML data. The crawling engine then crawls these sites and returns the collected HTML content to the server (input: parsed search keywords, output: collected HTML data).
[1640] Step 3:
[1641] Information Filtering
[1642] The filtering system in the server receives the HTML data collected in the previous step as input. The filtering system runs an algorithm to evaluate the reliability of the information and eliminate hoaxes and unreliable information. For example, it analyzes the source domain and calculates a reliability score. Information from unreliable sites and anonymous social media posts will be eliminated due to their low scores. (Input: Collected HTML data, Output: Filtered information)
[1643] Step 4:
[1644] Classification and chronological organization of information
[1645] The server then sorts the filtered information into chronological order and categories. This process uses SQL queries to retrieve information stored in the database, sorts it based on the collection date and time, and categorizes it by content. Specifically, the information is classified into latest news, side effects, and research data, and then organized in chronological order within each category (input: filtered information, output: classified and organized information).
[1646] Step 5:
[1647] Shaping information with generative AI
[1648] The server takes the classified and organized information as input and uses a generative AI model (e.g., GPT-4) to format it into a portal site. The generative AI model understands the context of the search keywords and lays it out in a format that is easy for users to read. For example, the latest news is placed at the top of the page, and side effects and research data are placed in dedicated sections (input: classified and organized information, output: formatted HTML content).
[1649] Step 6:
[1650] Sending information
[1651] The formatted HTML content is sent from the server to the user's device. The server generates an HTTP response and delivers the information to the user's device in HTML or JSON format. The user's browser receives this and displays it optimally (input: formatted HTML content, output: information sent to the user's device).
[1652] Step 7:
[1653] Receiving and displaying information
[1654] The user's device receives the HTML or JSON data sent from the server and renders it in the browser. In the case of HTML format, the browser's rendering engine displays the page, allowing the user to easily browse the categorized information. In the case of JSON format, JavaScript manipulates the DOM to dynamically display the information (input: information sent from the server, output: information displayed on the device).
[1655] Step 8:
[1656] User review and use of information
[1657] The user checks the displayed information and accesses the information they need, such as reading the latest news about vaccines, checking side effect data, and scientific research results (input: information displayed on the device, output: information obtained by the user).
[1658] Through these steps, the system is able to efficiently collect and organize reliable information and provide it to users.
[1659] (Application example 1)
[1660] 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."
[1661] Conventional information retrieval systems have issues such as low reliability of information collected based on search keywords and lack of an appropriate display format for users to easily view and use information. Furthermore, especially when providing information for smart glasses and other wearable devices, the lack of a voice-based interface limits the user experience. To address these issues, a system is needed that can provide real-time, reliable information and display that information in an optimized audio and visual format.
[1662] 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.
[1663] In this invention, the server includes means for receiving search keywords from users, crawling engine means for collecting information related to the search keywords from online sources, means for filtering the collected information and eliminating false information, means for classifying the filtered information in chronological order and by type, generation AI means for displaying the generated information in a format suitable for a display device, means for analyzing voice input and generating search keywords, a speech synthesizer for outputting the classified information by voice, and means for providing the information to the user's display device. This allows users to obtain reliable information in real time and easily view and use it in a format optimized for audio and visuals.
[1664] A "user" is an entity that searches for and uses information.
[1665] "Search keywords" are words or phrases related to the information a user wants to find.
[1666] "Online sources" are information sources such as websites and databases that can be accessed via the Internet.
[1667] A "crawling engine" is software that automatically collects specific information from online sources.
[1668] "Filtering" is the process of removing unreliable information or false information from collected information.
[1669] "Chronological order" refers to arranging information in order based on the date and time it was collected.
[1670] "By type" refers to classifying information into categories according to its content.
[1671] "Generative AI" is artificial intelligence that formats collected and filtered information into a format suitable for the user's device or display device.
[1672] "Voice input" is a method in which a user provides information to a system by speaking into a microphone.
[1673] A "speech synthesizer" is a device that converts text information into speech and makes it audible to the user.
[1674] A "display device" is a device for visually presenting collected and generated information to a user.
[1675] This invention is a system that collects necessary information based on search keywords entered by the user through input or voice, eliminates unreliable information, and uses a generation AI to provide the information in a format suitable for the user's device. Specific embodiments of this system are described below.
[1676] Server Processing
[1677] 1. Receiving search keywords from users
[1678] The user enters a search keyword by voice or text, and the request is sent to the server. An example search keyword is "new spring clothes." The server receives this request and starts the search process.
[1679] 2. Starting the crawling engine
[1680] The server uses a crawling engine to collect information related to the search keywords from online sources, such as fashion websites, official brand websites, and social media.
[1681] 3. Information filtering
[1682] The information collected by the crawling engine is processed through a filtering system within the server. The filtering system evaluates the reliability of the information and removes unnecessary data such as false information. For example, anonymous social media posts and data from less reliable sources are excluded.
[1683] 4. Classification and chronological organization of information
[1684] The server then organizes the filtered information chronologically and by type, specifically into categories like tops, bottoms, and outerwear.
[1685] 5. Information Shaping with Generative AI
[1686] The organized information is then formatted into a suitable format for the display device by the generation AI, which understands the meaning of the search keywords and determines the layout so that the user can easily view related information.
[1687] 6. Transmission of Information
[1688] The formatted information is sent from the server to the user's display device using web technologies such as HTML and JSON.
[1689] Terminal handling
[1690] 1. Enter search keywords
[1691] The user inputs keywords related to the information they want to search for into a search box on the device or provides them by voice. For example, the user inputs or provides by voice a specific keyword such as "new spring clothes."
[1692] 2. Receiving and Displaying Information
[1693] The terminal receives the formatted information sent from the server, displays the information in a format optimal for the display device, and uses a speech synthesizer to provide the information to the user aloud.
[1694] User Activity
[1695] 1. Enter search keywords
[1696] Users input the information they want to know into the terminal. Voice input is also supported, so for example, keywords such as "new spring clothes" can be provided by voice.
[1697] 2. Verification and Use of Information
[1698] Users can access the information they need by viewing the information displayed in a portal format on the display device. Audio output is also provided, allowing users to view and hear the information.
[1699] Specific examples
[1700] For example, if a user uses smart glasses to say, "I want new spring clothes," the system will analyze the speech, collect relevant fashion information, filter and organize it, and use the generative AI to visually display the information in a format that is optimal for the smart glasses, while the speech synthesizer will provide the information aloud.
[1701] Prompt Sentence Examples
[1702] What's the latest information on spring fashion?
[1703] The hardware used includes smart glasses (e.g., microphone, speaker, display), and the software includes Python, Requests, BeautifulSoup, OpenAI GPT-3, and google_speech, allowing users to efficiently receive up-to-date, reliable information through visual and audio.
[1704] The flow of the specific processing in the application example 1 will be described with reference to FIG.
[1705] Step 1:
[1706] Enter and submit search keywords
[1707] The user enters search keywords using the voice input function of the smart glasses.
[1708] For example: "I want some new spring clothes."
[1709] Speech recognition software converts this voice input into text data.
[1710] The converted text data is sent to the server.
[1711] Step 2:
[1712] Receiving search keywords
[1713] The server receives the text data sent by the user.
[1714] Input: A keyword spoken by the user (e.g., "new spring clothes")
[1715] Output: Search keyword data for analysis
[1716] Step 3:
[1717] Starting the crawling engine
[1718] The server launches a crawling engine targeting multiple online information sources (fashion sites, official brand websites, social media, etc.).
[1719] Input: Search keyword ("new spring clothes")
[1720] Data processing: Obtain relevant information from sources and collect it as text data.
[1721] Output: Collected text data
[1722] Step 4:
[1723] Information Filtering
[1724] The server evaluates the reliability of the collected information using a filtering system.
[1725] Input: Collected text data
[1726] Data calculation: Remove unreliable information and false information.
[1727] What it does: Identify and filter out low-trust sources (e.g., anonymous social media posts).
[1728] Output: Filtered reliable information data
[1729] Step 5:
[1730] Classification and chronological organization of information
[1731] The server categorizes the filtered information chronologically and by type (tops, bottoms, outerwear, etc.).
[1732] Input: Filtered reliable information data
[1733] Data processing: Sort the information based on the date and time it was collected and divide it into categories.
[1734] Output: Categorized and organized information data
[1735] Step 6:
[1736] Shaping information with generative AI
[1737] The server uses generative AI to format the organized information into a format suitable for display on smart glasses.
[1738] Input: Classified and organized information data
[1739] Data calculation: Generative AI determines the optimal layout for the user and converts it into a format that is easy to display visually.
[1740] What it does: Sorts information by category, with the most recent information at the top.
[1741] Output: Formatted information data
[1742] Step 7:
[1743] Sending information
[1744] The server sends the formatted information to the user's smart glasses.
[1745] Input: Formatted information data
[1746] Data processing: Converts data into a transmission format (e.g., HTML, JSON) and executes transmission logic.
[1747] Output: Information data sent to the user terminal
[1748] Step 8:
[1749] Display and speech output of information
[1750] The terminal provides the received information to the user visually and audibly.
[1751] Input: Information data sent from the server
[1752] Specific operation: Information is displayed on the HUD of the smart glasses, and at the same time, the voice synthesizer plays the information aloud.
[1753] Output: Visual and audio information provided to the user
[1754] 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.
[1755] The present invention is a system that collects information on the Internet based on search keywords entered by users, eliminates unreliable information, and recognizes the user's emotions using an emotion engine to provide the user with appropriate information in the form of a portal site. A specific embodiment of this system is described below.
[1756] Server Processing
[1757] 1. Receiving search keywords from users
[1758] When a user enters search keywords on their device and presses the search button, the request is sent to the server. The server receives this request and starts the search process. For example, if a user enters "COVID-19 vaccine information," the server analyzes the keywords and prepares to collect related information.
[1759] 2. Starting the crawling engine
[1760] The server starts a crawling engine to collect information related to the search keyword from multiple sources on the Internet (news sites, official websites, social media, blogs, etc.). The engine searches sources as needed to obtain the latest data.
[1761] 3. Information filtering
[1762] The collected information is processed through a filtering system within the server. The filtering system checks the reliability of the information, the reputation of the source, and the consistency of the content, and filters out rumors and misinformation. For example, anonymous posts on social media and information from unreliable sources are filtered out.
[1763] 4. Classification and chronological organization of information
[1764] The filtered information is organized chronologically and by category. Specifically, it is organized in chronological order based on the date and time of collection, and then divided into categories such as "latest news," "side effects," and "research data" according to its content.
[1765] 5. Information Shaping with Generative AI
[1766] The organized information is then laid out in a portal site format by a generative AI. This generative AI understands the meaning of search keywords and arranges relevant information so that users can easily browse it. For example, the latest news is placed at the top of the page, and information on side effects and research data are displayed as the main content.
[1767] 6. User Emotion Recognition by Emotion Engine
[1768] The server uses an emotion engine to analyze the user's search keywords and behavioral history to recognize the user's emotions. Based on this emotion recognition, the priority of the information displayed is changed. For example, if the user is worried, information that gives a sense of security is displayed at the top.
[1769] 7. Transmission of Information
[1770] The formatted information is then converted into a format appropriate for the user's device (such as HTML or JSON), and the converted data is ready to be sent to the user's device.
[1771] Terminal handling
[1772] 1. Enter search keywords
[1773] Users enter keywords related to the information they want to find into the device's search box. For example, they enter "COVID-19 vaccine information."
[1774] 2. Receiving and Displaying Information
[1775] The device receives the formatted information sent from the server. This information is displayed in a portal site format, allowing users to easily browse categorized information. For example, the top of the screen displays the latest news, the middle section displays side effect statistics, and the bottom section displays links to research papers. In addition, information appropriate to the user's current emotional state is displayed based on the analysis results of the emotion engine.
[1776] User Activity
[1777] 1. Enter keywords
[1778] The user inputs the information they want to know into the device. For example, they input specific keywords such as "COVID-19 vaccine information."
[1779] 2. Verification and Use of Information
[1780] Users can review the displayed portal-style information and access the information they need. For example, they can read the latest news about vaccines, check side effect data, and check scientific research results. Based on the analysis results of the emotion engine, information appropriate to the user's emotional state is recommended, providing information to reduce stress.
[1781] Specific examples
[1782] (Search keyword: "COVID-19 vaccine information")
[1783] 1. Receiving a request
[1784] The server receives a search request for "COVID-19 vaccine information."
[1785] 2. Starting the crawling engine
[1786] The crawling engine starts up and collects information about vaccines from news sites, official websites, social media, etc.
[1787] 3. Information filtering
[1788] The filtering system evaluates the reliability of collected information and filters out false information, for example, giving priority to information from reliable news sites and official organizations.
[1789] 4. Classifying and Formatting Information
[1790] The filtered information is categorized into the latest news, side effect data, research results, etc. and organized in chronological order. Generative AI then formats this information into a portal site format and prepares it for delivery to users.
[1791] 5. Emotion Recognition by Emotion Engine
[1792] If a user is worried, information that provides reassurance (e.g., vaccine effectiveness or reassuring statistics) will be prioritized based on that emotion.
[1793] 6. Transmission of Information
[1794] The formatted information is sent to the user's terminal, where the user can view the information displayed in a portal site format and access the information they need.
[1795] As described above, this invention is designed to enable users to efficiently obtain the information they need. In addition to eliminating false information and classifying information by time series and category, reliability, ease of use, and psychological compatibility are improved by recognizing user emotions using an emotion engine. This system allows users to quickly access appropriate information.
[1796] The processing flow will be explained below.
[1797] Server Processing
[1798] Step 1:
[1799] The server receives the search keywords sent by the user, along with the user's settings information (language settings, region settings, etc.).
[1800] Step 2:
[1801] The server starts a crawling engine to collect information, and the crawling engine prepares to collect information related to the specified keyword from a wide range of sources.
[1802] Step 3:
[1803] The crawling engine accesses predefined sources such as news sites, official sites, social media, blogs, etc. to gather information related to the search keyword, exploring multiple sources in parallel during this process.
[1804] Step 4:
[1805] The collected information passes through a filtering system within the server, which evaluates the reliability of the information and filters out hoaxes and unreliable information. For example, unreliable anonymous social media posts are filtered out.
[1806] Step 5:
[1807] The filtered information is sorted chronologically and by category. Specifically, it is organized in chronological order based on the date and time the information was collected, and then divided into categories such as "latest news," "side effects," and "research data" according to its content.
[1808] Step 6:
[1809] The AI then organizes the information and lays it out in a portal site format. The AI understands the meaning of the search keywords and places relevant information in key positions. For example, the latest news is placed at the top of the page, and detailed research results are displayed in the center.
[1810] Step 7:
[1811] The server starts the emotion engine, which analyzes the user's search keywords and history to recognize the user's emotions. Based on the recognition results, the display priority is changed.
[1812] Step 8:
[1813] The formatted information is then converted into a format appropriate for the user's device (such as HTML or JSON), and the converted data is ready to be sent to the user's device.
[1814] Terminal handling
[1815] Step 1:
[1816] The user enters keywords into the search box on the device and presses the search button. This input information is sent to the server.
[1817] Step 2:
[1818] The terminal receives the formatted information sent from the server, parses the received data, and prepares it for display.
[1819] Step 3:
[1820] The device displays the received information in a portal site format. For example, the top of the screen displays the latest news, side effect statistics in the middle, and links to research papers in the bottom. Based on the analysis results of the emotion engine, the device also displays information appropriate to the user's current psychological state.
[1821] User Action
[1822] Step 1:
[1823] Users enter the keywords they want to know in the search box on their device and press the search button. For example, they could enter "COVID-19 vaccine information."
[1824] Step 2:
[1825] Users can check the portal site-style information displayed on their device and access the information they need. For example, they can click on the latest news headline to read a detailed article. They can also view recommended reassuring information based on the analysis results of the emotion engine.
[1826] Specific examples
[1827] (Search keyword: "COVID-19 vaccine information")
[1828] Server details:
[1829] Step 1:
[1830] The server receives a search request for "COVID-19 vaccine information."
[1831] Step 2:
[1832] The crawling engine starts up and collects information about vaccines from news sites, official websites, social media, etc.
[1833] Step 3:
[1834] The collected information is filtered out using a filtering system to remove false information, such as anonymous posts on social media and unreliable information.
[1835] Step 4:
[1836] The filtered information is categorized into the latest news, side effect data, research results, etc. and organized in chronological order.
[1837] Step 5:
[1838] The generating AI lays out the classified information in a portal site format, placing important information at the top.
[1839] Step 6:
[1840] The server activates an emotion engine, analyzes the user's search history and behavior, and prioritizes displaying information that will reassure worried users.
[1841] Step 7:
[1842] The formatted information is converted into HTML or JSON format and sent to the user's device.
[1843] Detailed terminal processing:
[1844] Step 1:
[1845] The user enters "COVID-19 vaccine information" and presses the search button.
[1846] Step 2:
[1847] The terminal receives the information sent from the server and analyzes it.
[1848] Step 3:
[1849] The terminal displays the received information in a portal site format, and displays information according to the user's emotions with priority.
[1850] User detailed processing:
[1851] Step 1:
[1852] A user enters "COVID-19 vaccine information" into the search box and presses the search button.
[1853] Step 2:
[1854] Users can check the displayed information in a portal site format and click on the item they are interested in to see more details. For example, a worried user will be given priority in seeing reassuring information demonstrating the effectiveness of vaccines.
[1855] As described above, the present invention not only enables users to efficiently obtain the information they need, but also improves user satisfaction and reliability, since the information provided by the emotion engine is adapted to the user's emotional state.
[1856] Example 2
[1857] 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."
[1858] In today's Internet environment, a huge amount of information is generated every day, making it difficult for users to quickly obtain reliable information from that information. Furthermore, if the information is unreliable or contains false information, users may act based on the incorrect information. Furthermore, conventional information gathering systems lack a mechanism for changing the priority of information taking into account the user's emotions, which makes it difficult for users to quickly access the information they want to see.
[1859] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving a search keyword from a user, information collection means for collecting information related to the search keyword from online information sources, means for filtering the collected information and eliminating unreliable information, means for classifying the filtered information in chronological order and by category, generation AI means for displaying the generated information in a portal site format, emotion recognition means for recognizing the user's emotion and changing the priority of information based on the emotion, and means for providing information to the user's terminal. This allows the user to quickly access reliable information and provides appropriate information according to the user's emotional state, thereby improving the convenience and reliability of information.
[1860] The "means for receiving search keywords from a user" is a function that allows the server to receive search keywords that a user has input on a terminal.
[1861] "Information gathering means for gathering information related to search keywords from online sources" is a function for gathering information related to specific search keywords from multiple sources on the Internet.
[1862] "Means for filtering collected information and eliminating unreliable information" refers to a function for removing unreliable information or false information from the collected information.
[1863] "Means for classifying filtered information chronologically and by category" refers to a function for organizing filtered information in chronological order based on the date and time of collection and classifying it into different categories depending on its content.
[1864] "Generative AI means for displaying generated information in a portal site format" is a function that utilizes artificial intelligence technology to format organized and classified information into a portal site format and display it in a way that is easy for users to view.
[1865] "Emotion recognition means that recognizes the user's emotions and changes the priority of information based on those emotions" is a function that analyzes the user's search keywords and behavioral history to recognize the user's emotions, and dynamically changes the priority of the information to be displayed based on the results.
[1866] The "means for providing information to a user's terminal" is a function for transmitting formatted information to a user's terminal in an appropriate format.
[1867] This invention is a system that collects information on the Internet based on search keywords entered by users, eliminates unreliable information, recognizes the user's emotions using an emotion engine, and provides appropriate information in the form of a portal site. This system is composed of three entities: a server, a terminal, and a user.
[1868] Server Action:
[1869] First, the server receives the search keywords entered by the user on the device. For example, if a specific keyword such as "COVID-19 vaccine information" is entered, the server analyzes this information using a dedicated analysis module.
[1870] The server then launches a crawling engine to gather relevant information from multiple sources on the Internet (news sites, official websites, social media, blogs, etc.) using software such as Apache Nutch or Scrapy. The engine generates a list of URLs as needed and selects relevant information.
[1871] The collected information is then passed through a filtering system to check its reliability. This system analyzes the reputation of the information source and the consistency of its content, and eliminates unreliable information and hoaxes. For example, anonymous posts on social media and unreliable blogs are excluded.
[1872] After filtering, the information is further sorted chronologically and by category. Specifically, it is organized based on the date and time of collection and divided into categories such as "latest news," "side effects," and "research data" according to its content.
[1873] The information is then formatted using generative AI and laid out in the form of a portal site. Models such as GPT-3 and BERT are used for this generative AI. The generative AI understands the context of the search keywords and arranges the information so that it is easy for users to view. The layout is designed with usability in mind, such as placing the latest news at the top of the page and displaying information about side effects and research data as the main content.
[1874] Furthermore, the server uses an emotion engine to analyze the user's search keywords and behavioral history to recognize the user's emotions. The emotion engine uses natural language processing technology to determine whether the user is anxious. For example, if the user is anxious, it can prioritize the display of information that will give a sense of reassurance (e.g., the effectiveness of vaccines or reassuring statistical data).
[1875] Finally, the formatted information is converted into a format suitable for the user's device (such as HTML or JSON) and sent to the device using the HTTP protocol.
[1876] Terminal handling:
[1877] Users enter keywords for the information they want to find into the search box on their device. For example, they might enter "COVID-19 vaccine information." This information is sent to the server, where it is analyzed and processed.
[1878] The device receives and displays the formatted data sent from the server. This information is displayed in a portal site format, allowing users to easily browse categorized information. For example, the top of the screen displays the latest news, the middle section displays side effect statistics, and the bottom section displays links to research papers. Additionally, information appropriate to the user's current psychological state is displayed based on the analysis results of an emotion engine.
[1879] User Activity:
[1880] Users enter the information they want to know into their device and perform a search. They then check the portal-site-style information displayed on the device and access the information they need. For example, they can browse the latest news about vaccines and check side effect data and scientific research results. Based on the analysis results of the emotion engine, information that is appropriate to the user's emotional state is highlighted, allowing the user to use the information with peace of mind.
[1881] Examples of prompts:
[1882] "New Coronavirus Vaccine Information"
[1883] The flow of the identification process in the second embodiment will be described with reference to FIG.
[1884] Step 1: User enters keywords
[1885] Users enter keywords for the information they want to find into the device's search box, such as "COVID-19 vaccine information."
[1886] Input: The search term entered by the user.
[1887] Output: The data sent to the server as a search request.
[1888] Step 2: Receiving and analyzing search keywords
[1889] The server receives search keywords sent by the user, which are then input to an analysis module and analyzed as search keywords.
[1890] Input: The search keyword submitted by the user.
[1891] Output: Parsed search keywords data format.
[1892] Specifically, the server extracts keywords as string patterns using an analysis module and passes them to the next processing step.
[1893] Step 3: Start the crawling engine and gather information
[1894] The server runs a crawling engine to collect relevant information from online sources (news sites, official websites, social media, blogs, etc.) using crawling engines such as Apache Nutch and Scrapy.
[1895] Input: The parsed search keywords.
[1896] Output: Collected raw information data.
[1897] Specifically, the crawling engine generates a list of URLs to search and sends HTTP requests to collect relevant information.
[1898] Step 4: Filtering the information
[1899] The collected information is then input into a filtering system to check its reliability, which evaluates the source of the information and eliminates unreliable or false information.
[1900] Input: Collected information data.
[1901] Output: filtered reliable information data.
[1902] Specifically, the filtering system evaluates information sources using reputation scores and eliminates information from unreliable social media posts and unreliable blogs.
[1903] Step 5: Categorize and organize information
[1904] The filtered information is sorted chronologically and by category. Specifically, it is organized based on the date and time of collection and sorted into categories such as "latest news," "side effects," and "research data" according to its content.
[1905] Input: Filtered information data.
[1906] Output: Information data organized in categorized and chronological order.
[1907] Specifically, the server uses the information's metadata (collection date and time, information category, etc.) to store the information in a database and then categorizes and organizes it appropriately.
[1908] Step 6: Formatting information with generative AI
[1909] The organized information is then laid out in a portal site format by a generative AI, using models such as GPT-3 and BERT. The generative AI understands the context of search keywords and arranges the information in a way that makes it easy for users to browse.
[1910] Input: Information data organized in categorized and chronological order.
[1911] Output: Information data formatted for a portal site.
[1912] Specifically, the generative AI model analyzes the input data, places the latest news at the top of the page, and arranges information about side effects and research data as the main content.
[1913] Step 7: Recognizing user emotions with the emotion engine
[1914] The server uses an emotion engine to analyze the user's search keywords and behavioral history to recognize the user's emotions. Based on the emotion recognition, the priority of the information to be displayed is changed.
[1915] Input: User search keywords and behavioral history data.
[1916] Output: Emotion recognition results and reprioritized information data.
[1917] Specifically, the emotion engine uses natural language processing technology to analyze the user's emotional state (e.g., worry, relief), and optimizes the display order of information based on the results.
[1918] Step 8: Send and view information
[1919] The formatted information is converted into a format suitable for the user's device (HTML, JSON, etc.) and sent to the device. The device displays the received information in portal site format.
[1920] Input: Information data with changed priority.
[1921] Output: Data sent to the user's device.
[1922] Specifically, the server converts the information into an appropriate format and sends it to the user's device using the HTTP protocol. The device then renders the received data using a display engine, allowing the user to view the information in a portal site format.
[1923] (Application example 2)
[1924] 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."
[1925] In today's information society, a huge amount of information is circulating on the Internet, including unreliable and false information. It is difficult for users to quickly and accurately obtain the information they need. Furthermore, there are few systems that prioritize and provide appropriate information based on the user's search behavior and psychological state. This can cause users to feel stressed or anxious during the information acquisition process. The present invention aims to solve these problems and efficiently provide reliable information that corresponds to the user's emotional state.
[1926] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: means for receiving a search keyword from a user; crawling engine means for collecting information related to the search keyword from online information sources; means for filtering the collected information and eliminating false information; means for classifying the filtered information in chronological order and by category; generation AI means for displaying the generated information in a portal site format; means for providing the information to the user's terminal; emotion engine means for recognizing the user's emotions and changing the priority of the information to be displayed based on the emotions; means for classifying the filtered information by reliability, content, and category and rearranging it in chronological order; and means for prioritizing the generated portal site-formatted information based on the user's emotions. This allows users to efficiently obtain reliable information without stress and make decisions based on appropriate information.
[1927] "User" refers to an end user who uses this system to search for information.
[1928] A "search keyword" is a word or phrase that a user enters to identify information in which they are interested.
[1929] A "crawling engine" is software that automatically collects information related to specified search keywords from various sources on the Internet.
[1930] "Filtering" is the process of evaluating the reliability and relevance of collected information and eliminating or filtering out false or unnecessary information.
[1931] "Chronological order" is a method of ordering information based on the date and time it was collected, with the most recent information being displayed first.
[1932] "Categorization" is the process of separating collected information into different categories based on its content.
[1933] "Generative AI" is artificial intelligence that takes collected and organized information and formats it into a form that users can easily understand and use (e.g., portal site format).
[1934] An "emotion engine" is a system that analyzes a user's input and behavioral history to recognize the user's current emotions.
[1935] "Changing the priority" means appropriately adjusting the display order of information based on the emotional state of the user.
[1936] The "portal site format" is a format that integrates various information and allows users to access all information on a single interface.
[1937] This invention is a system that collects information on the Internet based on search keywords entered by a user, evaluates the reliability of that information, and provides appropriate information according to the user's emotional state. Specific embodiments of this invention are described below.
[1938] System Overview
[1939] The system includes a server that receives search keywords from users and collects related information from the internet, and a terminal that allows users to access the information. The information is classified through reliability evaluation and filtering, and then formatted into a portal site format by a generative AI. In addition, the emotional engine recognizes the user's emotional state and adjusts the priority of the information to be displayed.
[1940] Hardware used
[1941] 1. Server: Oversees information collection, filtering, classification, formatting by generative AI, and processing by the emotion engine.
[1942] 2. Terminal: A device such as a smartphone or tablet that allows users to input, receive, and view information.
[1943] Software used
[1944] 1. Crawling engine: Collects data related to the search keywords from sources.
[1945] 2. Filtering system: Evaluate the reliability of collected data and eliminate false information.
[1946] 3. Classification system: Organize the filtered data chronologically and by category.
[1947] 4. Generative AI: Layout information into a portal site format for easy viewing by users.
[1948] 5. Emotion engine: Analyzes the user's search behavior and history, recognizes the user's emotional state, and adjusts the information that is displayed as a priority.
[1949] Data processing and calculation
[1950] 1. Receiving search keywords:
[1951] The user enters search keywords into the terminal and sends them to the server.
[1952] 2. Collection of Information:
[1953] The crawling engine on the server starts up and collects relevant information from news sites, official websites, social media, blogs, and other sources on the Internet.
[1954] 3. Information filtering:
[1955] The filtering system evaluates the reliability of collected information and filters out false or inaccurate information.
[1956] 4. Classifying and organizing information:
[1957] The filtered information is sorted chronologically and into categories such as "latest news" and "word of mouth."
[1958] 5. Emotion recognition:
[1959] The emotion engine analyzes users' search history and behavior to recognize their emotional state and prioritize information based on the recognized emotion.
[1960] 6. Formatting Information:
[1961] The organized information is laid out in a portal site format by the generation AI and converted into a format suitable for the user's device.
[1962] 7. Transmission and Display of Information:
[1963] The converted data is sent to the user's device and displayed in a portal site format, allowing the user to easily browse the classified information and access the information they need.
[1964] Specific examples
[1965] When a user enters the keyword "COVID-19 vaccine information," the server receives this keyword and the crawling engine collects related information. The filtering system eliminates unreliable information and classifies it into categories such as "latest news," "side effect information," and "research data," and the generative AI layouts the information appropriately. The emotion engine analyzes the user's emotions and prioritizes displaying information that provides reassurance if the user is feeling anxious.
[1966] Example prompt for generative AI model:
[1967] Provide reliable, up-to-date information on the following keyword: "COVID-19 vaccine information." Categorize information into news, reviews, and statistics, and prioritize information that reassures users based on emotion recognition.
[1968] Based on this example, it becomes easier for users to efficiently obtain reliable information and access data categories that have already been specified.
[1969] The flow of the specific processing in the application example 2 will be described with reference to FIG.
[1970] Step 1:
[1971] Input: The user enters a search keyword into the device's search box. Example: "COVID-19 vaccine information."
[1972] How it works: The user presses the search button and a search request is sent to the server.
[1973] Output: The server receives the search keywords and starts the search process.
[1974] Step 2:
[1975] Input: The server is ready to process the search keywords received from the user.
[1976] How it works: The server launches a crawling engine to collect information related to the search keywords from numerous online sources (news sites, official websites, social media, blogs, etc.).
[1977] Output: The server generates an initial list of the relevant information it has collected.
[1978] Step 3:
[1979] Input: The server proceeds based on the initial list of relevant information it has collected.
[1980] How it works: The filtering system is activated and evaluates the reliability of the collected information. Unreliable information and hoaxes are filtered out. For example, anonymous posts on social media and data from low-rated sources are filtered out.
[1981] Output: A filtered list of trusted information is generated.
[1982] Step 4:
[1983] Input: The server receives the filtered list of information.
[1984] How it works: Organizes information chronologically and by category. It categorizes information into breaking news, side effects, research data, etc., and sorts them chronologically within each category.
[1985] Output: A complete list of information organized chronologically and by category.
[1986] Step 5:
[1987] Input: The server performs the following process based on the organized information list.
[1988] How it works: The generative AI starts up and lays out the organized information in a portal site format. To make it easier for users to understand visually, the latest news is placed at the top of the page, side effect information in the middle, and research data at the bottom.
[1989] Output: Information in the form of a portal site is generated, laid out by the generation AI.
[1990] Step 6:
[1991] Input: The server receives the information laid out by the generation AI.
[1992] How it works: The emotion engine analyzes the user's search keywords and behavioral history to recognize their emotions. It then changes the priority of the information displayed based on their emotional state. For example, if the user is worried, it displays reassuring information at the top.
[1993] Output: A prioritized list of information tailored by the emotion engine is created.
[1994] Step 7:
[1995] Input: The server receives the priority list adjusted by the emotion engine.
[1996] What it does: Converts information into the appropriate data format (e.g. HTML, JSON, etc.) to send to the user's device.
[1997] Output: Data is formatted for sending to the terminal.
[1998] Step 8:
[1999] Input: The user's terminal receives the data sent from the server.
[2000] Operation: The device displays the received data in a portal site format. The user browses through the categorized information and accesses the information they need. In addition, the emotion engine prioritizes display, providing information appropriate to the user's psychological state.
[2001] Output: The portal site that the user can view is displayed.
[2002] 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.
[2003] 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.
[2004] 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.
[2005] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific m...
Claims
1. means for receiving search keywords from a user; a crawling engine means for collecting information related to the search keyword from online sources; A means of filtering the collected information and eliminating misinformation; a means for sorting the filtered information chronologically and by category; A generating AI means for displaying the generated information in a portal site format; means for providing information to a user terminal; A system including:
2. The system of claim 1 further comprising means for receiving the filtered information and displaying it in a portal site format.
3. The system according to claim 1, further comprising means for inputting and transmitting a search keyword.
Citation Information
Patent Citations
Persona chatbot control method and system
JP2022180282A